Results of Ground Creel Surveys Completed in 2001 T. Zimmerman BC Ministry of Water, Land and Air Protection #325, 1011 4th Ave., Prince George, British Columbia V2L 3H9 L. Kardash BC Conservation Foundation Suite 202, 1940 Third Ave, Prince George, British Columbia V2M 1G7

The Ministry of Water, Land, and Air Protection presently stocks 92 small lakes in the Omineca Region in order to provide quality recreational fishing opportunities to the angling public. In recent years, efforts to understand the status of these fisheries resources have primarily relied upon stocking assessment techniques. While stocking assessments allow the fisheries manager to quickly evaluate the size distribution, growth, health, and relative abundance of fish stocks in a particular system, they do not reveal the degree of fishing pressure that has been applied to a waterbody, nor the effects of that pressure on fish stocks. Since recreational harvesting alters the growth and abundance of fish populations, angler exploitation needs to be understood to correctly interpret stocking assessment results. The combination of stocking assessment and exploitation data can then be used to evaluate if stocks are being harvested in a sustainable manner, or if angling quality targets are being met.

Creel surveys provide a direct measure of the activity of anglers, their preferences, and their effect on fisheries resources. Catch rates (fish captured per unit time), and yield (number and mean size or weight of fish captured), are used as indicators of angler satisfaction and fishing quality, and estimates of total harvest derived from creel data can be used to assess the sustainability of an exploited fish stock.

In the Omineca Region, the lack of up to date creel data has precluded the development of specific angler success or conservation targets for recreational fisheries. Waterbody specific stocking rates and fishing regulations have consequently been developed based on perceived expectations of angling quality and stock sustainability, which may not be consistent with the actual performance or resilience of that waterbody's fishery.

To partly address these data shortcomings, we conducted a series of ground-based creel surveys at selected stocked lakes within the Omineca Region during the 2001 angling season. We focused our efforts on stocked lakes because the pool of candidate lakes was familiar to us, and because most had recently received stocking assessments against which the creel data could be compared. We identified 4 lakes that were representative of various stocking scenarios currently employed within the Region :

• stocked with rainbow trout, with no other species present (Amanita Lake);

• co-stocked with rainbow trout and brook trout, with non-piscivorous non-game species present (Cobb Lake);

• stocked with rainbow trout to supplement wild rainbow trout stocks, with piscivorous non-game species including northern pikeminnow present (Eulatazella Lake) and;

• stocked with rainbow trout to supplement wild rainbow trout stocks, with non-piscivorous non-game species present (Grizzly Lake West)

The results of these surveys, an interpretation of the results, and management recommendations are presented in the following report.

 

Methods

We used a two-stage, stratified random survey design (Dixon 1986, Neilsen and Johnson 1983) to conduct creel surveys at each of the 4 lakes. This design is suitable for single-access lakes where the probability of encountering anglers following the completion of their trips is 100%, a condition which was met for all 4 lakes. Stratification was employed to improve the precision of estimates for the study period.

At Cobb, Amanita, and Eulatazella lakes, the survey period started on the first day of the Victoria Day long weekend (May 19), and ended on the last day of the Labour Day long weekend (September 3). While it was recognized that some angling likely occurred before and after this period, for the purposes of this study we assumed this activity to be negligible in relation to the sampling period. Due to limitations in staff availability, the survey period for Grizzly Lake (West) was restricted to the late summer period, commencing July 12 and ending on September 3.

The angling season was stratified by morning and afternoon, and weekend and weekday, for a total of 4 strata. The Monday following a long weekend was included in the weekend stratum. Sampling was weighted towards weekends, whereby at least one weekend day was randomly selected for sampling. However, due to staffing limitations, Amanita Lake was not sampled during the last two weekends of the survey period, and the first and last weekends of the survey period were not sampled at Eulatazella Lake. Weekday strata were randomly selected for sampling as staff time permitted.

Creel shifts were 6 hours in length at Cobb, Eulatazella, and Amanita lakes. The proximity of Grizzly Lake West to Eulatazella Lake permitted us to conduct surveys at both lakes simultaneously, however the shift length at Grizzly Lake West was reduced when anglers were encountered at Eulatazella Lake. The morning shift (stratum) commenced at 0800 hours, and the afternoon shift commenced at 1500 hours. The period between 1400 hours and 1500 hours was not sampled. Data recorded during angler interviews included time spent fishing per rod, number of fish captured, number of fish released, and number of fish retained. If fish were retained by the angler, their catch was sampled for fork length (to the nearest mm), weight (to the nearest g), sex, and maturity.

