Alcova Reservoir

 

Objective:  Maintain RBT CPUE in spring FS above 0.85 and WAE CPUE in fall ES below 0.40 fish per hour.

 

·         Lake Survey. 512

Alcova Reservoir was netted during the weeks of 5/18/2009 (FG) and 9/8/2009 (ES) (Tables 4-7).

 

Table 4.  Number, CPUE (stdev), mean length (n; stdev) with ranges, and mean weight (n; stdev) with ranges of fish captured in FG Alcova Reservoir May 18-21, 2009.

 

Species	Number	CPUE	Mean Length	Range	Mean Weight	Range
BNT	4	0.02(0.03)	18.9(4;1.9)	16.4-21.0	2.43(4;0.73)	1.45-3.00
CRP	2	0.01(0.02)
LNS	1	0.01(0.02)
RBT	146	0.68(0.40)	14.1(144;1.9)	10.0-17.6	1.20(144;0.39)	0.48-1.99
WAE	2	0.01(0.02)	14.9(2;1.2)	14.0-15.7	1.07(2;0.25)	0.89-1.25
WHS	71	0.33(0.23)

 

 

Table 4a.  Relative Stock Density

 

Species	N>=S	RSD-Q	RSD-P	RSD-M	RSD-T
BNT	4	100	100	25	0
RBT	144	144	18	0	--
WAE	2	2	50	0	--

 

 

Table 5. Mean relative weights (n; stdev) with ranges and relative weights by length category (n; stdev) for selected fish species in ES, Alcova Reservoir, May 2009.

 

Species	Mean Wr	Range	S-Q	Q-P	P-M	M-T
BNT	77(4;9.9)	70-92			80(3;10.4)	70(1;--)
RBT	96(144;11.2)	64-131	98(118;9.9)	83(26;7.8)
WAE	87(2;1.4)	86-88	86(1;--)	86(1;--)

 

 

       Table 6. Number, CPUE (stdev), mean length (n; stdev) with ranges, and mean weight (n; stdev) with       

       ranges of fish captured in ER Alcova Reservoir, September 2009.

 

Species	Number	CPUE	Mean Length	Range	Mean Weight	Range
BNT	20	0.09(0.08)	19.5(20;2.22)	16.1-24.7	3.24(20;0.95)	1.70-5.46
CRP	4	0.02(0.03)
LNS	57	0.24(0.25)
RBT	41	0.18(0.13)	14.2(40;1.57)	11.7-17.6	1.17(40;0.36)	0.68-2.13
WAE	98	0.42(0.17)	18.4(97;5.98)	9.3-32.6	3.07(97;3.40)	0.21-14.94
WHS	540	2.34(1.03)

 

 

 

Table 6a.  Relative Stock Density

 

Species	N>=S	RSD-Q	RSD-P	RSD-M	RSD-T
BNT	20	100	100	35	5
RBT	40	20	0	--	--
WAE	93	77	33	20	4

 

 

Table 7.  Mean relative weights (n;stdev) with ranges and mean relative weights (n;stdev) by length category for fish captured with ER Alcova Reservoir, September 2009.

 

Species	Mean Wr	range	S-Q	Q-P	P-M	M-T
BNT	95(20;15)	76-132			99(13;17.3)	89(6;3.9)
RBT	91(40;6)	79-102	93(32;5.7)	86(8;5.3)
WAE	87(97;9)	69-111	83(21;6.7)	85(41;7.1)	91(12;7.0)	95(15;10.5)

 

 

Trout

There was a significant decrease in gillnet CPUE for RBT in 2009 (two-sample T, t = 3.14, p = 0.005, df = 21).  The large decrease in CPUE is likely the result of dwindling numbers of fish stocked (Figure 4).  Currently, Alcova is not meeting the management objective of RBT CPUE > 0.85 in standard spring sampling  (one-sample T, t = -1.47, p = 0.92).

 

Given the large variation is size in stocked RBT and number stocked since 2003, multiple regression was again employed to analyze the relationship between both size and number of stocked fish on RBT CPUE.  Multiple regression was run using individual nets as replicates (N = 84) with weighted mean size of RBT (number per pound) and number stocked the previous fall as predictors.  Based on analysis of studentized residuals from the initial regression, five nets were flagged as significant outliers.  These individual nets were excluded from the analysis as it was assumed the extremely high (1 net) or low (4 nets) catch rates were due to extraneous local circumstances and not a reflection on the actual density of RBT.  There is a significant relationship between size and number of RBT stocked on CPUE (p = 0.001) and is described by the regression equation:

CPUE = 1.16 – 0.0619(Mean size) + 0.0000001(Number), suggesting within the range of variables tested, the importance of size at stocking far outweighs the importance of number stocked.  Model precision, conversely is poor with size and number explaining only 14.4% of the overall variation in net catch.  The low r-square is predictable however, given the inherent variability in CPUE among individual nets.  The model will continue to be adjusted as successive data is collected.

 

 

 

Figure 4.  Interpolated surface depicting the relationship between weighted mean size of stocked RBT (number/pound), the total number stocked and spring floating gillnet CPUE.

 

 

Walleye

Since 2001, the walleye population has exhibited an overall significant positive trend as measured by ES CPUE (p = 0.027, r-square = 58.4%) (Figure 5).  Currently the management objective of WAE CPUE < 0.40 is not being met (T = 0.41, p = 0.65) and has not been met since 2003.  The only silver lining from the perspective of trout management is that very high WAE recruitment in 2005 and 2006 has been buffered by average to below average year-class production in 2007 and 2008.  The resulting WAE population is stable in terms of numbers, but ever-increasing in terms of size structure (Figure 6).

 

 

Figure 5.  Mean WAE CPUE ± 1 standard deviation in ES.  Alcova Reservoir.

 

 

 

Figure 6.  RSD categories for WAE captured in fall ES sampling on Alcova Reservoir, 2005 – 2009.

