ArticlePDF Available

An approach to minimise bias in estimation of the length-frequency distribution of bowhead whales (Balaena mysticetus) from aerial photogrammetric data

Authors:

Abstract

Past attempts to estimate the length structure of the Bering-Chukchi-Beaufort (B-C-B) stock of bowhead whales have yielded quite different results from one study to the next because of size segregation on the summering areas or because of size segregation during the spring migration combined with inconsistent sampling during the migration period. A new approach is presented to document the length-frequency distribution of the B C B stock using length measurements from 3107 whale images collected during the spring migrations of 1985, 1986, 1989-1992 and 1994. This method provides estimates of the proportion of calves (length <6m), subadults (length 6-<13m), and adults (length ≥13m). The data from all years are combined by weekly period, and a bootstrap sampling procedure is used to construct the length-frequency by week. The distributions for each week are then combined to obtain the overall distribution, with each week’s contribution being in proportion to the fraction of the migration estimated from ice-based census studies to pass during that week. Corrections for differential detectability of mother/calf pairs and for calves born after they pass Pt Barrow, Alaska, are allowed for in the analysis. This new approach eliminates some of the biases that affected past attempts to estimate the length-frequency distribution for the B-C-B population of bowhead whales. It is robust to inclusion or exclusion of data for any given year and the time interval chosen to define repeat images. The new approach estimates a slightly higher proportion of subadults and lower proportion of adults in the population than most previous studies. The proportion of calves is also lower, but that is suspected to result from our inability to accurately estimate the proportion of the migration late in the season when many of the mother-calf pairs pass Pt Barrow. These late season migrants have not been accounted for during past photogrammetry studies or the ice-based census. Although the results do not differ substantially from those of most previous studies, sensitivity analyses indicate that several biases existed in the previous methods, but largely cancelled each other out.
A preview of the PDF is not available
... (4) the length-frequency of the population in recent years (e.g. Angliss et al., 1995;Koski et al., 2006); and (5) the estimates of abundance from photogrammetry (e.g. da Silva et al., 2000;Schweder, 2003) were not included in the likelihood function used when estimating model parameters during the 1998 assessment. ...
... The estimable parameters of this growth model are the lengths at ages 1 and 40 (ages chosen to encompass the bulk of the ages represented in the length-at-age data set), k, b and s. The mean length of a calf is set to 4.54m, the mean length of calves in the data set analysed by Koski et al. (2006). This assumption is, however, inconsequential for the analyses of this paper because the population dynamics model is fitted to data for animals aged 1 and older only. ...
... A variety of papers have documented the methods employed (e.g. Koski et al., 2006), but briefly the surveys were conducted from fixed-wing aircraft with search effort focused along open water areas, especially near the land-fast ice edge. A variety of ways exist for analysing the data from these surveys. ...
Article
Full-text available
The 1998 assessment of the Bering-Chukchi-Beaufort (B-C-B) Seas stock of bowhead whales (Balaena mysticetus) was conducted using a Bayesian estimation framework. That assessment ignored information on the length-frequency and age-composition of the harvests and the detailed length-frequency information from photogrammetry studies. The modelling framework used to assess the B-C-B Seas bowhead whales is therefore extended to make use of these data. The results indicate that selectivity is not uniform, as assumed in previous assessments, but rather domed-shaped, with young animals most vulnerable to harvest. The length-frequency, proportion, age-composition and abundance data are inconsistent to some extent. Fitting the model to the age-composition data leads to the most pessimistic estimates of stock status and productivity. The results of projections based on these assessments in which strike limits are set using the Bowhead Strike Limit Algorithm (SLA) suggest that none of the data sets are such that the scenarios considered when testing the Bowhead SLA should now be considered implausible.
... Other evidence, such as the current rate of population increase (Givens et al., 2016), percent calves in surveys (Clarke et al., 2022;Koski et al., 2006), the current harvest rate, and survival rate (da Silva et al., 2007;Schweder et al., 2010;Zeh et al., 2002), all suggest that a 3-year interval is the most plausible. ...
... It has long been known from earlier ice-based and aerial surveys that calf production varies widely among years, and that these differences are real and not an artifact of sampling or survey effort (George et al., 2004;Koski et al., 2006). If half of the BCB population are females, and about 40% are mature (Angliss et al., 1995), then a 3-year calving interval suggests calf production should be approximately 7%. ...