A subsample of angling parties were also asked to participate in a voluntary angler questionnaire. We asked anglers to provide information on the number of days spent fishing at the lake where they were interviewed, the number of days fished elsewhere in British Columbia, the type of angling gear they used, the location of their home, what their reasons were for fishing on the day of the interview, and whether they were satisfied with their fishing experience. In this report, we present selected responses from this survey that are germaine to the subject content of the paper.

Estimates of total effort, total number of rods fished, catch rate, retained rate, release rate, and total harvest were developed for Eulatazella, Cobb, and Amanita lakes. Estimates were restricted to rate values for Grizzly Lake (West), due to the reduced sampling schedule for that lake. Estimates were developed for the entire angling season for Eulatazella, Amanita, and Cobb lakes, and for the sampling period at Grizzly Lake (West). In addition, complete estimates were made for three arbitrarily designated fishing sessions at Eulatazella, Cobb and Amanita lakes. The sessions were defined as "Early" (May 19 - June 29), "Middle" (June 30 - August 3), and "Late" (August 4 - September 3). We used sessional estimates instead of monthly estimates to evaluate within-season changes to the parameters measured, because sample sizes were too low to permit stratification within months. To minimize variances for estimates within sessions, we pooled the morning and afternoon strata, and stratified the sessional results by weekend and weekday only.

Sessional estimates of the total biomass of fish removed were calculated as the product of the sessional estimate of total catch and the mean mass of fish captured for that session. We did not use pooled seasonal data to calculate an independent seasonal estimate of biomass removed, because fishing effort and the mean mass of fish captured were highly variable between sessions. We instead summed the sessional values and their variances to obtain a more precise estimate of the total biomass of fish removed during the fishing season.

To estimate total fishing effort and the total number of rods fished, we computed total values and their variances within each stratum, then summed these values across strata to obtain seasonal and sessional estimates, as described in Dixon 1986. We accounted for the time period not surveyed between 1400 and 1500 hrs by applying a correction factor of 13/12 to these estimates and variances. To calculate 95% confidence intervals (CI) for each estimate, we first calculated the degrees of freedom for a stratified sample using the formula :

     

where is the degrees of freedom, is the total variance across all strata, is the variance of the ith stratum, and is the total number of sample events in the ith stratum (Gasaway et al. 1986). We then derived the t statistic from and calculated the CI as .

Catch rate, retention rate, and release rate were computed using the ratio of means estimator (Pollock et al. 1997). This estimator has been shown to be more precise than alternative estimators when anglers are interviewed at the end of their fishing trip. The formula for computing the ratio of means is:

     

where is the ratio being estimated, is the number of fish caught, retained, or released by the jth angler, and is the time fished in hours by the jth angler.

The variance for the ratio of means estimator was calculated using the formula (Cochran 1977):

     

where is the sample mean of the angler effort, is the finite population correction factor, is the sample size, is the variance of effort, is the variance of catch, and is the covariance of catch and effort. Because our survey did not include a census of the total number of anglers, we estimated by dividing by our estimate of the number of rods (i.e. anglers) for the session or season.

We estimated the number of fish caught, retained, and released for each session and by season, as the product of the effort estimate for each time period, and the associated rate parameter. Variances for each estimate were calculated as the variance of a product (Heard 1987).

 

Results and Discussion

TABLE 1. Total number of possible sample events (N), relative to sampled events, per session stratum. WD=weekday, WE=weekend   Early   Middle   Late   WD WE   WD WE   WD WE N 26 58   22 48   22 40 Cobb 5 7   6 5   5 5 Eulatazella 5 6   5 5   5 3 Amanita 7 5   5 6   4 2

Sampling Schedule

For any one stratum, we attempted to sample at least 5 events in order to minimize variances. Due to unforeseen reductions in staff availability, we were unable to meet this target during the Late session at Amanita (both strata) and Eulatazella (weekend stratum) lakes (Table 1).