 

 

Walleye recruitment has been highly variable in Alcova in recent years.  As reported in the 2007 progress report, the 2005 and 2006 cohorts were very strong in Alcova.  When studentized residuals from weighted catch curve regression on 2005 – 2009 pooled relative frequency by age are inspected for the 2007 and 2008 cohorts, it is apparent that both cohorts were below average (in terms of proportion of total net catch) at age 1+ (Figure 7).  What is unclear is why the 2007 year-class was very poor at age 1+ but above average at age 2+, but could be indicative of a decrease in total annual mortality.

 

 

Figure 7.  Age 1+ through age 4+ year-class strength of the 2001 – 2008 cohorts from standardized fall ES catch, 2005 – 2009.

 

 

While geographically, Alcova is much closer to the upper system reservoirs, growth rates of WAE are more similar to Glendo (Figure 8).  WAE growth to age 5 is slightly lower in Alcova than Glendo, but after age 6, WAE are consistently larger in Alcova than in Glendo.  This is likely attributable to the RBT stocking program, especially in light of the fact that RBT in the 6 – 12 per pound range have been frequently stocked since 2002.

 

 

 

 

 

 

Figure 8.  Walleye growth curves from Seminoe, Alcova and Glendo Reservoirs.

 

 

The escalating WAE population and size structure is likely negatively impacting the RBT management program.  Prior to the expansion of the WAE population, the relationship between angler PAS and the total number of 8-inch and larger RBT stocked the previous year showed a strong positive relationship.  When the data points for 2008 and 2009 are added representing an expanded WAE population in terms of both number and size, it seems that the presence of large WAE may have redefined the relationship between RBT stocking and PAS of RBT (Figure 9).

 

 

 

Figure 9.  Relationship between the total number of 8-inch and larger RBT stocked with the following year PAS, 1995 – 2009.  The circle denotes the 2008 and 2009 data points.

 

Hydroacoustic survey.  512

Alcova Reservoir was sampled with hydroacoustics on August 17, 2009.  Sampling occurred as it has for the previous eight years, following 15 parallel transects, equally spaced throughout the reservoir.  Sampling occurred between 0956 and 1459 hours under calm conditions.

 

Mean total pelagic fish density was 7.35/acre (90% CI: 6.3, 8.4), which was significantly lower than the 2007 estimate of 37.1/acre (one-way ANOVA; P < 0.001; Figure 10).  Hydroacoustic sampling was not conducted in 2008.  Based on a pelagic surface area of 1,974 acres (i.e., the same area as used for historic estimates), the estimated pelagic fish population was 14,516 individuals (90% CI: 12,364, 16,668).  Historically, hydroacoustic population estimates have not accounted for the length of individual transects when calculating the mean fish density.  To remedy this, the 2009 population estimates, as well as all future estimates, will use a weighted mean based on transect length.  Accounting for transect length, the mean total pelagic fish density was 7.53/acre (90% CI: 5.5, 9.5) and the estimated pelagic fish population size was 14,870 (90% CI: 10,903, 18,838).  The weighted values were not significantly different from the un-weighted values (one-way ANOVA; P = 0.89).

 

Historically, density and population estimates have been partitioned by species using data from purse seine sampling.  This was the first year since 2003 that purse seine sampling has occurred on Alcova Reservoir.  Although the purse seine catch was low (see details below), results were similar to those found in 2003 and prior.  It should be noted that the purse seine only samples fish within the top 60 ft of water; therefore, extrapolating the results to the entire water column may not be appropriate.  Hydroacoustic sampling indicated that approximately 94% of the pelagic fish population resided in the top 60 ft of water during daylight hours.  Purse seine samples indicate that 97% of the fish in the top 60 ft of the water column were RBT.  Assuming that purse seining does effectively sample the top 60 ft of the water column, the pelagic RBT population in Alcova Reservoir was approximately 13,994 individuals (90% CI: 10,261, 17,728).

 

There are several possible explanations for the vast difference in estimated pelagic fish density between 2007 and 2009.  Hydroacoustic data requires post-processing to remove extraneous data (e.g., background noise, repeated bottom echoes) and identify individual fish.  Historically, the majority of this post-processing was done by the reservoir research biologist and personal biases in regard to indentifying fish could have a large impact on population estimates.  This process was extremely time consuming, not easily repeatable, comprised many steps where human error could occur, and did not allow for standardization in post-processing.  To facilitate standardization and improve post-processing speed, “rules” were setup in the software allowing for much of the process to be automated.  Historic data sets back to 2004 will be re-processed using current techniques so that data can be more accurately compared among years.  Personal communication with FMCR indicates that the 2009 hydroacoustic population estimate follows predicted trends based on gill net and angler catch rates, and that the 2007 population estimate does not.  It is possible that the unusually high 2007 population estimate was a result of a difference in post-processing technique, an error in calculation, and/or a malfunction of the hydroacoustic equipment.  Further insight will likely be gained after the 2004–2007 datasets are reprocessed using current techniques.

 

The length of fish sampled with hydroacoustics can be estimated based on the strength of the returned echo.  Back-calculation of length can only be done for fish sampled by the downlooking transducer because the orientation of fish sampled by the sidelooking transducer is unknown.  In 2009, 41 fish were sampled with the downlooking transducer in Alcova Reservoir.  Back-calculated lengths were estimated and individuals were separated by those sampled shallower than 60 ft and those greater than 60 ft, because this is the sampling depth of the large purse seine.  There was no significant difference between individuals sampled shallower than 60 ft and those greater than 60 ft (one-way ANOVA; P = 0.82).  However, the majority of individuals sampled shallower than 60 ft were 3–5 in in length with the exception of a few larger (i.e., 14–22 in) fish (Figure 11).  Individuals found greater than 60 ft deep tended to be a little larger (i.e., 5–11 in) than individuals found shallower, and no fish greater than 15 in were sampled greater than 60 ft.  Because the species composition is unknown below 60 ft, the differences in size structure may be due to species composition or may be due to habitat preferences by certain size groups.

 

 

 

 

Figure 10.  Density estimates for pelagic fish from hydroacoustic surveys in Alcova Reservoir, 1998–2009.  All surveys were conducted in August, no survey was conducted in 2008, and error bars indicate 90% confidence intervals.  Note that the value indicated for 2009 is the un-weighted mean which is consistent with previous years.