Article
Data from Bering‐Chukchi‐Beaufort Seas bowhead whales ( Balaena mysticetus ), harvested during 1973–2021 by aboriginal subsistence hunters, were used to estimate reproductive parameters: length at sexual maturity (LSM), age at sexual maturity (ASM), pregnancy rate (PR), and calving interval. Sexual maturity ( N = 187 females) was determined from the presence/absence of corpora in the ovaries, or a fetus. Using sampling bias‐corrected logistic regression, LSM was estimated at 13.5 m, 95% CI [13.0, 13.8]. There was a downward trend in LSM over time, statistically significant with one method but marginal with another. A growth model translated this estimate to an ASM estimate of 23.5 years, 95% CI [20.4, 26.7]. Pregnancy rate was determined from mature females ( N = 125), and from a subset limited to certain autumn‐caught whales ( n = 37) to reduce bias. The PR was estimated at 0.46 globally, 95% CI [0.36, 0.55] and 0.38 for the autumn sample, 95% CI [0.20, 0.51]. Both estimated PRs are consistent with a 3‐year calving interval, because the larger estimate includes two cohorts of pregnant whales harvested in spring, and bowhead whale gestation is longer than 12 months. These analyses represent the most conclusive empirical estimates of ASM, LSM, and PR for this bowhead whale stock from the largest available data sets to date.
... Frames were categorized as either "available" or "not available" based on whether the dugong was flat, horizontal, or straight (neither twisted nor inclined) and whether its extremities were distinct. These criteria are commonly used to estimate the body length of whales (Koski et al., 2006;Christiansen et al., 2018). Examples of rated photographs are provided in Figure 2. The length of the straightline from the tip of the snout to the medial notch of the fluke was measured. ...
Article
In this study, we describe the population characteristics and residency patterns of dugongs (Dugong dugon) across two intertidal seagrass beds in Talibong Island, Thailand: Site A, covering an area of 2.0 × 105 m2, and Site B, covering an area of 2.8 × 105 m2. Transect and individual identification surveys were conducted under clear water conditions using drones: 16 separate days over 11 months at Site A and 10 separate days over 3 months at Site B. Sixty-four individuals were identified from 180 videography sessions. The results confirmed at least two distinct patterns of seagrass habitat utilization among sites located approximately 5 km apart. Site A was characterized by a lower population density, higher year-round site fidelity, occupancy by relatively large individuals, and an absence of feeding aggregations. In contrast, Site B was characterized by a higher population density, lower site fidelity, occupancy by individuals with a wider range of body lengths, and the presence of feeding aggregations. The average population density at Site B was three to five times higher than that at Site A. Site A had a median nearest neighbor distance of 320 m with no significant bias in its distribution, whereas Site B had a median of 20 m with a significant bias. The mean site fidelity index for Site A (0.62 ± 0.08; n = 16) was significantly higher than that for Site B (0.39 ± 0.14; n = 10). Dugongs at Site A might have monopolized this site to some extent, while those at Site B might have benefited from increased opportunities for social interaction provided by aggregations. These findings highlight the importance of fine-scale monitoring of feeding ground utilization by dugongs, taking into consideration individual-specific details such as body lengths and resighting rates for a better understanding of their spatial distribution.
... However, maximum lengths of whales measured photogrammetrically are about 8.2% shorter than those of hunter-harvested whales, as there appears to be some "stretching" when whales are hauled onto shore (George et al. 2004). The longest whale measured in photographs was 17.57 m (Koski et al. 2006). ...
Chapter
Full-text available
Bowhead whales ( Balaena mysticetus ) have adopted growth and reproductive strategies to survive in a challenging environment where no other mysticete whales reside. They grow slowly, become sexually mature at around 25 years (later than other mammals), and do not give birth until they have sufficient energy reserves for the best possible chance of survival of the calf to weaning and long-term survival of the mother. To compensate for late maturity and long inter-birth intervals, some seem to have the capability to live to 200+ years of age, making them the longest-lived mammal known to date. Bowhead whale males have large testes per body size, and it is hypothesized that the basic polygynandrous system of females and males mating with multiple partners per estrous cycle allows for males to not compete violently against each other. Instead, they use sperm competition by volume of sperm for enhanced capability to father as many offspring as possible. Also, as in right whales ( Eubalaena spp.), the length of the penis is proportionally longer than those of balaenopterids. Details of sperm volume, potential variabilities of sperm viabilities, and actual paternities are unknown, but some patterns can be inferred from the closely related right whales with similar morphologies.