Sample sizes at Amanita and Cobb lakes were sufficient to calculate seasonal estimates using full stratification by shift and day type. Eulatazella Lake was undersampled during the weekday morning and weekend afternoon strata, so these strata were pooled and seasonal estimates were stratified by day type only. Complete sampling schedules for Cobb, Amanita, Eulatazella, and Grizzly lakes are illustrated in appendices 1 through 4 respectively.

TABLE 2. Total number of angling parties interviewed by lake surveyed, per session and by season.   Session   Early Middle Late Season Cobb 94 57 60 211 Eulatazella 7 1 2 10 Amanita 16 4 3 23 Grizzly W NA NA NA 17

We conducted a total of 261 angler interviews across all of the lakes surveyed (Table 2). The low sample of interviews obtained during the Middle and Late sessions at Amanita and Eulatazella lakes reflect very low utilization rates that we observed for these lakes (see below).

 

Angler Effort

We estimate that Cobb Lake received 2,700 angler visits during the 2001 fishing season (Table 3), the highest level of angler use of the three lakes for which estimates were derived.

TABLE 3. Estimated number of anglers fishing by session and season in 2001, with 95% confidence intervals in parentheses.   Session   Early Middle Late Season Cobb 1,170 820 790 2,700   (730-1,610) (520-1,120) (500-1,080) (2,240-3,160) Eulatazella 120 21 60 200   (0-270) (0-75) (0-300) (30-360) Amanita 110 23 24 170   (10-220) (0-58) (0-72) (50-280)

Eulatazella and Amanita lakes received very little angler use, at 200 and 170 angler visits respectively.

Anglers spent 7,300 hours fishing at Cobb Lake, but only 240 and 250 hours at Eulatazella and Amanita lakes (Table 4). Fishing time per trip (95% CI) averaged 1.2 hours at Eulatazella Lake (0 - 2.6), 1.5 hours at Amanita Lake (0 - 3.1), and 2.7 hours at Cobb Lake (2.1 - 3.3). Angling activity was the highest during the Early session for all three lakes. Variances for all estimates were lower at Cobb Lake than at Eulatazella and Amanita lakes, suggesting that the latter two lakes received sporadic, unsustained angler use.

The shorter fishing times observed at Eulatazella and Amanita lakes likely reflect the low CPUE values experienced at each lake (see Catch Success below) and may also be influenced by the perceived expectations of fishing quality at the two lakes. For example, if anglers that targeted Amanita Lake had reduced expectations of fishing success due to its reputation as a poor quality fishery, they may have spent less time fishing than they would at lakes whose success rates are known to be high. Conversely, anglers that visit Cobb Lake may have high expectations of success and might therefore allocate more time to fishing regardless of their individual catch success.

TABLE 4. Estimated fishing effort in angler-hours by session and season in 2001, with 95% confidence intervals in parentheses.   Session   Early Middle Late Season Cobb 3,600 1,700 2,200 7,300   (2,000-5,300) (900-2,500) (1,300-3,200) (6,000-8,600) Eulatazella 110 60 70 240   (0-280) (0-230) (0-310) (30-450) Amanita 190 13 40 250   (0-440) (0-33) (0-150) (10-500)

Distance from home was an important determinant affecting lake selection by anglers at all three study lakes (Table 5), however sample sizes for Amanita and Eulatazella lakes were very low. Cobb Lake is located near the Highway 16 corridor, and attracts anglers due to its relatively quick and straightforward access. Amanita Lake is located within an hour's drive of Prince George. Ministry of Forests user statistics indicate that the Amanita Lake Forest Recreation Site receives between 2,400 and 3,500 user days per year (data on file), suggesting that its distance from Prince George does not negatively influence the recreational use of the lake by the public. Access to Eulatazella Lake is more complex and requires knowledge of forest service roads, however this did not deter the anglers that we interviewed from accessing the lake.

The reputation of a lake's fishery also influenced its selection by anglers. Cobb Lake has a reputation for providing anglers with good yields of brook and rainbow trout. Users of the Amanita Lake Forest Recreation Site informed the creel clerk that fishing was known to be poor, even though few of those users actually fished. Opinions regarding expectations of catch success at both Amanita and Cobb lakes are partially supported by gill net CPUE data we derived from stocking assessments completed in 1999 (Zimmerman 1999a, 1999b), however the relationship between stocking assessment results and angling success can be variable.