 

Figure 11.  Estimated size distribution of fish sampled in Alcova Reservoir with downlooking hydroacoustics in 2009 separated by depths shallower and deeper than 60 ft.

 

 

Purse seine survey.  512

Alcova Reservoir was sampled with the large purse seine (LP; 775 ft long, 60 ft deep) on August 10 through August 13, 2009.  Historically, there were 10 standardized sites where purse seine sampling occurred; however, GPS locations for these sites were not available.  The 2009 sampling consisted of sampling eight of the historic sites and two sites which were moved slightly from historic sites to better distribute sampling throughout the reservoir.  Global positioning system locations were noted for all sample locations to facilitate standardization in the future.

 

In 10 purse seine hauls, 102 fish were sampled consisting of 99 RBT, one BNT, one WHS, and one CRP.  Although this catch seems low, the result was similar to historic catches.  The mean catch per haul for the fall LP from 1988 to 2003 was 18.7 ± 10.9 (mean ± SD) compared to the 2009 result of 10.2 ± 4.7 fish/haul (Figure 12).  The 2009 purse seine catch was not significantly different from the 1998–2003 fall LP catch (one-way ANOVA; P = 0.199).  The LP encircles an area of 1.1 acres; therefore, the estimated pelagic density was 9.3 ± 4.3 fish/acre. 

 

As RBT comprised the majority of the catch, the results focus on that species.  The length of RBT measured (n = 96) varied from 12.0 to 17.4 in (13.8 ± 1.2 in).  The majority (71%) of the RBT sampled were between 12.0 and 13.9 in (Figure 13).  The PSD of RBT sampled in the 2009 LP catch was 10; PSD-SQ was 90.  No preferred size or larger RBT were sampled in 2009. 

 

The weight of RBT measured (n = 44) varied from 0.69 to 1.68 lbs (1.03 ± 0.23 lbs).  Wr values varied from 77.2 to 109.8 (90.2 ± 6.2) indicating good overall condition.

 

 

Figure 12.  Mean number of individuals sampled per year in the large purse seine on Alcova Reservoir during the fall by species.  Error bars indicate one standard deviation.

 

Figure 13.  Length-frequency distribution of RBT sampled by LP from Alcova Reservoir, August 2009.

 

 

 

Seminoe Reservoir

 

Goal: Maintain Seminoe Reservoir to provide a trout and walleye sport fishery.

 

Objective: Maintain RBT FS PSD of 25 or greater and WAE ES PSD of 45 or greater.

 

·         Lake Surveys. 512

 

Annual standardized gill net sampling was conducted in June (floating) and September (sinking) targeting RBT and WAE respectively (Tables 11-14).

 

 

      Table 11. Number, CPUE (stdev), mean length (n; stdev) with ranges, and mean weight (n; stdev)           with ranges of fish captured in FG Seminoe Reservoir, June 2009.

 

Species	Number	CPUE	Mean Length	Range	Mean Weight	Range
BNT	25	0.09(0.09)	15.2(25;2.9)	10.8-21.0	1.43(25;0.82)	0.41-3.62
CRP	4	0.01(0.03)
LNS	1	0.01(0.01)
RBT	145	0.49(0.39)	15.7(143;2.0)	9.9-20.6	1.68(143;0.57)	0.47-4.89
WAE	36	0.12(0.20)	13.4(36;2.1)	10.2-21.0	0.85(36;0.58)	0.33-3.27
WHS	10	0.03(0.05)

 

 

Table 11a.  Relative Stock Density

 

Species	N>S	RSD-Q	RSD-P	RSD-M	RSD-T
BNT	25	80	48	4	0
RBT	142	60	1	0	--
WAE	36	8	3	0	--

 

 

      Table 12. Mean relative weights (n; stdev) with ranges and relative weights by length category (n;           stdev) for selected fish species in FG, Seminoe Reservoir, June 2009.

 

Species	Mean Wr	Range	S-Q	Q-P	P-M	M-T
BNT	84(25;7.5)	67-98	86(5;4.4)	83(8;7.1)	83(11;9.4)	85(1;--)
RBT	98(143;17.0)	62-253	101(57;9.3)	95(84;20.2)	67(1;--)
WAE	89(36;6.2)	78-101	88(33;6.1)	98(2;1.4)	90(1;--)

     

      Table 13. Number, CPUE (stdev), mean length (n; stdev) with ranges, and mean weight (n; stdev)           with ranges of fish captured in ER Seminoe Reservoir, September 16, 2009.

 

Species	Number	CPUE	Mean Length	Range	Mean Weight	Range
BNT	11	0.05(0.06)	17.3(10;2.3)	11.5-19.3	2.08(10;0.76)	0.48-2.86
CRP	5	0.02(0.04)
LNS	71	0.33(0.53)
RBT	22	0.10(0.09)	17.5(22;1.0)	14.9-19.1	2.27(22;0.33)	1.60-2.85
WAE	210	0.92(0.38)	14.6(210;4.8)	7.6-31.0	1.63(196;2.51)	0.18-13.80
WHS	318	1.41(0.81)

 

 

Table 13a.  Relative Stock Density

 

Species	N>S	RSD-Q	RSD-P	RSD-M	RSD-T
BNT	10	90	80	0	--
RBT	22	91	0	--
WAE	190	28	12	8	3

 

 

      Table 14. Mean relative weights (n; stdev) with ranges and relative weights by length category (n;           stdev) for selected fish species in all gear, Seminoe ReservoirWednesday, September 16, 2009.

 

Species	Mean Wr	Range	S-Q	Q-P	P-M	M-T
BNT	85(10;7.4)	76-95	77(1;--)	77(1;--)	87(8;6.9)
RBT	98(22;11.0)	78-127	96(20;9.3)	96(20;9.3)
WAE	87(196;8.9)	68-125	88(31;8.6)	88(31;8.6)	92(7;2.6)	95(11;6.8)

 

 

Walleye

WAE relative abundance has displayed an upward trend since 2004 (Figure 16).  While the difference in CPUE between adjacent years is not statistically significant, both 2008 and 2009 WAE CPUE is significantly greater than 2004 (One way ANOVA with post-hoc multiple comparisons (Fisher’s LSD) F=7.39, DF = 5, p < 0.001).  Given the year to year variability in year-class strength Seminoe is likely experiencing a long-term cyclical WAE abundance pattern, with 2008 likely representing the peak of abundance.

