Introduction
Marine mammals are the top predators in their respective ecosystems, and changes in their abundance can have cascading impacts on ecosystems (Trites et al. 1999). There is a need to understand changes in abundance over time to provide harvest advice, monitor population recovery, and to evaluate how future perturbations in the ecosystem may impact the population. The estimated historical abundance of the Cumberland Sound (CS) beluga whale (Delphinapterus leucas (Pallas, 1776)) population in southeast Baffin Island, Nunavut, Canada, pre-commercial whaling, was approximately 8500 individuals (DFO 2005). Commercial whaling between 1880 and 1920, primarily by large offshore vessels from Europe, removed approximately 2000 beluga whales, and another 5700+ were removed from 1921 to 1945 during drives organized by the Hudson’s Bay Company (Stewart 2018). The majority of the commercial hunt occurred using drive-hunts, which drove beluga whales into the shallow waters in Clearwater Fiord, where they were stranded at low tide and harvested (Stewart 2018). From 1945 to 1960 most hunting of beluga whales was conducted by local Inuit for sale to Hudson’s Bay Company and approximately 2000 whales were hunted. This commercial harvest ended in 1966, but subsistence harvesting by community members from Pangnirtung has continued (Stewart 2018). The CS beluga population is co-managed by Fisheries and Oceans Canada (DFO), the Nunavut Wildlife Management Board (NWMB), Nunavut Tunngavik Inc. (NTI), and the Pangnirtung Hunters and Trappers Organization (HTO). Through co-management, decisions on quota and beluga whale sustainability are made considering all knowledge streams and through consultation between organizations. An annual quota has been in effect since 1980, and is currently set at 41.
Genetic and satellite telemetry data indicate that CS beluga whales form a distinct population that stay within CS year round (de March et al. 2002, 2004; Richard and Stewart 2009; Turgeon et al. 2012); however, traditional knowledge indicates that different types of beluga whales visit the Sound each year (Kilabuk 1998). In summer, beluga whales congregate in Clearwater Fiord on the northwest side of CS (Richard and Stewart 2009). The Pangnirtung HTO recognized the importance of Clearwater Fiord as a calving and nursing ground for beluga whales and implemented a ban on hunting in the area in 1985 (Richard and Pike 1993). Despite the implemented quota and regulated hunt location, recent population abundance estimates have been <2000 animals, and a 2009 survey had large confidence intervals (310–1679 belugas (Richard 2013)).
Population dynamics are typically modelled using catch records in conjunction with survey or mark–recapture-derived abundance estimates (Holt 2004). Accurate abundance estimates are particularly important as the CS beluga population was listed in 2017 as threatened under Canada’s Species at Risk Act (SARA), with an imminent recovery strategy set to be released in 2020. Reconstruction of the population dynamics since commercial exploitation has ended is important to accurately assess the impact whaling had on the ecosystem, and to establish a baseline for assessing population recovery. Accurate catch records and abundance estimates can provide an upper limit on historic population size, which can then be corrected for changing reproductive rates and natural mortality that are often assumed to be density dependent (Baker and Clapham 2004). These models can provide insight into how the population has changed over time. The goal of the present study was to update abundance estimates for the CS beluga population using new aerial survey data collected in 2014 and 2017, and to model the population trajectory since 1960.
Materials and methods
Aerial surveys
Visual and photographic aerial surveys were conducted in CS in 2014 and 2017 (Table 1). The study area was divided into three strata in 2014: North CS, which included Kangilo Fiord, West CS, and Clearwater Fiord (Fig. 1). In 2017, Kangilo Fiord was surveyed with greater coverage as a separate stratum, and the west CS stratum was extended farther south (Fig. 1). Surveys in both years were flown in a de Havilland Twin Otter (DH-6) equipped with four bubble windows and an optical glass-covered camera hatch on the rear underbelly of the plane.
[Image omitted: See PDF]
[Image omitted: See PDF]
Visual survey transects were flown in all but the Clearwater Fiord stratum at a target altitude of 1000 ft (305 m) and a ground speed of 100–110 knots (185–204 km/h). A double platform design in which two observers were seated at bubble windows on each side of the aircraft was used for visual surveys (Buckland et al. 2001). All four observers remained in their respective positions throughout the survey. The two observers on the same side of the aircraft were visually and acoustically isolated from one another to ensure independent observations. Sony PCM-D50 audio recorders were used to record beluga whale sightings including species, group size, and perpendicular declination angle to the center of each group, which was measured using a Peco DCC1 Digital Compass/Clinometer when the group was at beam (90° to plane) of the observer. A “group” was defined as animals within one body length of each other and behaving cohesively. The two primary observers at the front of the aircraft also recorded environmental conditions such as ice concentration (in tenths), sea state (Beaufort scale), fog (% of field of view), glare (% of forward field of view), and cloud cover (%).
Due to the high density of beluga whales that were anticipated in Clearwater Fiord, full coverage photographic surveys were conducted in that stratum and were flown 3–11 August in 2014 and 29 July–12 August in 2017, at a target altitude of 610 m and speed of 100–110 knots (185–204 km/h). Two Nikon D810 cameras, equipped with 25 mm lenses, mounted at the rear of the aircraft and directed straight down, captured complete photographic coverage of the stratum. Photographs were georeferenced using a GPS receiver (Bad Elf GPS Pro+) using a Bluetooth module accessory (Foolography Unleashed D200+ Bluetooth Module). At the target altitude of 610 m, the ground area covered by each photograph was 875.4 m × 585.2 m, resulting in 20% side overlap between photographs on adjacent transects. Variation in speed, altitude, and pitch of the aircraft resulted in the need to use a photographic interval of 7–8 s to achieve 20% overlap among photographs. For better water clarity conditions, surveys of Clearwater Fiord were flown to coincide with high tide when possible (Table 1).
Analysis
Visual survey
Distance sampling, which models sighting probability as a function of perpendicular distance from the trackline and adjusts counts for differences in detectability as a function of distance (Thomas et al. 2010), requires at least 60–80 observations for reliable estimation; when there are a low number of detections only one or two additional sightings close to the trackline can affect estimation substantially (Buckland et al. 2001). With <60 beluga whale sightings across all visual surveys, we decided to analyse the survey data using a strip-transect design. To determine strip width, we estimated perpendicular distances and used detection functions to assess the distance at which sightings were missed. Perpendicular distance was calculated from the declination angle to each sighting, which was measured by observers for all but one of the observations. The perpendicular distance of that one observation was estimated from an aerial photograph using:
[Formula omitted: See PDF]
where Ds is the distance of the sighting, XT is the total number of pixels in image widthwise, XS is the widthwise pixel count from the image’s outer edge to the sighting, and β is half the field of view of the lens.