TABLE 5. Number of responses by category to the question, "What is your main reason for fishing on this lake today?" from anglers residing in B.C.   Distance from Home Reputation Natural Setting Other Cobb 56 33 12 48 Eulatazella 2 0 1 3 Amanita 5 2 1 4

We have no data on the perceived quality of fishing at Eulatazella Lake, and the few anglers who fished there did not have preconceived opinions regarding its fishing quality. Aerial boat counts conducted in the summer of 2001 (Zimmerman in prep) showed that Grizzly Lake (West), which is located approximately 15 km from Eulatazella Lake, received over 4 times as much fishing effort as did Eulatazella Lake. This suggests that while proximity may influence overall angler activity, that expectations of fishing success rates likely play an important role in lake selection at more remote locations.

 

 

Catch Success Rates: Rainbow Trout

The Cobb Lake rainbow trout fishery sustained the highest sessional and seasonal catch rates of all the lakes surveyed. Anglers averaged 0.49 fish per angler hour during the fishing season, with catches up to 0.62 fish per hour during the Early session (Table 6). Catch success dropped to half this rate by the July long weekend (i.e. the Middle session) and remained at a reduced level for the remainder of the season. Interestingly, the reduction in catch rate from the Early to Middle sessions corresponded to the reduction in angling effort over the same period. Whether these two results are correlated is unknown, however it is generally accepted that fishing success rates in the Omineca diminish by mid summer, once thermal stratification has occurred, and that this reduction in success leads to reduced angling effort.

TABLE 6. Estimated rainbow trout fishing success rates by session and season in 2001, in catch per angler-hour, with 95% confidence intervals in parentheses. NC=not calculated, rationale given below.   Session   Early Middle Late Season Cobb 0.62 0.31 0.42 0.49   (0.50-0.74) (0.16-0.46) (0.29-0.55) (0.41-0.57) Eulatazella 0.6 NC1 NC1 0.3   (0-1.7)     (0-0.9) Amanita 0.1 0 0 0.07   (0-0.2) (NC2) (NC2) (0-0.17) Grizzly W NC3 NC3 NC3 0.29   (NC3) (NC3) (NC3) (0.18-0.41) 1Sample size too low to estimate catch rate. 2No fish captured by any angler during session. 3Sessional values not calculated for Grizzly Lake West.

We were unable to estimate catch rates for Eulatazella Lake during the Middle and Late sessions due to the paucity of anglers that fished during these periods (Table 2, Table 6). The 7 angling parties we interviewed during the Early session averaged 0.6 fish per angler hour, but this result was inluenced by one angling party that succeeded in obtaining 8 rainbow trout over 4 hours of effort, while the remainder of the parties did not catch any trout. This variability in fishing success is reflected in the wide confidence interval for the estimate (Table 6). These results suggest that Eulatazella Lake is capable of providing a successful fishing experience to anglers, but that highly variable catch rates may be negatively influencing the selection of this lake by anglers.

Amanita Lake exhibited extremely low catch success rates during all sessions and for the angling season. This lake receives annual releases of 1,500 hatchery rainbow trout (Zimmerman 1999a) and should sustain a moderate recreational fishery capable of providing higher catch rates than what we observed during our survey. Our results lead us to believe that Amanita Lake rainbow trout stocks are experiencing higher than normal natural mortality, perhaps resulting from sustained winter kills or possibly due to low lake productivity.

The Grizzly Lake West seasonal catch rate estimate of 0.29 fish per hour was consistent with catch rates reported in the early 1980s of between 0.31 and 0.47 fish per hour (Dixon 1983). Because we did not begin to sample Grizzly Lake West until July 12, this result may underestimate the true seasonal value of the mean catch rate for the fishery, since spring rates are typically higher than summer rates. We therefore predict that catch success during the spring period may approach or exceed 0.5 fish per hour.

 

Catch Success Rates: Other Species

In addition to rainbow trout, eastern brook trout were captured at Cobb Lake, and kokanee and mountain whitefish were taken at Eulatazella Lake. Incidental catches of northern pikeminnow were also reported at Eulatazella Lake. Catch rates for these species were lower than those reported for rainbow trout (Table 7).