 

The management goal pertaining to WAE is to maintain a PSD of 45.  This goal is not currently being met (one sample test for proportion, p < 0.001).  In fact, the goal has been met only 2 out of the last 8 years (Figure 16).  The reasons for size structure meeting the goal was the result of two strong year-classes followed by one weak and one average year class (FMCR progress report 2007).

 

In order to test whether high mortality rates (presumably due to angling) are the basis for the general failure of the WAE population to meet management goals, otoliths were once again collected from Seminoe with emphasis on larger (older) fish.  The 2009 samples were pooled with otoliths collected in 2006 and 2008 (N=245) to construct both a von Bertalanffy growth curve (Figure 17) and an age-length key.  Given the high degree of variability in year-class strength, all WAE from 2005 through 2009 were pooled and treated as an average representation of the WAE population.  The age-length key was applied to the pooled data set (N= 847) to construct a catch curve.  Estimates of mortality (Z and A) were calculated using the weighted regression method (Maceina and Bettoli 1998) (Figure 18).  The target mortality cap (maximum Z at which PSD=45 can be maintained) was calculated from the von Bertalanffy coefficients (Miranda 2002).

 

 

Figure 16.  Mean WAE CPUE (1 s.d.) in fall ES and fall WAE PSD from 2002 through 2009.  Seminoe Reservoir.

 

 

 

Figure 17.  Von bertalanffy growth curve for Seminoe Reservoir WAE.

     

 

 

 

Figure 18.  Catch curve for Seminoe Reservoir WAE (pooled 2005 – 2009)

 

 

Given the slow growth of WAE in Seminoe, a PSD = 45 can only be met at Z < 0.35 or A < 0.30.  The mortality estimate for the entire WAE population age 3 and up (gear selectivity precludes WAE less than age 3) is A = 0.25 (r2 = 78.2%).  However, creel data from 2007 shows harvest is focused on fish 10-24 inches in length (ages 3 to 12).  Hence fish escaping this size range are effectively released from harvest pressures.  Visual inspection of Figure 18 seems to substantiate this argument as it appears the trend in data points hinges around age 13.  Taking this into account, mortality was recomputed for fish age 3 through 12 to better represent mortality of the fished component of the population.  For age 3-12 WAE, A = 0.34. 

 

Based on this cursory analysis, assuming additive not compensatory processes are in play, total fishing mortality rate would have to be reduced by at least 0.04 in order to meet the management goal for WAE size structure.  Depending on the proportion of total annual mortality that is fishing mortality (F), various decreases in total harvest would be needed (Table 15).

 

 

Table 15.  Theoretical harvest reductions needed to achieve PSD=45 under various levels of total annual fishing (F) and natural (M) mortality rates.

 

F	M	Harvest Reduction
0.10	0.24	40%
0.15	0.19	27%
0.20	0.14	20%
0.25	0.09	16%
0.30	0.04	13%

 

 

While exploitation is unknown from Seminoe, FAST incorporates models developed by various researchers predicting the instantaneous rate of natural mortality based on different population metrics; hence a reasonable estimate of the range of fishing mortality could be predicted.  Using these models, total annual natural mortality (M) would be expected to fall into the range from 0.15 – 0.23 (Table 16).

 

 

Table 16.  Model, parameters used, natural mortality (M) and fishing mortality (F).

 

Model	Parameters	M	F
Quinn and Deriso (1999)	Survival to Tmax and Tmax	0.23	0.19
Hoenig (1983)	Tmax	0.17	0.25
Jensen (1996)	K(Von bertelanffy growth coefficient)	0.15	0.27
Pauly (1980)	Linf (von Bertalanffy), K and mean temp	0.17	0.25
Chen and Watanabe (1989)	Initial age, Final age, K, T0	0.16	0.26

 

 

The dynamic pool model in FAST was used to model the Seminoe WAE population under various regulation scenarios to determine if achieving the goal for PSD is even realistic.  Given the varied year-class strengths exhibited, recruitment was customized within the model using the number of 10.0 – 10.9 inch fish captured in fall nets from 2004 – 2009 as an index of year to year recruitment variability.  Conditional fishing mortality (cf) was presumed to be zero for age 0 through age 2 fish.  Fishing mortality of age 3 through age 13 fish was modeled under low fishing (F = 0.19), moderate fishing (F = 0.23) and high fishing (F = 0.27).  Fishing mortality for age 14 and older fish was assumed to be negligible (F = 0.05).  The model was run for a period of 50 years, with the predicted PSD for the last eight years of each model compared with PSD from the previous 8 years of netting data.  No significant differences were evident either between models or compared with actual data (one way ANOVA with post-hoc multiple comparisons (F = 0.11, df = 3, p = 0.96) indicating all of the models give a reasonable approximation of actual conditions (Figure 19).  Both a 13 inch and 15 inch minimum length limit causes an increase in modeled PSD (Figure 19).  However, given the high degree of recruitment variability, PSD would be expected to fall below the goal approximately 5 out of every 10 years for the 13 inch minimum and 3 out of ten years for the 15 inch minimum.

 

 

 

Figure 19.  Modeled WAE PSD (1 s.d.) under high, medium and low fishing effort; current WAE PSD and modeled PSD under 13 and 15 inch minimum size limits.

 

 

While a 15 inch minimum size limit may improve size structure to the point where the management goal is realized more than 70% of the time, the restrictive nature of the regulation may not be received favorably by the angling contingent (the bulk of harvest is less than 15 inches currently).  Further compounding the argument is the fact that mercury levels in fish over 15 inches are quite high.  Requiring folks to restrict harvest to fish that are mercury laden poses an escalated level of philosophical debate, yet to be undertaken.  The most probable regulation type for Seminoe given high recruitment variability and slow growth rate would be a size range where harvest would be prohibited (would capture the 15 inch mark) with a restrictive harvest over the slot.  In order to model the potential regulations and scenarios, data on the actual fishing mortality partitioned by size class is needed.  Should management action be recommended in order to meet the goal for size structure, a fishing mortality study will have to be carried out to gauge size specific mortality rates including creel surveys to determine  angler sentiments regarding any potential  regulation change.

