Population Dynamics of Lake Whitefish in Lake Pend Oreille, Idaho

Abstract

To evaluate the biological potential of a commercial fishery for lake whitefish in Lake Pend Oreille, I estimated population attributes that regulate production and yield and constrain compensatory responses to exploitation. The length-frequency distribution based on trap netting in autumn 2005 was unimodal with a mean of 448 mm, whereas the length-frequency distribution based on gillnetting in spring 2006 was bimodal with a mean of 390 mm. The sex composition (proportion females) was skewed toward females (0.66; 95% CI = 0.63 – 0.70). The shape parameter of the weight-length equation (β = 3.363) was similar to other unexploited whitefish populations and indicated that whitefish grew plumper as length increased. The instantaneous growth coefficient for whitefish in Lake Pend Oreille (K = 0.12/year) was among the lowest recorded for unexploited whitefish across their range. Maturation occurred at an age (6 years, range = 4–12 years) and length (383 mm, 340–440 mm) that was relatively high for unexploited populations of whitefish. The instantaneous natural mortality rate (M = 0.167) was among the lowest observed among unexploited populations of the species. Population density of adult whitefish was low based on total surface area (1.35 fish/acre; 95% CI = 1.11–1.78 fish/acre), but average based on expected whitefish habitat ≤ 230 feet deep (4.07 fish/acre; 95% CI = 3.35–5.35 fish/acre). Density of immature and adult whitefish over expected whitefish habitat (11.93 fish/acre; 95% CI = 8.42–17.53 fish/acre) was within the range observed for unexploited populations of the species. Natural mortality was low, growth was slow, and maturity was late for whitefish in Lake Pend Oreille, which indicate a high biological potential for yield to a commercial fishery. To determine if subpopulations of whitefish were present in Lake Pend Oreille, Idaho, I compared population attributes and movement among putative subpopulations. To enable comparison, the population was divided into five putative subpopulations based on bathymetry and geographical separation. Mean length increased from north (445.3 mm) to south (459.2 mm) and size-structure differed significantly among putative subpopulations. Female proportion increased from north (0.59) to south (0.76). Body condition differed significantly among putative subpopulations and was highest in the south and lowest in the north. Length-at-age differed significantly among putative subpopulations, though the average maximum difference in length-at-age among putative subpopulations ranged from only 13 mm (SD = 5.2 mm) for the first 20 age classes to 19 mm (SD = 7.7 mm) for all age classes combined. Length- and age-at-50% maturity differed significantly among putative subpopulations, and increased from north (367.97 mm; 5.44 yr) to south (395.5 mm; 7.36 yr). Mortality rates did not differ significantly among putative subpopulations. Whitefish tended to be recaptured in the sampling area where they were tagged and overall fidelity was 81%. Recaptured fish tagged in the north, midwest, and mideast areas were predominantly recaptured (96.4%) within these three areas, and recaptures tagged in the mid-south were predominantly recaptured (85.1%) in the mid-south area, which suggests the presence of two discrete spawning subpopulations. Population attributes generally differed most from north to south, but were relatively similar among north, midwest, and mideast areas, further suggesting the presence of at least two spawning subpopulations. To provide harvest recommendations for optimizing yield of lake whitefish in Lake Pend Oreille, Idaho, I used a Beverton-Holt yield/recruit model to estimate yield for each fully recruited whitefish, calculated catch rates of whitefish in trap nets spatially and temporally, and estimated annual yield based on yield/recruit results and an estimate of adult age-structured abundance. Maximum yield/recruit (YPRMAX) was 111.37 g at an instantaneous fishing mortality rate of FMAX = 3.59 (annual fishing mortality rate = 63.2%) and tr = age 6. For tr ≥ age 6, yield/recruit could not be maximized for reasonable values of fishing mortality. Increases in yield/recruit above age 4 were small and yield/recruit was similar for tr = ages 4–8. Whitefish reached 50% sexual maturity between the ages of 6 and 8 years, and 95% of YPRMAX was at F = 0.50 (105.4 g) for tr = age 6, F = 0.70 for tr = age 7 (105.1 g), and F = 0.80 for tr = age 8 (101.1 g). Catch rate was highest at Sunnyside Bay (CPE = 69.7) in the north (CPE = 34.0) and lowest at Idlewilde Bay (CPE = 0.9) in the south (CPE = 1.8). Catch rate peaked during the weeks of 3 and 17 October, 14 November, and 12 December 2005. Harvesting fish older than age 7 at a fishing mortality rate of F = 0.5 would result in a 48.6% total annual mortality rate and an estimated annual yield of 86,512 lbs, similar to the annual commercial lake whitefish harvest from Red Lakes Minnesota (93,136 lbs). I conclude that the whitefish population in Lake Pend Oreille can sustain a commercial fishery if harvest is maintained within sustainable limits.