TABLE 7. Estimated catch per angler hour of species other than rainbow trout, by session and season in 2001. 95% confidence intervals are given in parentheses. EBT=eastern brook trout, KO=kokanee, MW=mountain whitefish, NSC=northern pikeminnow. NC=not calculated, rationale given below.   Session Lake (Species) Early Middle Late Season Cobb (EBT) 0.084 0.036 0.038 0.059   (0.038-0.132) (0.013-0.060) (0.014-0.061) (0.034-0.084) Eulatazella (KO,MW) 0.17 NC1 NC1 0.08   (0-0.42)     (0-0.22) Eulatazella (NSC) 0 NC1 NC1 0.17   NC2     (0-0.37) 1Sample size too low to estimate catch rate. 2No fish captured by any angler during session.

We did not differentiate anglers that targeted brook trout from those that were fishing for rainbow trout at Cobb Lake. Thus our estimate of the brook trout catch rate may be lower than the actual catch rate, since the effort data were pooled for both species. Higher accuracy for the summer brook trout fishery could be achieved by refining the survey method to include the type of fish targeted in the fishery.

We assumed that anglers that fished at Eulatazella Lake targeted rainbow trout, but would consider mountain whitefish or kokanee for retention as a 'sport' fish. The catch rates reported for these species suggest that the wild non-rainbow salmonid populations in Eulatazella Lake currently supplement the rainbow trout fishery, but do not support a directed fishery. Seasonal catch rates of northern pikeminnow exceeded those of kokanee and whitefish, but were lower than rainbow trout catch rates.

 

TABLE 8. Estimated number of trout captured, harvested, and released by anglers at Cobb, Eulatazella, and Amanita lakes. 95% confidence intervals are given in parantheses. RB=rainbow trout, EB=eastern brook trout, NC=not calculated due to small sample size.   Session   Early Middle Late Season   Cobb RB Captured 2,300 520 940 3,600 (1,300-3,200) (200-840) (500-1,370) (2,800-4,400) Harvested 1,010 350 500 1,800 (580-1,440) (150-560) (250-750) (1,370-2,240) Released 1,250 170 430 1,790 (670-1,840) (30-310) (190-670) (1,280-2,310)   Eulatazella RB Captured 70 NC NC 80 (0-220)     (0-240) Harvested 50 NC NC 50 (0-110)     (0-120) Released 28 NC NC 30 (0-82)     (0-90)   Amanita RB Captured 17 0 0 18 (0-46)     (0-47) Harvested 6 0 0 6 (0-27)     (0-22) Released 11 0 0 12 (0-35)     (0-36)   Cobb EB Captured 300 62 80 430 (100-510) (17-107) (20-140) (240-630) Harvested 170 41 30 230 (80-260) (4-79) (2-62) (140-320) Released 140 21 54 200 (0-280) (0-44) (16-92) (70-340)

Capture, Retention, and Release of Sport Fish

Our estimates of the total number of fish captured, retained, and released in each of the study lakes illustrate that Cobb Lake rainbow trout are highly exploited, whereas Eulatazella Lake and Amanita Lake stocks receive very little angling pressure (Table 8).

Fifty-six percent of the total number of rainbow trout retained at Cobb Lake were taken during the Early session, which corresponds to the high angler effort seen during the same period. Retention rates (number retained divided by number captured) varied from a low of 44% during the Early session to 67% during the Middle session. The seasonal retention rate was 50%. While it is beyond the scope of this study to evaluate the effects of exploitaiton on fish stocks, these results strongly suggest that the Cobb Lake adult rainbow trout population is subject to significant angling pressure. We estimate that 1,800 fish were removed from the lake during the 2001 fishing season; this likely represents a high proportion of the total number of catchable trout, given that the lake is stocked with 10,000 yearlings per year.

Not surprisingly, our exploitation estimates for rainbow stocks in Amanita and Eulatazella lakes are very low. The variances around each estimate are high owing to high variances in effort and catch rates for the two lakes. The upper 95% confidence limit of the estimate of fish harvested is 120 and 22 rainbow trout for Eulatazella and Amanita lakes, respectively. These low harvest estimates are attributable to low angling pressure at both lakes, however catch rates for those anglers that fished Eulatazella Lake were much higher (Table 6) than for those that fished at Amanita Lake. These results support our view that different factors are inluencing angler use and harvest at the two lakes. We believe that Eulatazella Lake is capable of supporting a recreational rainbow trout fishery, however either its location, the presence on non-game fish, or other deterrents are responsible for low angler use seen at this lake. Conversely, Amanita Lake, while aesthetically pleasing, appears to not support viable populations of trout. The causal factors for this poor survival are unknown at present, but may be attributable to the lake's low productivity.