For the 2014 survey, a hazard-rate detection function with a simple polynomial adjustment showed sightings were missed within the first 100 m of the aircraft and increasingly missed beyond 600 m (Fig. 2); thus, the strip was defined from 100 to 600 m (w = width) on either side of the aircraft. In 2017, a hazard-rate function with a cosine adjustment was fitted to the sightings data and showed there was a relatively uniform distribution of sightings from 0 to 300 m, but animals were increasingly missed beyond this distance (Fig. 2). Therefore, beluga whales observed within a 300 m (w) strip on either side of the aircraft were used to estimate near surface abundance.
[Image omitted: See PDF]
Duplicate observations, defined as observations between the primary and secondary observers occurring within 10 s of one another (Pike and Doniol-Valcroze 2015) and less than a 5° declination difference (Southwell et al. 2002), were removed. For the 2017 survey, 11 probable duplicate sightings between observers 1 and 2 were removed. Five of these were confirmed as duplicates from aerial photographs that were taken throughout the visual survey. The other six were outside of the frame of the photographs. After removal of duplicates, the probability density function of the perpendicular distances of beluga whale groups near the surface (f(y)) was estimated using an uniform function:
[Formula omitted: See PDF]
And the encounter rate (E(n)) was calculated using:
[Formula omitted: See PDF]
where [Formula omitted: See PDF] is the total number of detections from line i, where i is 1 to k (total number of lines). The total survey effort (L) is [Formula omitted: See PDF], where l is transect length. The cluster encounter rate variance was calculated following Fewster et al. (2009) for systematic variance where:
[Formula omitted: See PDF]
The density of beluga whales at or near the surface was estimated by:
[Formula omitted: See PDF]
And the total estimate of beluga whales at or near the surface is:
[Formula omitted: See PDF]
where A is the area of the survey stratum.
Double observer methods used in the 2017 survey allowed for estimation of perception bias (p(0)), which results from missed whales that are visible at the surface (Richard et al. 2010). To determine the value of p(0), duplicate sightings between the primary and secondary observers on each side of the aircraft were identified as defined above. Detection probabilities of observers can be correlated because of factors such as groups size and, therefore, a point independence model that assumes detections are independent only on the track line was used (see Buckland et al. (2009) for details). The 2014 survey was not adjusted for perception bias as there were too few sightings (n = 25 observations in total) and only a 14% re-sighting rate.
Near-surface abundance estimates were also adjusted to account for beluga whales that were diving too deep to be seen by observers (i.e., availability bias). Availability bias was estimated by taking the weighted average of time spent in the 0–1 m depth bin, 0–2 m bin, 0–3 m bin, 0–4 m bin, and 0–6 m bin for three beluga whales tagged with satellite linked time depth recorder tags in Clearwater Fiord in July 2006 (n = 1) and August 2007 (n = 2) (Richard and Stewart 2009; for tagging methodology see Orr et al. (2001)). Weighted averages were based on the number of 6 h blocks collected for each beluga whale. Standard errors were calculated using a weighted standard deviation divided by the square root of the number of beluga whales used in each calculation. Beluga whales can be seen to depths of 5 m when the water is clear (Richard et al. 1994), but tags were not programmed to collect information from 0 to 5 m; thus, an average of the availability bias for 0–4 m and 0–6 m was used to estimate the 0–5 m availability bias. The larger of the two standard errors was used as the standard error for the interpolated 0–5 m bin (Richard 2013). The availability bias correction factor, Ca, was calculated as:
[Formula omitted: See PDF]
For visual surveys, which were conducted over water we classified as clear (i.e., able to see the entire body of beluga whale under the water up to 5 m depth (Richard 1991)), we used a correction factor calculated using the interpolated 0–5 m bin.
The total estimate of beluga whales at or near the surface was adjusted to account for perception and availability biases using:
[Formula omitted: See PDF]
where Cp in 2014 was one by default as it could not be calculated.
The final abundance estimate had an associated variance calculated using the delta method (Buckland et al. 2001) where:
[Formula omitted: See PDF]
Confidence intervals were calculated assuming a log-normal distribution as suggested in Buckland et al. (2001).
Photographic survey
Photographs were georeferenced and examined in ArcMap 10.1 (ESRI) by two experienced readers. Water clarity was evaluated in each photograph by looking at the proportion of the beluga whale body that could be seen in the photograph. An instantaneous availability bias correction factor for the 0–1 m bin was used if the photograph reader could only see the portion of the beluga whale body that was breaking the water surface, whereas the 0–2 m bin was used for photographs where the reader could see the tail and (or) the head of the beluga whales under the water up to a depth of 0–2 m (Richard 1991). Photograph readers made a decision about whether a 0–1 or 0–2 m adjustment should be used for each photograph; however, we also compared their assessment to RGB pixel colour of the water in each photograph. RGB pixel colours of brown hues matched readers 0–1 m adjustment decision, whereas blue hues aligned with their 0–2 m adjustment.
The surface area covered by each photograph was calculated as:
[Formula omitted: See PDF]
where length = altitude/Fs × Ls and width = altitude/Fs × Ws, and Fs is the focal length of the camera sensor (25 mm), Ls is the length of the camera sensor (35.9 mm) and Ws is the width of the camera sensor (24 mm).
To accurately determine beluga whale density, photograph area with sun glare, which masked beluga whale presence and was not searched, and land cover were cropped from total photograph area using shapefiles. Overlapping sections between adjacent photographs were also cropped from the first of each pair. The remaining area of searchable water was termed Anoglare.
The number of beluga whales detected near or at the surface in each photograph was adjusted for the instantaneous availability bias, Ca described above, according to water clarity.
An adjusted abundance estimate was calculated using:
[Formula omitted: See PDF]
where [Formula omitted: See PDF] is the total number of beluga whales detected at the surface for each photograph.
Beluga whale density was calculated by dividing the summed beluga whale count by the summed area of glare-free water. Density was then multiplied by the stratum area to obtain near-surface abundance estimates:
[Formula omitted: See PDF]
where Astratum is the area covered by the survey and I is the number of photographs per survey.