... musculus) Aerial photos; this study 0.90%, n = 11 3.51%, n = 8 Blue whale (B. musculus) an age-class is over-represented (Angliss et al. 1995;Perryman and Lynn 2002;Koski et al. 2006), or when the total length is underestimated because the animal is not observed in a perpendicular position from the aircraft (Cubbage and Calambokidis 1987;Angliss et al. 1995;Cosens and Blouw 2003). Likewise, aerial photogrammetry is limited in obtaining length information, because it is generally not possible to identify individuals from the air, with exception of bowhead whales (Balaena myticetus; Rugh et al. 1992) and right whales (Eubalaena sp.; Kraus et al. 1986;Khan et al. 2022) and relate them with sighting and biological data base. ...
Article
Full-text available
Length measurement of individuals provides useful information for biological and ecological studies and is instructive in estimating and parameterization of population dynamics and to identify changes in population structure. This study presents a photogrammetric method to estimate the total length of blue whales from a boat, using sequential images taken to measure the exposed body flank of the whale, from the blowhole to the dorsal fin, while simultaneously measuring the perpendicular distance between the whale and the camera to determine the scale of the image. The photo sequences were joined and measured with specialized software. Total length was predicted from blowhole-dorsal fin length using body proportion, mainly from whaling data. A Bayesian model calculated uncertainty in predicted total length due to errors in the photo-sequence process and variability in body proportion. Nearly all the uncertainty in total length was due to variability in the photo-sequence measurements of blowhole-dorsal fin length rather than in body proportion. The precision of estimates of total length depended on the number of replicate photo-sequences and size of whale but generally were in the range of 2-3%. A comparison of the total lengths of four individuals simultaneously measured by replicated photo-sequences and aerial pho-togrammetry showed that the photo-sequence method was unbiased. The photo-sequence method applied to 169 individuals sighted during 2006-2015 in the Gulf of California revealed an average length of 20.49 m and a range of 7.45-29.01 m, which falls within the known length range for Northeast Pacific blue whales. The photo-sequence method can be carried out in conjunction with other field studies of the species as it does not modify sampling logistics, nor add to the cost of the study. It also has the advantage of being adaptable to other cetaceans that show natural flank marking.
... Historically, aerial imagery from crewed aircraft and a manual system where field biologists document visual observations represent an accepted method to estimate individual animal size (Koski et al. 2006). High-resolution, low-altitude UAVs can be used to determine individual animal size, depending on the species and the survey environment (Watts et al. 2010). ...
Article
Full-text available
Data extraction from unmanned aerial vehicle (UAV) imagery has proved effective in animal surveys and monitoring, but to date, it has scarcely been used for detailed population analysis and individual animal feature extraction. We assessed the zoometric and feature extraction of Arabian oryx (Oryx leucoryx) using data acquired from a captive population for comparison with reintroduced populations monitored by UAVs. Highly accurate scaled and geo-rectified imagery derived from UAV surveys allowed precise morphometric measurements of the oryx. The scaled top view imagery, combined with baseline data from known sex, age, weight, and pregnancy status of captive individuals, were used to develop predictive models. A bracketed index developed from the predictive models showed high accuracy for classifying the age group ≤ 16 months, animals with a weight >80 kg, and pregnancy. The pregnancy classification decision tree model performed with a 91.7% accuracy. The polynomial weight predictive model performed well with relatively high accuracy when using the total top view surface measurement. Photogrammetrically processed UAV acquired imagery can yield valuable zoometric data feature extraction and modelling; it is a tool with a practical application for field biologists that can assist in the decision-making process for species conservation management.
Article
Full-text available
Effective management of animal populations requires knowledge of life history parameters and estimates of population abundance. One method commonly used to estimate abundance is capture–recapture analyses of photographs. Small, relatively inexpensive, rotary-wing drones have become an effective platform for obtaining high-quality aerial photographs of whales. To conduct capture–recapture analyses the animal needs to be defined as marked or unmarked and the photographs must be of high quality. While a system for scoring quality and markedness has previously been developed for bowhead whales (Balaena mysticetus Linnaeus, 1758) (Rugh et al. 1998. Rep. int. Whal. Commn. 48: 501–512), a revised scoring system was needed to incorporate increased information in photographs taken by drones. We present a revised scoring system that enlarges two of the previously defined areas of the whale examined for markings and incorporates smaller markings into the definition of marked whales. We scored 30 whales using the previous criteria and the revised criteria developed in this paper. More whales were identified as marked (23%) and mark scores were higher for 30% of the zones scored using the new system. Increasing the number of marked whales during capture–recapture studies increases the precision of estimated parameters and permits us to make those estimates with smaller samples of photographs.