 

Trout

 

Abundance of RBT  in Seminoe continues its downward trajectory.  As reported in the 2008 progress report, the decreasing number of fish stocked annually, has continually negatively impacted the RBT population (Figure 20).  The model developed in 2008 to predict CPUE from the sum of the previous 2 years stocking predicted a mean CPUE in 2009 of 0.55 fish/hour.  The measured value of 0.49 is not statistically dissimilar from the predicted value (one sample t, t = -0.62, p = 0.55).  When incorporated into the existing model, data from 2009 resulted in an increase in model precision.  The updated model (Figure 21), predicts gillnet CPUE in 2010 to be 0.10.

 

 

 

Figure 20.  Gill net CPUE and number of RBT stocked the previous 2 years for Seminoe Reservoir, 2004 – 2009.

 

 

 

Figure 21.  Relationship between stocking and gill net CPUE for Seminoe Reservoir.

 

 

The goal for size structure (PSD >=25) is currently being exceeded.  The reason is, not surprisingly, a decrease in numbers stocked in 2008.  Given the short life span of RBT in Seminoe (3-5 years), PSD will be strongly influenced by numbers stocked.  Until stocking can be stabilized, PSD is not a good measure of management success.  Rather, goals pertaining to gillnet CPUE and angler PAS or catch rate should be explored.  Given the lack of creel data for this water, relationships between angler catch and either gillnet CPUE or number stocked cannot be developed at this time.  Future emphasis on programmatic collection of spot creel data should be incorporated into work schedules.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Pathfinder Reservoir

 

Objective: Maintain RBT FS PSD at least 40 and WAE ES PSD of at least 20.

 

·         Lake Survey. 512

Trend gill netting was conducted on Pathfinder Reservoir in June targeting trout with FS and in September targeting WAE with ES to conform to Standing Water Fishery Assessment guidelines.  Results are presented in Tables 25-28.  Pathfinder Reservoir water levels in 2009 were the highest they have been in since 2001, peaking at 71% capacity in November. The lowest water level in 2009 was 34% capacity, which was near the peak water level in 2008 (36%) and above the peak level in 2007 (28%).

 

 

 

 

 

 

 

 

 

 

 

Table 25. Number, CPUE (stdev), mean length (n; stdev) with ranges, and mean weight (n; stdev) with ranges of fish captured in FS, Pathfinder Reservoir, June 26 -29, 2009 (15 nets, 285 hours).

 

Species Number CPUE    Mean Length Range Mean Weight Range BNT 22 0.08 (0.09) 15.5 (22;1.76) 11.2 - 18.4 1.39 (22;0.33) 0.48 - 1.69 CRP 1 0.00 (0.01)                     LNS 1 0.00 (0.02)                     RBT 174 0.61 (0.41) 14.6 (173;2.33) 10.7 - 19.6 1.47 (173;0.66) 0.52 - 2.94 SRC 1 0.00 (0.01) 19.6   19.6 - 19.6 2.68   2.68 - 2.68 WAE 5 0.02 (0.03) 15.9 (5;3.40) 11.3 - 20.0 1.54 (5;0.94) 0.49 - 2.85 WHS 12 0.04 (0.08)                     TOTAL 216

 

 

Table 25a. Traditional Relative Stock Density.

 

Species	n >= S	RSD-Q	RSD-P	RSD-M	RSD-T
BNT	22	95	45
RBT	173	34
SRC	1	100	100
WAE	5	60

 

 

Table 26.  Mean relative weights (n; stdev) with ranges and relative weights by length category (n; stdev) for fish in FS, Pathfinder Reservoir, June 26 – 29, 2009.

 

Species	Mean Wr		Range			S - Q		Q - P		P - M		M - T
BNT	85	(22;14.0)	58	-	113	84		95	(11;9.3)	74	(10;9.3)	85	(22;14.0)
RBT	104	(173;11.1)	79	-	159	106	(115;10.7)	100	(58;10.8)
SRC	82		82	-	82					82
WAE	92	(5;4.9)	86	-	98	91	(2;7.1)	93	(2;6.4)	91

 

 

Table 27.  Number, CPUE (stdev), mean length (n; stdev) with ranges, and mean weight (n; stdev) with ranges of fish captured in ES, Pathfinder Reservoir,  September 21 - 24, 2009 (12 nets, 229 hours).

 

Species	Number	CPUE		Mean Length		Range			Mean Weight		Range
BNT	15	0.07	(0.06)	18.1	(14;1.68)	15.0	-	20.7	2.56	(14;0.74)	1.43	-	3.87
CRP	29	0.13	(0.15)
LNS	41	0.19	(0.18)
RBT	42	0.18	(0.11)	16.7	(40;1.93)	12.5	-	20.5	2.05	(40;0.71)	0.80	-	3.87
WAE	192	0.86	(0.40)	17.2	(192;3.93)	8.5	-	31.4	2.20	(191;2.14)	0.19	-	12.68
WHS	457	2.09	(1.54)
TOTAL	776

Table 27a. Traditional Relative Stock Density.

 

Species	n >= S	RSD-Q	RSD-P	RSD-M	RSD-T
BNT	14	100	86	14
RBT	40	63	3
WAE	188	79	13	5	2

 

 

Table 28.  Mean relative weights (n; stdev) with ranges and relative weights by length category (n; stdev) for fish in ES, Pathfinder Reservoir, September 21 - 24, 2009.