Within a single complete coverage photographic survey there is no variance associated with the encounter rate as all observable whales are counted. As a result, the only variance within a single repeat is the variance in the availability bias correction factor. However, four repeats of the Clearwater Fiord Stratum were conducted in 2014 and during the second visual survey in 2017 (5–12 August) and therefore the abundance estimate was an average of the four repeats. The variance of this mean estimate was calculated using (equation 8.8, Buckland et al. 2001):
[Formula omitted: See PDF]
where Ai is the area of the ith repeat of the photographic survey.
Total population abundance for 2014 and 2017 was estimated by adding the individual stratum estimates for each survey, with a variance calculated as the sum of the individual stratum variances.
Population modelling
A stochastic surplus-production model, assuming density dependence acting on the population growth rate was fitted by Markov Chain Monte Carlo (MCMC) Bayesian methods to aerial survey (1990–2017) and reported harvest data (1960–2018) (Tables 2 and 3) (Doniol-Valcroze et al. 2012). All aerial survey data were adjusted for availability bias, but only the 2017 survey was adjusted for perception bias (Table 2). For years when harvest numbers were not reported (n = 8) it was assumed that the quota was taken (41 animals; Table 3). Observation error (associated with data collection and abundance estimation) was separated from the process error (arising from natural variability in population dynamics) using a hierarchical state-space model that considers survey data to be the outcome of two distinct stochastic processes: a state process and an observation process (de Valpine and Hastings 2002). The state process describes the underlying population dynamics and the evolution of the true stock size over time, using a discrete formulation of the Pella–Tomlinson model (Pella and Tomlinson 1969; Innes and Stewart 2002) modified to allow the process error to have either positive or negative impact on the growth rate:
[Formula omitted: See PDF]
where λmax is the maximum growth rate or rate of population increase, K is the environmental carrying capacity, θ defines the shape of the density-dependent function, εp is a stochastic term for the process error, Rt are the removals for that year, calculated as reported catches, Ct, that are corrected for the proportion of animals that were struck and lost, S&L:
[Formula omitted: See PDF]
[Image omitted: See PDF][Image omitted: See PDF]
The observation process describes the relationship between true population size and observed data. In our model, aerial survey estimates St are linked to population size Nt by a multiplicative error term, [Formula omitted: See PDF]:
[Formula omitted: See PDF]
The model was run 200 000 times with a 20 000 burn-in and 30 thinning.
A series of trial models were run to narrow in on parameters for the final population model, but the model was not sensitive to changes in the prior parameters (DFO data on file). For marine mammals, maximum productivity is thought to occur between 50% and 85% of carrying capacity (Taylor and DeMaster 1993), but different studies have used theta values ranging between 1 and 3, which results in maximum productivity occurring between 50% and 63% (Wade 1998; Hobbs et al. 2006; Jackson et al. 2016; Heide-Jørgensen et al. 2017). We examined if theta was fixed and if it was allowed to vary between 1 and 3, and set theta to 1 in the presented model. For odontocetes, the maximum rate of population increase λmax is thought to lie between 1.02 and 1.06, with most studies using 1.04 (Wade and Angliss 1997; Wade 1998; Lowry et al. 2008). We allowed the model to estimate λmax using a uniform prior with minimum and maximum values of 1.01 and 1.05.
Reported harvests underestimate the number of beluga whales killed because some animals are wounded or killed but cannot be recovered (i.e., S&L). There are no data on S&L rates for CS; thus, we allowed the model to estimate S&L, basing it on the eastern Hudson Bay beluga whale assessment, using a moderately informative prior following a Beta(3, 4) distribution, with a median of 0.42 (42%) and quartiles of 0.29 (29%) and 0.55 (55%) (Marcoux and Hammill 2016).
The stochastic process error terms εp were given a log-normal distribution with a zero location parameter. The precision parameter for this log-normal distribution was assigned a moderately informative prior following a Gamma (1.5, 0.00005) distribution. These parameters were chosen so that the resulting error would have a coefficient of variation of 1% as we do not have evidence that large mammal stock dynamics are highly variable (Erb et al. 2001).
Uncertainty associated with each aerial survey was incorporated into the fitting process by guiding the formulation of the prior distribution of the survey error. The survey error term [Formula omitted: See PDF]followed a log-normal distribution with a zero location parameter. Its precision parameter was given a moderately informative prior following a Gamma (2.5, 0.4) distribution. These parameters were chosen so that the resulting CV of the survey estimates would have quartiles Q1 and Q3 of 0.35 and 0.55.
A Gibbs sampler algorithm, implemented in JAGS (Plummer 2003), was used to define posterior estimates for all model parameters. R2jags (Plummer et al. 2006) and coda (Su and Yajima 2015) packages developed in the R programming language (R Core Team 2019) were used to examine the results. We tested for mixing of the chains using Geweke’s test of similarity between different parts of each chain (Geweke 1996). The Brooks-Gelman-Rubin (BGR) diagnostic, which compares the width of the 80% credible interval of pooled chains with the mean of widths of the 80% credible interval of individual chains was assessed for convergence between chains (Brooks and Gelman 1998). The relative contributions of the parameters to the model were examined by estimating the pD value, which is the “effective” number of parameters being fitted (Spiegelhalter et al. 2002).
Results
Availability and perception biases
Availability bias adjustment factors were 2.54 (CV = 0.052) for the 0–5 m bin, 4.46 (CV = 0.117) for the 0–1 m bin, and 2.06 (CV 0.056) for the 0–2 m bin.
In 2017, both the primary and secondary observers missed observations at the track line (g(0)), but based on all observations from 0 to 550 m, the observers had probabilities of detection (p(0)) of 0.78 ± 0.158. The estimated p(0) of the two observers combined was 0.95 ± 0.070, which resulted in a perception bias adjustment factor (Cp) of 1.05 (CV = 0.077).
Aerial surveys
In 2014, the estimated number of beluga whales near the surface for two repeats of the visual survey outside of Clearwater Fiord was 389 (CV = 0.48) and 41 (CV = 0.57) (Fig. 1). The average estimated number of beluga whales near the surface from photographic surveys inside Clearwater Fiord was 228 beluga whales (Fig. 3). Abundance estimates adjusted for availability bias resulted in an estimated total abundance of 1151 (CV = 0.214; 95% CI = 760–1744 (Table 4)) beluga whales.
[Image omitted: See PDF][Image omitted: See PDF]
In 2017, the estimated number of beluga whales near the surface for two visual surveys outside of Clearwater Fiord was 543 (CV = 0.502) and 35 (CV = 0.487) (Fig. 1). The average estimated number of beluga whales near the surface from photographic surveys inside Clearwater Fiord was 601 beluga whales (Fig. 3). Average abundance estimates adjusted for perception and availability bias produced a corresponding abundance estimate of 1381 (CV = 0.043; 95% CI = 1270–1502 (Table 4)) beluga whales.