With the exception of Cobb Lake eastern brook trout, we did not calculate harvest statistics for other sport fish species because catches were very low. We estimate that during the 2001 fishing season anglers caught 430 brook trout at Cobb Lake, and retained 230 of these fish (Table 8) for a retention rate of 53%. Anglers retained approximately 1 brook trout for every 4 rainbow trout retained. Cobb Lake supports a summer and winter fishery, the latter being primarily targeted at brook trout. Exploitation of brook trout during the summer months appears to be moderate relative to current stocking rates; if more detailed catch statistics are required for fish supply planning needs, a full winter creel survey would be necessary to fully document brook trout harvest rates throughout the year.

 

Parameters of Fish Harvested and Estimates of Biomass Removed

Because very few fish were captured, we are unable to describe the characteristics of exploited fish populations in either Eulatazella or Amanita lakes. Grizzly Lake (West) was sampled on an opportunistic basis, so few fish were obtained from this fishery for sampling. Thus we report only on the Cobb Lake eastern brook trout and rainbow trout fisheries, for which large sample sizes were available.

TABLE 9. Mean length, mean weight, and estimated total biomass of trout retained from Cobb Lake, 2001. Values in parantheses are standard deviations for length and weight, and 95% confidence intervals for biomass estimates.   Session   Early Middle Late Season Rainbow Trout Mean Length (mm) 424 363 371 378   (52.9) (33.2) (39.7) (40.4) Mean Weight (g) 618 492 523 529   (236.8) (107.7) (155.8) (164.4) Biomass (kg) 625 172 263 1,060   (85-1,165) (47-298) (61-465) (470-1,650) Brook Trout Mean Length (mm) 384 352 321 348   (29.3) (37.7) (30.1) (40.0) Mean Weight (g) 560 544 397 489   (126.8) (137.8) (102.3) (137.0) Biomass (kg) 94 22 12 129   (29-160) (0-46) (0-26) (58-200)

Retained Cobb Lake rainbow trout averaged 378 mm in length and 529 g in weight for the angling season (Table 9). The mean length and weight of fish retained was highest during the early session, then stabilized during the latter two sessions of the fishing season. These results suggest that Cobb Lake rainbow trout stocks are fished down during the angling season. Those fish that escape retention or are released without recapture continue to grow during the marginal season when fishing pressure subsides, then become available to the retainable fishery during the following season. Thus we believe that Cobb Lake rainbow trout angling success is regulated by the degree of angling pressure that occurs during the previous year, rather than by density dependent factors that appear to influence catch success in lakes with low angling activity.

We estimate that a total of 1,060 kg of rainbow trout were removed from Cobb Lake during the 2001 fishing season. This equates to a harvest of 5.0 kg per ha, which is consistent with harvest rates seen in moderately productive lakes in other jursidictions, that are stocked at low to moderate densities.

The numbers of brook trout sampled by the creel clerk were 4, 5, and 6 for the Early, Middle, and Late season respectively. These low sample sizes were due to low numbers that were captured and retained by anglers. The attribute data for each sessional sample is presented in Table 9, however these data should be viewed with caution because of the low sample sizes obtained. The mean length and weight of brook trout captured declined throughout the fishing season, however this trend was less distinct than that observed for rainbow trout, and may be an artefact of the small sample size. We estimate that 129 kg of brook trout were removed from the lake during the 2001 season, representing a harvest of 0.61 kg/ha.

 

Management Recommendations

The creel surveys that we conducted at Eulatazella and Amanita lakes reveal that these two lakes receive very little angling effort. Because few anglers were encountered during our surveys, we were unable to obtain sufficient data to draw firm conclusions regarding the cause for the low level of activity that we observed. However, the data we were able to retrieve from those anglers that did visit the lakes provide some evidence that different factors are responsible for their low use.