 

Species	Mean Wr		Range			S - Q		Q - P		P - M		M - T
BNT	94	(14;8.2)	82	-	110			98	(2;9.2)	93	(10;9.0)	94	(2;0.7)
RBT	98	(40;8.3)	79	-	118	101	(15;7.0)	96	(24;8.5)	86
WAE	93	(191;7.2)	68	-	118	92	(39;7.1)	93	(125;6.8)	93	(15;5.9)	101	(6;9.4)

 

 

Trout

Rainbow trout RSD-Q was 34 in 2009, a 45% decrease from the RSD-Q of 62 in 2008. Length frequencies indicate a greater number of RBT from the previous year’s stocking in 2009 and 2007 compared to 2006 and 2008 (Figure 25). The number of stocked RBT > 7 inches (< 5/lb) has varied widely from 2002-2008 and this variation influences rainbow trout RSD-Q the following year (Figure 26). Linear regression indicates that stocking of RBT > 7 inches from the previous two years also influences RBT CPUE (Figure 27; R² = 0.97;  p = 0.002). Similar strong relationships between stocking of RBT > 7 inches and relative abundance has been observed in sonar surveys of Pathfinder Reservoir (2007 Progress Report) and gill net catch in Seminoe Reservoir (2009 Progress Report). In September of 2009, 98,611 RBT > 7 inches were stocked in Pathfinder reservoir, an increase from 61,213 stocked in 2008. Applying the CPUE regression model, CPUE is predicted to increase to 0.68 fish per hour in 2010. Rainbow trout RSD-Q is expected to decrease in 2010.

 

Mean relative weight of RBT caught in spring netting was 104 in 2009, the highest value ever reported for Pathfinder Reservoir (Figure 28).  Rainbow trout were in excellent condition in 2009 despite high survival of fish stocked in 2008, suggesting that intraspecific competition is not limiting relative weight. Increased water levels and/or water clarity in Pathfinder Reservoir could have led to high Wr in 2009 by increasing forage resources. Improved condition may have accelerated growth, as peaks in the length frequency appear to have shifted to the right, suggesting increased length-at-age (Figure 25).

 

 

Figure 25. Length frequency of RBT captured in spring netting, Pathfinder Reservoir, 2006-2009.

 

Figure 26. RSD-Q and number of RBT > 7 inches stocked the previous fall, 2003-2009.

 

 

 

Figure 27. Linear regression of CPUE (fish/net/hour/) versus number of RBT > 7 inches stocked previous two years, 2004 and 2006-2009. Other years were excluded from CPUE analysis because netting did not conform to Standing Water Fishery Assessment guidelines.

 

 

 

Figure 28. Mean relative weights (Wr) for RBT captured in spring gill netting 2000-2009. Error bars are 95% confidence intervals.

 

 

Walleye

Walleye CPUE in 2009 (0.86) was the highest CPUE since standard fall netting protocols were implemented on Pathfinder Reservoir in 2004 (Figure 29). Most of the walleye captured were over 15 inches (quality size; Figure 30). RSD-Q in 2009 was 79. The proportion of the walleye population that are of quality to preferred size has generally increased in fall gill netting since 2004 (Figure 30). Walleye Wr was also the highest it has been since 2004 (Figure 31).  It appears, like the RBT population, Pathfinder Reservoir walleye are doing well and in good condition. There has been some interest in stocking GZS in Pathfinder Reservoir to provide supplemental forage for walleye. GZS have been stocked in the reservoir in the past, but do not overwinter. If water levels remain high, walleye should continue to be in good condition, and it will probably not be necessary to stock GZS.

 

 

Figure 29. Mean CPUE for walleye captured in fall gill netting of Pathfinder Reservoir, 2004-2009. Error bars are 95% confidence intervals.

 

 

Figure 30. Percent of walleye captured in sinking gill nets during fall that fit in to stock to quality (S-Q), quality to preferred (Q-P), preferred to memorable (P-M), memorable to trophy (M-T), and trophy (T) length classes.

 

 

Figure 31. Mean relative weight (Wr) for walleye captured in fall gill netting of Pathfinder Reservoir, 2004-2009. Error bars are 95% confidence intervals.

 

• Hydroacoustic survey.  512

Pathfinder Reservoir was sampled with hydroacoustics on July 21, 2009.  Sampling occurred as it has for the previous seven years, sampling the central portion of the reservoir where depths exceed 25 ft, with a zig-zag, 15 transect pattern.  Sampling occurred between 0839 and 1514 under mostly calm conditions with increasing winds in the afternoon.  Sampling was completed before wind speeds increased to the point of having a negative impact on sampling.

 

Historically, hydroacoustic transects on Pathfinder Reservoir have been partitioned into 500-m bins for statistical analysis, and only data from the sidelooking transducer were used to estimate density; to remain consistent, the 2009 data were analyzed in this manner.  Mean pelagic fish density in the top 20 ft was 4.59/acre (90% CI: 3.4, 5.8).  The data were not normally distributed, so a natural log plus one transformation was used to satisfy this assumption for statistical analysis; values were then back-transformed for reporting.  Mean pelagic fish density in the top 20 ft was significantly lower in 2009 than in 2007 (one-way ANOVA; P < 0.001; Figure 5).  Hydroacoustic sampling was not conducted in 2008.  To account for changes in reservoir content between years, population estimates are based on the surface area of water greater than 25 ft deep at the time of sampling.  The estimated pelagic fish population was 28,325 individuals (90% CI: 20,920, 35,730) which was significantly less (one-way ANOVA; P < 0.001; Figure 6) than the 2007 estimate of 54,100 individuals (90% CI: 46,500, 63,000).

 

Direct comparisons of the 2009 population estimate to past estimates should be viewed carefully.  Although the method used to estimate the population from the hydroacoustic data was the same, the method for post-processing the hydroacoustic data differed.  Personal communication with FMCR indicates that the 2009 hydroacoustic population estimate follows predicted trends based on gill net and angler catch rates.  Historic data sets back to 2004 will be re-processed using current techniques so that data can be more accurately compared among years.

 

Pathfinder Reservoir is the only waterbody in the state where hydroacoustic transects are partitioned into bins.  Although binning has a number of statistical advantages (e.g., increased sample size, reduced variability between samples), its most notable disadvantage is that binning does not use all of the data collected.  When a transect is divided into 500-m bins, the remainder of each transect is ignored.  In the 2009 data set, 16% of the data collected were not used in the analysis.  To remedy this, the 2009 population estimates, as well as all future estimates, will use entire transects as a sampling unit, weighted based on transect length (i.e., the same technique used on all other waters in the state).  Population estimates for Pathfinder Reservoir will continue to use only the data from the sidelooking transducer to minimize the inclusion of demersal, non-trout species (e.g., WAE, WHS).  Using this method, the mean total pelagic fish density in the top 20 ft was 4.45/acre (90% CI: 2.1, 6.8) and the estimated pelagic fish population size was 27,445 (90% CI: 13,190, 41,700).  Population estimates based on weighted transects were not significantly different from those based on binned units (one-way ANOVA; P = 0.65).