Population modelling
Priors and posteriors for the population dynamics model are shown in Fig. 4 and Table 5. In this model, each chain for the variables carrying capacity (K), population size in 2018 (N2018), process error, initial population size (N1960), and S&L rate showed rapid convergence. The BGR values for this model were all equal to one, indicating convergence of the chain (Brooks and Gelman 1998), and the Geweke’s Convergence Diagnostic Z-scores were between −1.96 and 1.96.
[Image omitted: See PDF][Image omitted: See PDF]
The model shows the population has been declining since 1960 (Fig. 5) and some updating of the starting population prior with an estimated median starting population of 2884 and a carrying capacity (K) of 7875. There was no updating of the prior for λmax with a median of 1.03, but there was some updating to the prior for S&L, from a median of 0.42–0.36. The estimated median population abundance for 2018 (N2018) is 1090 animals. The median abundance estimate for beluga whales in 2018 was not sensitive to changes in prior parameters for theta, carrying capacity, growth rate, or initial population size (DFO data on file).
[Image omitted: See PDF]
Discussion
Survey estimates from this study were very similar between 2014 and 2017, which is not unusual for large-bodied mammals with low rates of population increase (Wade and Angliss 1997; Erb et al. 2001). This provides confidence in our estimates, and suggests most of the population was surveyed. The model suggests a decline in the population over ∼60 years from approximately 2900 in 1960 to 1090 in 2018; from 1960 to 1989 when surveys are lacking, the model relies on the harvest data, although the harvest data are less informative in recent years since the quota has been reduced, and therefore the survey results constrain the estimation. The confidence intervals on the model projections are large and encompass the confidence limits of all the survey estimates.
Availability bias adjustment factors were based on only three tagged animals from 2006 to 2007. As there were only three tags, the variability in the adjustment factor is relatively low; therefore, for areas outside Clearwater Fiord we used the larger standard error of the 0–4 m and 0–6 m depth bins, rather than averaging them, which resulted in a slightly larger variability in the availability bias adjustment factor. Future efforts to deploy tracking devices on a larger number of beluga whales, particularly at the same time as an aerial survey, are needed to provide a better representation of dive behaviour of the whole population to improve adjustment factors. In addition, time–depth recorder dive data would provide detailed dive cycle data for these beluga whales and would be more appropriate for adjusting perception bias of visual surveys (Hagihara et al. 2014). The 2017 survey abundance estimate was the only survey that was adjusted for perception bias. As perception bias can be impacted by individual observers and weather conditions, we did not adjust other surveys using the adjustment factor for the 2017 survey. Adjusting for perception bias generally increases the estimated abundance by ∼5% compared with the other survey estimates and, thus, this should not impact the trajectory of the population model because it only changes the abundance estimate by ∼70 individuals, which is within the coefficient of variation for the abundance estimate.
The estimated abundance from these surveys only pertains to the areas where the transects occurred, and there has not been extrapolation to other areas. The 2014 and 2017 surveys covered what has previously been identified as the CS beluga population’s summer range (Richard and Stewart 2009); however, after a request by the Pangnirtung HTO, the 2014 survey was expanded south, and in 2017 another three transect lines were added to the southwest portion of CS. In 2017, whales were observed on the most southerly transect line early in the survey period suggesting that whales may have still been moving into Clearwater Fiord in late July (Fig. 1). Previous reconnaissance surveys farther south have failed to find beluga whales, but no reconnaissance surveys were flown in 2014 or 2017. Surveying a greater area of CS, in particular, farther southeast has been suggested for future surveys by the Pangnirtung HTO who think the survey may be missing whales at the southern edge of their range (Pangnirtung HTO, personal communication, 2019). If the survey has not captured the entire summer range for this population, there may be an underestimate of the population size.
Various iterations of the population model were examined before narrowing in on parameters that reduced the correlation between model parameters and made biological sense regarding what is known about the CS beluga population. We allowed the model to estimate the maximum growth rate, but limited the upper range to 5% to reflect potential stress-induced impacts on reproductive success (discussed below (Hobbs et al. 2006; Wade et al. 2012; Trana 2014)). We set upper limits on the priors for carrying capacity taking into account earlier assessments that suggested the starting population of CS beluga in 1960 was much lower than carrying capacity. Based on hunt records, abundance in 1923 was estimated at over 5000 beluga whales (Mitchell and Reeves 1981). Commercial whaling, which ended in 1939, resulted in a depleted CS beluga population (Mitchell and Reeves 1981). Carrying capacity for this population was estimated in a previous assessment by Alvares-Flores (DFO, unpublished data, 2004) as between 7000 and 8000 individuals. The population model presented here also suggests a carrying capacity of almost 8000 individuals. All models we investigated indicated the population has declined since 1960.
In the population model we only used abundance estimates for surveys that had been conducted systematically and from 1990 onwards. Surveys completed from 1980 to 1990 were re-examined for possible inclusion in the model fitting. Overall, they had little impact on our current understanding of the population. Prior to 1990, it was thought that most beluga whales were distributed within Clearwater Fiord, so coverage outside of this area was irregular and often limited to coastal areas. In addition, these surveys did not provide complete photographic coverage of Clearwater Fiord. The surveys flown in 1985 and 1986 provided more extensive systematic coverage outside of Clearwater Fiord and similar coverage within the Fiord, but whales appear to have been detected within a very restricted area and no estimate of survey variance was provided (Richard et al. 1990). Overall, these surveys are likely negatively biased and provide support for a minimum population size estimate (see Table 2 for survey estimates). In 1990, based on recommendations from hunters, the survey coverage was extended to other parts of CS and the extended coverage has continued since then (Kilabuk 1998; Richard 2013). Since 1990, the number of whales outside of Clearwater Fiord but within surveyed areas has varied, ranging from 15% of the total population estimate in 1999 to >60% in 2009 (Table 1), underlining the importance of timing of the survey and contemporaneously surveying areas outside of Clearwater Fiord.