Amanita Lake is used by local residents for recreational activities, but with a few exceptions those users did not include angling as part of their recreational experience. Anglers that did attempt to fish experienced very low catch rates, even though the lake is stocked with 1,500 yearlings (42/ha) per year and should theoretically support a moderate fishery. We believe that this poor fishing success is explained by the lake's extremely low productivity. Simply put, low levels of nutrients in Amanita Lake constrain primary production, which in turn prevents zooplankton and benthic faunal communities from increasing to the levels that are required to support a healthy trout population. This shortage of food will be most acute during the winter months when terrestrial insects are not available as a diet supplement. Thus it is likely that many Amanita Lake rainbow trout experience high mortality during the winter months and are therefore not available to be angled during the summer fishing season. This situation cannot be easily remedied with conventional fisheries management tools. Given that other fishing opportunities are available in the vicinity of Amanita Lake, and that the lake does not appear to support stocked trout, we recommend that the lake be removed from the Omineca stocking program.

Angling success at Eulatazella Lake was variable but generally higher than what anglers experienced at Amanita Lake. The high variance in catch success can be attributed to the low sample size that we obtained, and possibly to the level of experience of the anglers that visited the lake. Most of the anglers we interviewed were not experienced fishers, but were visiting the lake as family outing or were "passing through." A larger sample of interviews is therefore required in order to sufficiently predict a mean catch rate for the Eulatazella fishery.

Since catch success may not be limiting angler interest, the low amount of use seen at Eulatazella is likely explained by other factors. The lake's proximity to other waterbodies that yield good fishing results (e.g. Grizzly Lake West) may divert fishing effort to those systems. The potential to capture non-game fish, such as northern pikeminnow, may reduce its appeal to the angling community. The size of rainbow trout available to the angler may be too small to warrant fishing effort, although we did not collect sufficient data to support this explanation.

Regardless of the reason for the low level of angling activity, our results suggest that the exploitation of rainbow trout stocks by anglers is far below the level needed to justify a supplemental stocking program. Indeed, the utility of supplemental stocking in mixed species lakes is questionable, and the Eulatazella data may be indicative of the level of angling effort that occurs at other mixed species lakes throughout the region. We therefore recommend that stocking be ceased at Eulatazella, and that consideration be given to the value of supplemental stocking in lakes with similar, naturally occurring fish communities.

Cobb Lake rainbow and brook trout stocks appear to support a robust recreational fishery that blends good catch rates with medium to large sized fish. Because no other lakes in the Omineca region have been subjected to creel surveys in recent years, we are unable to compare angling quality parameters with different stocking scenarios that are currently being employed in the region. Nevertheless, our results show that during the 2001 summer fishing season, Cobb Lake received 13 angler days per ha, with catch rates of up to 0.62 fish per angler-hour, and retention rates of up to 67%. This fishery is based on an annual stocking rate of 47 rainbow trout yearlings and 95 brook trout fingerlings per ha per year.

We conclude that this stocking strategy is well suited to both the productive capacity of Cobb Lake and needs of the angling community, and recommend that survey results from other lakes be gauged against the Cobb Lake fishery, because it represents a good compromise between catch rate and fish size given its productive potential. While our data indicates that Cobb Lake currently exhibits good fishing quality characteristics, we expect these traits to vary as biophysical conditions change and as the demographic makeup and behavior of the angling community shifts over time. To ensure that management strategies for the lake are optimized, we recommend that this survey be repeated within a 5 to 7 year time frame.

The Grizzly Lake (West) fishery was only partially sampled, yet our results show that recreational angling yielded moderate results during the mid summer, and likely provides good returns during the spring period. The lake contains naturally occurring stocks of non-piscivorous cyprinids which do not appear to detrimentally affect rainbow trout fish production. More work is needed to evaluate the relative performance of rainbow trout strains in mixed species lakes that do not contain piscivores, however the preliminary findings from Grizzly Lake (West) suggest that for these lake types, careful stocking strategies may provide good recreational opportunities if natural fish populations are accounted for.

 

Acknowledgments

This survey was conducted under a partnership arrangement between the Ministry of Water, Land, and Air Protection, the Carrier Sekani Tribal Council (CSTC), and the British Columbia Conservation Foundation. Funding was obtained from Fisheries Renewal B.C. and administered by the Upper Fraser-Nechako Fisheries Council. Lawrence Ward assisted with the fieldwork component of the assessment. Don Cadden (BC WLAP) and Todd French (UFNFC) reviewed the initial drafts of this report. Inquiries pertaining to this report should be directed to the author at the email and address located at the bottom of this page.