 

 

Figure 32.  Density estimates for pelagic fish in the top 20 ft from hydroacoustic surveys in Pathfinder Reservoir, 2003–2009.  All surveys were conducted in July, no survey was conducted in 2008, and error bars indicate 90% confidence intervals.  Note that the value indicated for 2009 was calculated using the method consistent with previous years.

 

 

Figure 33.  Hydroacoustic population estimates for pelagic fish in the top 20 ft from hydroacoustic surveys in Pathfinder Reservoir, 2003–2009.  All surveys were conducted in July, no survey was conducted in 2008, and error bars indicate 90% confidence intervals.  Note that the value indicated for 2009 was calculated using the method consistent with previous years.  The line plot represents the number of surface acres with a depth greater than 25 ft at the time of sampling.

 

 

 

 

 

 

 

 

 

 

 

Glendo Reservoir

 

 

Goal: Maintain angling opportunity for channel catfish.

 

Objective:  Evaluate natural and stocked channel catfish recruitment.

 

·         Stream Survey. 511

Stocking of CCF was halted in Glendo Reservoir and the N. Platte River upstream from 2001-2004 to investigate natural recruitment in the system. Early results of the study indicate that the CCF fishery cannot be sustained through natural recruitment (see 2006 and 2007 progress reports for more information).  Low numbers of CCF in 2009 Glendo Reservoir gill netting further corroborate evidence that natural CCF reproduction is limited (see Glendo Reservoir write-up in this progress report). However, CCF in the N. Platte-Glendo system are slow growing, making length-based age determination difficult. Also, sampling in the system has tended to be biased towards catching larger individuals.

 

A post-graduate researcher, Scott Carelton, with the Wyoming Cooperative Fishery and Wildlife Research Unit is investigating the use of stable isotopes as a tool for discerning hatchery-raised fish from wild fish. To collect samples for this study, FMCR, along with Dr. Carelton, electrofished the N. Platte River from the Pacificorp access area to the Bixby access area using two BF in close proximity. Over the 4-mile reach, 27 CCF were collected, all of which were sacrificed for the study. Otoliths from these fish have been sent to UC Davis for lazer ablation analysis.

 

 

Goal:  Maintain a good walleye fishery in Glendo Reservoir.

 

Objective:  Maintain April-July WAE catch rates of at least 0.3 per hour.

 

·         Spot Creel. 520

Spot creel surveys were conducted at Glendo Reservoir on May 14, May 17, and June 7, 2009.  The creel surveys were conducted using roadblock check stations. Catch rate (fish per angling hour), residency, and harvested fish lengths were obtained from interviewed anglers. Proportional Angling Success (PAS, proportion of anglers catching at least 0.3 walleye per hour) could not be calculated because some interviews were conducted on groups of anglers rather than individual anglers. Thus, the total number of anglers interviewed could not be determined. However, catch rate and harvest rates (fish/hour) could still be calculated. A total of 214 interviews were conducted over the three days (Table 29).

 

Table 29.  Number of interviews, WAE catch rate (stdev), and WAE harvest rate (stdev), and mean length of harvested WAE (n; stdev) by date for 2009 roadblock creel surveys conducted at Glendo Reservoir.

 

Date	Interviews	WAE catch rate	WAE harvest rate	Mean WAE length
5/14/2009	34	0.36 (0.58)	0.29 (0.51)	18.5 (28; 2.56)
5/17/2009	123	0.22 (0.42)	0.11 (0.23)	17.6 (74; 1.99)
6/7/2009	57	1.11 (1.17)	0.75 (0.85)	16.4 (25; 1.56)
Total	214	0.48 (0.81)	0.31 (0.58)	17.6 (127; 2.15)

 

 

Table 30.  Number of interviews, WAE catch rate (stdev) and WAE harvest rate (stdev), and mean length of harvested WAE (n; stdev) by year for June roadblock creel surveys conducted at Glendo Reservoir.

 

Year	Interviews	WAE Catch Rate	WAE Harvest Rate	Mean WAE length
2009	57	1.11 (1.17)	0.75 (0.85)	16.4 (25;1.56)
2008	183	1.15 (1.09)	0.56 (0.58)	15.2 (413;1.9)
2007	184	0.82 (1.03)	0.39 (0.50)	15.3 (301;2.2)
2004	284	0.37	0.15	16.2
2003	250	0.39	0.13	16.9
2002	267	0.83	0.46	15.5
2001	219	0.63	0.27	16.3
2000	272	0.74	0.38	16.7

 

 

Overall walleye angler catch rate (0.48 walleye/hour) exceeded the management goal of 0.30 walleye/hour (Table 29). Most of the walleye measured during the spot creel surveys were over 16 inches (Figure 34). Angler catch rates of WAE are often linearly related to measures of abundance such as mark-recapture estimates (Newby et al. 2000) or gill net CPUE (Isbell and Rawson 1989). To test for similar relationships between angler WAE catch rates from June roadblock creel surveys and standard gill netting CPUE at Glendo Reservoir, angler catch rate was regressed against gill net CPUE data from 2002-2004 and 2007-2009. No roadblock surveys were done in June in 2005-2006 and netting prior to 2002 did not conform to standards of the Standing Water Fishery Assessment manual. The linear regression indicated a positive relationship between catch rate of walleye and gill net CPUE. The equation of the regression line (R² = 0.669; p = 0.047) was angler catch rate = 0.9688(CPUE) - 1.0406. The intercept for this equation was not significantly different from zero (p = 0.183). Although this model is based on only six years worth of data, it appears that walleye gill net CPUE has the potential to be a good predictor of angler catch rates. More years of creel and gill net data will serve to refine the model and elucidate how angler catch rate is related to relative abundance. This data will also serve to track changes in angler catch and harvest of walleye under the new minimum length limit.