The population model presented here indicates the CS beluga population has declined since 1960, despite harvest controls implemented by the co-management team (Pangnirtung HTO, NTI, NWMB, and DFO) that have been in place since 1979. Unsustainable subsistence harvest levels are contributing to the decline. Given the estimated maximum rate of increase of 3% for the CS beluga population, which equals a net increase at best of ∼33 individuals annually, for an abundance estimate of approximately 1100 individuals from the population model, the current annual quota of 41 is too high to allow for the recovery of this population. The model estimated a S&L factor of approximately 0.36, which is lower than that reported for hunts on Baffin Bay beluga whales (0.41; Innes and Stewart (2002)); however, there are limited data from the CS beluga harvest to inform the priors on S&L. Data on how many animals are struck and survive are difficult to collect. However, many individuals within the CS beluga population display persistent hunting scars (DFO, unpublished data, 2019), suggesting some animals are struck and survive. The estimated population growth rate cannot compensate for harvests of 41 landed animals annually, plus an estimated 15 S&L (36% of 41 landed).
The current harvest in Pangnirtung is not sustainable; however, other factors may also be impacting the dynamics of this population. In Cooke Inlet, Alaska the beluga whale population declined in abundance by almost 50% between 1994 and 1998, and this decline was attributed to the subsistence harvest (Hobbs et al. 2006). However, the harvest in Cook Inlet was voluntarily ended in 1999, and the population has still not increased in size, suggesting other factors may be limiting population growth (Hobbs et al. 2006). St. Lawrence beluga whales have also not shown signs of recovery in spite of complete protection from harvesting since 1979 (Hammill et al. 2007). Changes in habitat and high levels of pollution and disturbance may also be affecting the recovery of these two populations. The St. Lawrence beluga population is exposed to shipping impacts in one of the busies waterways in the world, which negatively impacts their ability to forage through echolocation and communicate among each other (Gervaise et al. 2012). Cumberland Sound beluga are experiencing changes in habitat as a result of climate change, but have little exposure to pollution or shipping traffic. However, other factors, such as depensation (i.e., the Allee effect), and ecological and environmental factors such as predation (Higdon and Ferguson 2009), changes in the CS food web (Yurkowski et al. 2017), and competitive pressure with commercial fisheries (Galappaththi et al. 2019), may be resulting in additive mortality to the population. Significantly higher stress levels, as indicated by higher blubber cortisol concentrations, in CS beluga whales compared with whales from the eastern Beaufort Sea, High Arctic, or western Hudson Bay (Trana 2014), could reflect one or more of these factors. Blubber cortisol levels have been increasing since the 1980s (Trana 2014), and could now be an additional factor contributing to reproductive suppression (Dobson and Smith 2000).
Depensation, when a decrease in the number of breeding individuals in a population leads to reduced production or survival of offspring, results in the continued decline of a population (Liermann and Hilborn 2001). Depensation can arise when populations are so small that other factors such as reduced probability of fertilization results in population growth rates decreasing as abundance declines (Liermann and Hilborn 2001). Reduced probability of fertilization could result from individuals being unable to find mates at small population sizes. Breakdown of social structure and function in small populations may also contribute to lower rates of population growth and contribute to depensation (Wade et al. 2012; although see Hobbs et al. 2006). Genetic diversity and inbreeding may also prevent recovery of small populations; however, Lande (1991) found that populations with an effective size of even a few dozen individuals is usually large enough to avoid most deleterious effects of inbreeding. The CS beluga population, which still numbers over 1000 individuals, is likely too large for inbreeding depression or reduced fertilization to be limiting recovery.
Ecological and environmental factors, such as climate-driven shifts in forage fish abundance, have also been identified as potential factors that may be limiting recovery (Carter and Nielsen 2011). Cumberland Sound is located at the southern periphery of the Arctic, and there is evidence of climate-induced changes in lower trophic level dynamics over the past few decades. There has been a documented increase in capelin (Mallotus villosus) presence in CS (Rikardsen et al. 2007; McKinney et al. 2012; Yurkowski et al. 2017), and in beluga whale diet (Watt et al. 2016; Yurkowski et al. 2017). In Hudson Bay an increase in capelin has been correlated with a reduction in Arctic cod (Boreogadus saida) (Gaston et al. 2003) a common prey item of beluga whales (Loseto et al. 2009; Quakenbush et al. 2015). Although capelin and Arctic cod have similar caloric values (Hop and Gjøsæter 2013), further information about the energetic expense of capturing and handling Arctic cod and capelin is needed to understand the impacts of an increase of capelin in CS beluga whale diet. In addition to capelin and Arctic cod, dive behaviour suggests that Greenland halibut (Reinhardtius hippoglossoides) is another important prey item for CS beluga whales, particularly in the autumn and winter seasons (Watt et al. 2016). Competitive pressure from the active commercial halibut fishery in CS during winter, with an annual quota of 500 tonnes since 1998 (DFO 2008) is also possible, although interactions between the fishery and beluga whales have not been studied to date.
Higher than expected rates of mortality from predation or ice entrapments would also cause population decline. Increases in killer whale (Orcinus orca) sightings in the Canadian Arctic over the last decade have also been attributed to climate change (Higdon and Ferguson 2009), although the opposite of this general trend has been noted in CS, where killer whales were seen annually in the 60s and 70s, but less frequently since 1977 (Reeves and Mitchell 1988). In other areas of the Canadian Arctic, beluga whales and narwhals are the most commonly taken prey by killer whales (Higdon et al. 2011). In 2002, 10 killer whales were seen exiting Clearwater Fiord and hunters saw evidence of killer whale predation on both beluga and bowhead whales (Balaena mysticetus) (Stewart and Savard 2003). A group of four killer whales were also seen in the Sound in August 2008 (Diemer et al. 2011), but there were no reports of beluga whale carcasses after their exit. The magnitude of predatory pressure on CS beluga whales from killer whales remains unknown. Beluga whale ice entrapment events are not frequent but are known to occur in CS. In 1956, 100 beluga whales were reported entrapped but since then only small entrapments with a few individuals have been reported (Kilabuk 1998; Richard and Stewart 2009). Thus, it is unlikely entrapments have limited population recovery, although future recovery plans should consider entrapments as they could be significant given the high site fidelity of the entire CS beluga population.
The CS beluga population has continued to decline after cessation of the commercial harvest, which removed unsustainable numbers of whales from the population. Recovery strategies for the population will need to consider all factors that may be impacting this population and should create an integrated plan with the community of Pangnirtung, Nunavut, to ensure restoration of the ecological integrity of the CS ecosystem, including increasing the beluga whale population, while balancing food security. Future research should focus on gathering more information on demographic parameters such as reproductive rates, and age and sex composition of the harvest to improve the model estimates and allow for better co-management of the population.
Acknowledgements
Thanks to the community of Pangnirtung and the Pangnirtung Hunters and Trappers Association for their recommendations and assistance with the surveys. Thanks also to the Polar Continental Shelf Program, the Nunavut Implementation Fund (DFO), Species at Risk (DFO) and the Nunavut Wildlife Management Board for financial support. We thank the late M. Kingsley for developing the initial population model, and T. Doniol-Valcroze, A. Mosnier, and R. Hobbs for model improvements, L. Montsion for photographic analysis, and two anonymous reviewers for comments that improved the final paper.
Baker C.S. and Clapham P.J. 2004. Modelling the past and future of whales and whaling. Trends Ecol. Evol. 19: 365–371.
Brodie, P.F., Parson, J.L., and Sergeant, D.E. 1981. Present status of the white whale (Delphinapterus leucas) in Cumberland Sound, Baffin Island. Reports of the International Whaling Commission 31. pp. 579–582.
Brooks S.P. and Gelman A. 1998. General methods for monitoring convergence of iterative simulations. J. Comput. Graph. Stat. 7: 434–455.
Buckland, S.T., Anderson, D.R., Burnham, K.P., Laake, J.L., Borchers, D.L., and Thomas, L. 2001. Introduction to distance sampling: estimating abundance of biological populations. Oxford University Press, New York, N.Y., USA. 432 pp.
Buckland S.T., Laake J.L., and Borchers D.L. 2009. Double-observer line transect methods: levels of independence. Biometrics, 66: 169–177.
Carter B.T.G. and Nielsen E.A. 2011. Exploring ecological changes in Cook Inlet beluga whale habitat though traditional and local ecological knowledge of contributing factors for population decline. Mar. Policy, 35: 299–308.
de March B.G.E., Maiers L.D., and Friesen M.K. 2002. An overview of genetic relationships of Canadian and adjacent populations of belugas (Delphinapterus leucas) with emphasis on Baffin Bay and Canadian eastern Arctic populations. NAMMCO Sci. Publ. 4: 17–38.
de March B.G.E., Stern G., and Innes S. 2004. The combined use of organochlorine contaminant profiles and molecular genetics for stock discrimination of white whales (Delphinapterus leucas) hunted in three communities on southeast Baffin Island. J. Cetacean Res. Manage. 6: 241–250.
de Valpine P. and Hastings A. 2002. Fitting population models incorporating process noise and observation error. Ecol. Monogr. 72(1): 57–76.
DFO. 2005. Recovery potential assessment of Cumberland Sound, Ungava Bay, Eastern Hudson Bay and St. Lawrence beluga populations (Delphinapterus leucas). DFO Canadian Science Advisory Secretariat Science Advisory Report 2005/036.
DFO. 2008. Cumberland Sound Greenland halibut (turbot) inshore fishery. DFO Canadian Science Advisory Secretariat Science Advisory Report 2008/040.
Diemer K.M., Conroy M.J., Ferguson S.H., Hauser D.D.W., Grgicak-Mannion A., and Fisk A.T. 2011. Marine mammal and seabird summer distribution and abundance in the fjords of northeast Cumberland Sound of Baffin Island, Nunavut, Canada. Polar Biol. 34: 41–48.
Dobson H. and Smith R.F. 2000. What is stress, and how does it affect reproduction? Anim. Reprod. Sci. 60: 743–752.
Doniol-Valcroze, R., Gosselin, J.-F., and Hammill, M.O. 2012. Population modeling and harvest advice under the precautionary approach for eastern Hudson Bay beluga (Delphinapterus leucas). DFO Canadian Science Advisory Secretariat Research Document 2012/168. iii + 31 pp.
Erb J., Boyce M.S., and Stenseth N.C. 2001. Population dynamics of large and small mammals. Oikos, 92: 3–12.
Fewster R.M., Buckland S.T., Burnham K.P., Borchers D.L., Jupp P.E., Laake J.L., and Thomas L. 2009. Estimating the encounter rate variance in distance sampling. Biometrics, 65(1): 225–236.
Galappaththi E.K., Ford J.D., Bennett E.M., and Berkes F. 2019. Climate change and community fisheries in the Arctic: a case study from Pangnirtung, Canada. J. Environ. Manage. 250: 109534.
Gaston A.J., Woo K., and Hipfner J.M. 2003. Trends in forage fish populations in northern Hudson Bay since 1981, as determined from the diet of nestling thick-billed murres Uria lomvia. Arctic, 56: 227–233.
Gervaise C., Simard Y., Roy N., Kinda B., and Ménard N. 2012. Shipping noise in whale habitat: characteristics, sources, budget, and impact on belugas in Saguenay-St. Lawrence Marine Park hub. J. Acoustic Soc. Am. 132: 76–89.
Geweke, J. 1996. Evaluating the accuracy of sampling-based approaches to the calculation of posterior moments. In Bayesian statistics 4. Edited by J.M. Bernardo, J.M. Berger, A.P. Dawid, and A.F.M. Smith. Oxford University Press, Oxford, UK. pp. 169–193.
Global Administrative Areas. 2009. GADM database of Global Administrative Areas, version 1.0. Available from gadm.org.
Hagihara R., Jones R.W., Grech A., Lanyon J.M., Sheppard J.K., and Marsh H. 2014. Improving population estimates by quantifying diving and surfacing patterns: a dugong example. Mar. Mammal Sci. 30: 348–366.
Hammill, M.O., Measures, L.N., Gosselin, J.-F., and Lesage, V. 2007. Lack of recovery in St. Lawrence estuary beluga. DFO Canadian Science Advisory Secretariat Research Document 2007/026. i + 19 pp.
Heide-Jørgensen M.P., Hansen R.G., Fossette S., Nielsen N.H., Borchers D.L., Stern H., and Witting L. 2017. Rebuilding beluga stocks in West Greenland. Anim. Conserv. 20: 282–293.
Higdon J.W. and Ferguson S.H. 2009. Loss of Arctic sea ice causing punctuated change in sightings of killer whales (Orcinus orca) over the past century. Ecol. Appl. 19: 1365–1375.
Higdon J.W., Hauser D.D.W., and Ferguson S.H. 2011. Killer whales (Orcinus orca) in the Canadian Arctic: distribution, prey items, group sizes, and seasonality. Mar. Mammal Sci. 28: E93–E109.