 

 

 

 

 

 

Figure 34.  Length frequency of harvested walleyes measured during 2009 Glendo Reservoir spot creel surveys (n = 127).

 

 

Objective:  Maintain a WAE ES PSD of at least 45.

 

·         Lake Survey. 512

Trend gill netting was conducted on Glendo Reservoir from July 27 - 29, 2009 targeting WAE, CCF, and YEP with 12 ES to conform to Standing Water Fishery Assessment guidelines. One net became snagged and had to be cut. Data from the snagged net were excluded from the CPUE analysis. Results are presented in Tables 31 and 32.

 

 

Table 31. Number, CPUE (stdev), mean length (n; stdev) with ranges, and mean weight (n; stdev) with ranges of fish captured in overnight ES, Glendo Reservoir, July 27 - 29, 2009.

Species	Number	CPUE			Mean Length				Range			Mean Weight		Range
BLC	1	0.00	(0.02)			13.3			13.3	-	13.3	1.03		1.03	-	1.03
CCF	9	0.04	(0.06)			25.6	(9; 2.15)		23.3	-	30.0	6.93	(9; 2.10)	4.80	-	11.48
CRP	134	0.75	(0.40)
GZS	32	0.18	(0.19)
NRH	23	0.13	(0.09)
QBK	88	0.47	(0.55)
RBT	1	0.01	(0.02)			19.5			19.5	-	19.5	3.14		3.14	-	3.14
STC	1	0.01	(0.02)			9.9			9.9	-	9.9	0.37		0.37	-	0.37
WAE	421	2.25	(0.40)			15.0	(421; 3.76)		7.7	-	27.5	1.33	(413; 0.96)	0.11	-	7.78
WHS	6	0.03	(0.03)
YEP	35	0.20	(0.22)			7.8	(35; 1.58)		5.2	-	10.8	0.26	(33; 0.16)	0.06	-	0.64
TOTAL	752

 

Table 31a. Relative Stock Density

Species	n >= S	RSD-Q	RSD-P	RSD-M	RSD-T
BLC	1	100	100	100
CCF	9	100	78	11
RBT	1	100
WAE	344	68	6	2
YEP	35	40	11

 

 

Table 32.  Mean relative weights (n; stdev) with ranges and relative weights by length category (n; stdev) for selected fish species in overnight ES, Glendo Reservoir, July 27-29, 2009.

Species	Mean Wr		Range			S - Q		Q - P		P - M			M - T
BLC	68		68	-	68							68
CCF	104	(9;4.1)	96	-	109			104	(2;4.9)	104	(6;4.6)	106
RBT	99		99	-	99			99
WAE	88	(413;7.4)	61	-	136	88	(111;5.5)	87	(211;5.7)	79	(16;5.7)	79		(6;8.1)
YEP	95	(33;10.0)	77	-	117	97	(19;11.4)	91	(10;7.3)	93	(4;6.1)

 

 

Walleye

Excluding the snagged gill net, average WAE catch rate was 2.25 fish/hour (Table 31). Comparing mean CPUE and bootstrap confidence intervals (α = 0.05) from all years in which netting was done according to the Standing Water Fishery Assessment Manual, 2009 mean CPUE was statistically different only from that of 2003 (Figure 35). Variability in WAE CPUE among nets was low in 2009 compared to other years.

 

In the 2008 progress report, it was noted that the 2007 WAE cohort was likely a weak year-class. Length frequency of WAE netted in 2009 shows a relative paucity of age 2 fish (13- 15 inches in length), confirming that the 2007 year-class was indeed weak (Figure 36). Because the WAE population is currently dominated by fish ≥ 15, RSQ-Q in 2009 (68) was well above the objective of 45. Pair-wise Z-tests were used to determine if RSD-Q was significantly higher in 2009 than 2004-2008. RSD-Q was significantly higher in 2009 than any of the previous 5 years (Z = 5.07 – 13.13; p values < 0.0001).

 

Average walleye relative weight was 88, exceeding the objective of 85. Thirty-two Gizzard shad were captured during netting operations, indicating they successfully overwintered during 2008-2009. Glendo Reservoir has not been stocked with GZS since 2002.

 

 

 

Figure 35. Walleye CPUE from standard gill netting in Glendo Reservoir 2002-2009. Error bars are bootstrap 95% confidence intervals.

 

 

 

Figure 36.  Length frequency of WAE in overnight ES at Glendo Reservoir, July 27-29, 2009 (n = 421).

 

 

Channel Catfish

A total of 9 CCF were captured for an average CPUE of 0.04 fish/hour (Table 31), the lowest CPUE since Glendo Reservoir netting protocols were changed in 2002 to conform to the Standing Water Fishery Assessment Manual.  Linear regression revealed that catch of CCF in standard gill netting has declined steadily since 2002 (R² = 0.87, p = 0.007; Figure 37). The smallest CCF captured in 2009 was 23 inches long. No stocking of CCF occurred between 2001 and 2004 in Glendo Reservoir or the North Platte River upstream. Previous investigations of length and age data indicate that natural CCF reproduction is limited in Glendo Reservoir and the North Platte system (see the write-up on the evaluation of naturally-recruited and stocked channel catfish in this progress report for more information). Channel catfish were stocked in Glendo Reservoir in 2005-2008, but it is unlikely that these fish would have reached 23 inches in 2009. Gill netting in Glendo Reservoir tends to selectively sample larger age classes, but gill net catches have historically contained at least a few CCF < 23 inches. Channel catfish stocked 2005-2008 should begin to show up in Glendo Reservoir gill nets in 2010.

           

 

Figure 37. Linear regression of annual average CPUE of channel catfish in standard Glendo Reservoir gill netting versus year. Error bars are bootstrap 95% confidence intervals.

 

 

Yellow Perch

A total of 33 YEP were captured for an average catch rate of 0.2 fish/hour of netting time (Table 31).  Since 2002, annual yellow perch CPUE has varied in cyclical pattern with a period of one or two years (Figure 38). Lengths of YEP ranged from 5.2 – 10.8 inches (Table 31).