Hobbs, R.C., Shelden, K.E.W., Vos, D.J., Goetz, K.T., and Rugh, D.J. 2006. Status review and extinction assessment of Cook Inlet belugas (Delphinapterus leucas). AFSC Processed Report 2006-16. Alaska Fisheries Science Center, NOAA, National Marine Fisheries Service, Seattle, Washington, USA. 74 pp.
Holt S.J. 2004. Counting whales in the North Atlantic. Science, 303: 39–40.
Hop H. and Gjøsæter H. 2013. Polar cod (Boreogadus saida) and capelin (Mallotus villosus) as key species in marine food webs of the Arctic and the Barents Sea. Mar. Biol. Res. 9: 878–894.
Innes S. and Stewart R.E.A. 2002. Population size and yield of Baffin Bay beluga (Delphinapterus leucas) stocks. NAMMCO Sci. Publ. 4: 225–238.
Jackson J.A., Carroll E.L., Smith T.D., Zerbini A.N., Patenaude N.J., and Baker C.S. 2016. An integrated approach to historical population assessment of the great whales: case of the New Zealand southern right whale. R. Soc. Open Sci. 3: 150669.
Kilabuk, P. 1998. Final report on a study of Inuit knowledge of the southeast Baffin beluga. Report prepared for the Southeast Beluga Management Committee. Nunavut Wildlife Management Board, Iqaluit, Nun., Canada. 74 pp.
Lande, R. 1991. Applications of genetics to management and conservation of cetaceans. Reports of the International Whaling Commission Special Issue 13. pp. 301–311.
Liermann M. and Hilborn R. 2001. Depensation: evidence, models and implications. Fish Fish. 2: 33–58.
Loseto L.L., Stern G.A., Connelly T.L., Deibel D., Gemmill B., Prokopowicz A., et al. 2009. Summer diet of beluga whales inferred by fatty acid analysis of the eastern Beaufort Sea food web. J. Exp. Mar. Biol. Ecol. 374: 12−18.
Lowry L.F., Frost K.J., Zerbini A., DeMaster D., and Reeves R.R. 2008. Trend in aerial counts of beluga or white whales (Delphinapterus leucas) in Bristol Bay, Alaska, 1993–2005. J. Cetacean Res. Manage. 10: 201–207.
Marcoux M. and Hammill M.O. 2016. Model estimates of Cumberland Sound beluga (Delphinapterus leucas) population size and total allowable removals. DFO Can. Sci. Advis. Sec. Res. Doc 2016/077: iv + 35 p.
MacLaren Atlantic Ltd. 1978. Aerial surveys 77–2, 77–3, 77–4. Studies of seabirds and marine mammals in Davis Strait, Hudson Strait and Ungava Bay. February, 1978. Dartmouth, Nova Scotia.
McKinney M.A., McMeans B.C., Tomy G.T., Rosenberg B., Ferguson S.H., Morris A., et al. 2012. Trophic transfer of contaminants in a changing Arctic marine food web: Cumberland Sound, Nunavut, Canada. Environ. Sci. Technol. 46: 9914–9922.
Mitchell, E., and Reeves, R.R. 1981. Catch history and cumulative catch estimates of initial population size of cetaceans in the eastern Canadian Arctic. Reports of the International Whaling Commission 31. pp. 645–682.
Orr J.R., Joe R., and Evic D. 2001. Capturing and handling of White Whales (Delphinapterus leucas) in the Canadian Arctic for instrumentation and release. Arctic, 54: 299–304.
Pebesma E. 2018. Simple features for R: standardized support for spatial vector data. R J. 10(1): 439–446.
Pella, J.J., and Tomlinson, P.K. 1969. A generalized stock production model. Inter-Am. Trop. Tuna Comm. Bull. 13(3): 420–496. Available from aquaticcommons.org/id/eprint/3536.
Pike, D., and Doniol-Valcroze, T. 2015. Identification of duplicate sightings from the 2013 double-platform High Arctic Cetacean Survey. DFO Canadian Science Advisory Secretariat Research Document 2015/034. v + 22 pp.
Plummer, M. 2003. JAGS: A program for analysis of Bayesian graphical models using Gibbs sampling. In Proceedings of the 3rd International Workshop on Distributed Statistical Computing (DSC 2003), 20–22 March 2003, Vienna, Austria. Available from http://mcmc-jags.sourceforge.net/.
Plummer, M., Best, N., Cowles, K., and Vines, K. 2006. CODA: convergence diagnosis and output analysis for MCMC. R News, 6: 7–11. Available from cran.r-project.org/doc/Rnews/Rnews_2006-1.pdf#page=7.
Quakenbush L.T., Suydam R.S., Bruyan A.L., Lowry L.F., Frost K.J., and Mahoney B.A. 2015. Diet of beluga whales, Delphinapterus leucas, in Alaska from stomach contents, March–November. Mar. Fish. Rev. 77(1): 70–84.
R Core Team. 2019. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available from R-project.org/.
Reeves R.R. and Mitchell E. 1988. Distribution and seasonality of killer whales in the eastern Canadian Arctic. Rit Fiskideildar, 11: 136–160.
Richard P.R. 1991. Abundance and distribution of narwhals (Monodon monoceros) in northern Hudson Bay. Can. J. Fish. Aquat. Sci. 48: 276–286.
Richard, P.R. 2013. Size and trend of the Cumberland Sound beluga whale population, 1990 to 2009. DFO Canadian Science Advisory Secretariat Research Document 2012/159. iii + 28 pp.
Richard, P.R., and Orr, J.R. 1986. A review of the status and harvest of white whales (Delphinapterus leucas) in the Cumberland Sound area, Baffin Island. Can. Tech. Rep. Fish. Aquat. Sci. 1447: iv + 25 p.
Richard P.R. and Pike D.G. 1993. Small whale co-management in the Eastern Canadian Arctic: a case history and analysis. Arctic, 46(2): 138–143.
Richard, P.R., and Stewart, D.B. 2009. Information relevant to the identification of critical habitat for Cumberland Sound belugas (Delphinapterus leucas). DFO Canadian Science Advisory Secretariat Research Document 2008/085. iv + 24 pp.
Richard, P.R., Orr, J.R., and Barber, D.G. 1990. The distribution and abundance of beluga, Delphinapterus leucas, in eastern Canadian waters: a review and update. In Advances in research on the beluga whale, Delphinapterus leucas. Edited by T.G. Smith, D.J. St. Aubin, and J.R. Geraci. Can. Bull. Fish. Aquat. Sci. Vol. 224, pp. 23–38.
Richard P.R., Laake J.L., Hobbs R.C., Heide-Jørgensen M.P., Asselin N., and Cleator H. 2010. Baffin Bay narwhal population distribution and numbers: aerial surveys in the Canadian high Arctic, 2002–04. Arctic, 63(1): 85–99.
Richard P., Weaver P., Dueck L., and Barber D. 1994. Distribution and numbers of Canadian High Arctic narwhals (Monodon monoceros) in August 1984. Meddelelser om Gronland, Bioscience, 39: 41–50.
Rikardsen A.H., Dempson J.B., Amundsen P.-A., Bjorn P.A., Finstad B., and Jensen A.J. 2007. Temporal variability in marine feeding of sympatric Arctic char and sea trout. J. Fish Biol. 70: 837–852.
Southwell C., de la Mare W., Underwood M., Quartararo F., and Cope K. 2002. An automated system to log and process distance sight-resight aerial survey data. Wildlife Soc. Bull. 30: 394–404.
Spiegelhalter D.J., Best N.G., Carlin B.P., and van der Linde A. 2002. Bayesian measures of model complexity and fit. J. R. Stat. Soc. Ser. B Stat. Methodol. 64(4): 583–639.
Stewart, D.B. 2018. Commercial and subsistence catches of beluga whales (Delphinapterus leucas) from Cumberland Sound, Nunavut, 1840–2016. Canadian Technical Report of Fisheries and Aquatic Sciences 3250. 99 pp.
Stewart, D.B., and Savard, T. 2003. An overview of the ecology of Cumberland Sound, Baffin Island, Nunavut. Unpublished report prepared by North/South Consultants Inc., Winnipeg, Man., Canada, for Fisheries and Oceans Canada. 106 pp.
Su, Y.-S., and Yajima, M. 2015. R2jags: using R to Run ‘JAGS’. R package version 0.5-7. Available from CRAN.R-project.org/package=R2jags.
Taylor B.L. and DeMaster D.P. 1993. Implications of non-linear density dependence. Mar. Mammal Sci. 9(4): 360–371.
Thomas L., Buckland S.T., Rexstad E.A., Laake J.L., Strindberg S., Hedley S.L., et al. 2010. Distance software: design and analysis of distance sampling surveys for estimating population size. J. Appl. Ecol. 47: 5–14.
Trana, M.R. 2014. Variation in blubber cortisol as a measure of stress in beluga whales of the Canadian Arctic. M.Sc. thesis, University of Manitoba, Winnipeg, Man., Canada. 74 pp.
Trites, A.W., Livingston, P.A., Vasconcellos, M.C., Mackinson, S., Springer, A.M., and Pauly, D. 1999. Ecosystem change and the decline of marine mammals in the Eastern Bering Sea: testing the ecosystem shift and commercial whaling hypotheses. Fisheries Centre Research Reports. Vol. 7, No. 1, 106 pp.
Turgeon J., Duchesne P., Colbeck G., Postma L., and Hammill M. 2012. Spatiotemporal segregation among summer stocks of beluga (Delphinapterus leucas) despite nuclear gene flow: implication for the endangered belugas in eastern Hudson Bay (Canada). Conserv. Genet. 13(2): 419–433.
Wade P.R. 1998. Calculating limits to the allowable human-caused mortality of cetaceans and pinnipeds. Mar. Mammal Sci. 14(1): 1–37.
Wade, P.R., and Angliss, R.P. 1997. Guidelines for assessing marine mammal stocks: report of the GAMMS Workshop, April 3–5, 1996, Seattle, Washington. NOAA Technical Memorandum NMFS-OPR-12. U.S. Department of Commerce.
Wade P.R., Reeves R.R., and Mesnick S.L. 2012. Social and behavioral factors in cetacean responses to overexploitation: are Odontocetes less resilient than Mysticetes? J. Mar. Biol. 212: 567276.
Watt C.A., Orr J., and Ferguson S.H. 2016. A shift in foraging behaviour of beluga whales Delphinapterus leucas from the threatened Cumberland Sound population may reflect a changing Arctic food web. Endang. Species Res. 31: 259–270.
Yurkowski D.J., Hussey N.E., Fisk A.T., Imrie K.L., Tallman R.F., and Ferguson S.H. 2017. Temporal shifts in intraguild predation pressure between beluga whales and Greenland halibut in a changing Arctic. Biol. Lett. 13: 20170433.
Cortney A. Watt [email protected]
Freshwater Institute, Fisheries and Oceans Canada, Winnipeg, MB R3T 2N6, Canada.
Department of Biological Sciences, University of Manitoba, Winnipeg, MB R3T 2N2, Canada.
Marianne Marcoux
Freshwater Institute, Fisheries and Oceans Canada, Winnipeg, MB R3T 2N6, Canada.
Department of Biological Sciences, University of Manitoba, Winnipeg, MB R3T 2N2, Canada.
Steven H. Ferguson
Freshwater Institute, Fisheries and Oceans Canada, Winnipeg, MB R3T 2N6, Canada.
Department of Biological Sciences, University of Manitoba, Winnipeg, MB R3T 2N2, Canada.
Mike O. Hammill
Fisheries and Oceans Canada, Mont-Joli, QC G5H 3Z4, Canada.
Cory J.D. Matthews
Freshwater Institute, Fisheries and Oceans Canada, Winnipeg, MB R3T 2N6, Canada.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© 2021. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Abstract
Current scientific evidence indicates that the threatened Cumberland Sound beluga whale (Delphinapterus leucas (Pallas, 1776)) population is genetically differentiated and spatially segregated from other beluga whale populations. This population has been hunted for subsistence for centuries by Inuit who now live in the community of Pangnirtung, Nunavut, Canada, and was harvested commercially from 1860 until 1966. The commercial harvest removed at least 10 000 individuals from the population. Visual and photographic aerial surveys were flown during August 2014 and 2017 and produced beluga whale abundance estimates of 1151 (CV = 0.214; 95% confidence interval (CI) = 760–1744) and 1381 (CV = 0.043; CI = 1270–1502), respectively. Long-term trends in abundance were examined by fitting a Bayesian surplus-production population model to a time series of abundance estimates (n = 5), flown between 1990 and 2017, taking into account reported subsistence harvests (1960–2017). The model suggests the population is declining. Engaged co-management of the Cumberland Sound beluga population and information on demographic parameters, such as reproductive rates, and age and sex composition of the harvest, are needed to restore the ecological integrity of the Cumberland Sound marine ecosystem.