Chapter
Over the last 30 years, the use of telemetry, especially satellite telemetry, has improved our understanding of bowhead whale movements and behavior of the Bering-Chukchi-Beaufort (BCB) stock. Here, we summarize the findings of a collaboration between the Alaska Department of Fish and Game, the Alaska Eskimo Whaling Commission, the North Slope Borough, and Fisheries and Oceans Canada that deployed satellite tags (n=77) on bowhead whales during 2006–18. This study refined our knowledge of the distribution of BCB bowhead whales, especially outside the boundaries of aerial surveys and away from whaling communities. We found that whale movements are much more complex and wide-ranging than originally thought; for example, some whales that initially migrate to the Canadian Beaufort may return west in midsummer, sometimes traveling all the way to Russia and back to Canada prior to the autumn migration. Based upon whale behavior (i.e., the timing of movements and dive behavior) and the colocation of whales and oceanographic features that help aggregate prey, bowhead whales of the BCB stock likely feed year-round. We think the distribution of BCB bowhead whales is largely driven by food availability and that whales have spatial memory of where food was located in past years. Much of the data we collected on the distribution of BCB bowhead whales is already out of date, as environmental changes are altering the distribution of bowhead prey and sea ice.
Chapter
Bering–Chukchi–Beaufort Sea bowhead whales are born at about 4.2 m in length and 1000 kg, in lead systems along the NW Alaskan coast. Bowheads in the other northern stocks are also born in ice-covered seas. Maximum body lengths (standard measure) can reach 19 m. Regarding coloration: the skin is gray on neonates and young calves, and black on yearlings and adults. The size and shape of the chin patch is variable and does not change through life. After adulthood age, the specific areas of the skin lose pigment and turn white around the eye, the peduncle, genital groove, and base of the pectoral limb. Growth is rapid from birth to age 1, followed by a 3- to 4-year growth hiatus. Following the hiatus, the increase in body length is slow to sexual maturity at ~25 years and ~13.5 m in length. The Iñupiat recognize several growth forms associated with age; these include aġvaaq (neonate), ingutuq (yearling), qairaliq (postyearling), along with several names for large adults. Allometry of the head is marked and likely associated with extreme development of the baleen rack. The head is very large in relation to body—perhaps the largest in Mammalia. The blubber is among the thickest of any cetacean (to ~38 cm). Body mass is similar to right whales of similar length; large bowheads can reach 100 metric tons or larger. Longevity can reach 200 or more years. The age of physical maturity is unclear, and adults may increase in length through life. Their ability to navigate ice-covered seas and break through ice to breathe is singular among baleen whales.
Article
Full-text available
Outlines the problems of oil tanker traffic as a result of offshore oil and gas exploration which began in 1976, the bowhead whale Balaena mysticetus being threatened as a result of this activity. Several projects have been undertaken regarding bowhead reproductive biology, distribution of the calves, age segregation and the effects of age on bowhead behaviour. The study established a complete inventory of all re-identifiable calves in the Western Arctic summering grounds and made observations of micro-scale movements and established fidelity within a season and between sites. -S.J.Yates
Article
Full-text available
The method of Cooke (1996) and Punt and Butterworth (1999) for computing abundance estimates for bowhead whales of the BeringChukchi-Beaufort Seas stock is reviewed. These abundance estimates are computed from estimates N4 of the number of whales that passed within the 4km visual range of the observation ‘perch’ from which the whales are counted, the estimated proportions P4 of the whales that passed within this range and the estimated standard errors (SE) of N4 and P4. Errors discovered while assembling the data used in developing previous estimates were corrected, and new estimated detection probabilities, N4 and P4 values and SEs were computed using the corrected data. The method of Cooke (1996) and Punt and Butterworth (1999) was then applied. The resulting 2001 abundance estimate was 10,545 (95% confidence interval 8,200 to 13,500), extremely close to the 2001 N4/P4 abundance estimate of 10,470 (95% confidence interval 8,100 to 13,500) (George et al., 2004). The estimated rate of increase of this population from 1978 to 2001 was 3.4% per year (95% confidence interval 1.7% to 5%).
Article
Aerial photogrammetric techniques were used in 1985, 1986, and 1989 through 1992 to establish the length structure of the Bering-Chukchi-Beaufort Seas stock of the Balaena mysticetus population, producing a sample size of 1898 measured individuals. The perents of adults, juveniles and calves using revised length frequency data are 41.1%, 53.7%, and 5.2%, respectively. -from Authors
Article
The 1988 census began on 17 April and continued until 10 June. A total of 2087 sightings of "new' (not duplicate) whales was made, with an additional 452 "conditional' whales. The total of new whales sightings is higher than in any previous season. By 15 May, 95% of the whales counted on the visual census had passed Pt. Barrow suggesting a relatively early migration. In 1988, Eskimo whale hunters made 29 strikes, resulting in 23 landed whales. -from Authors
Article
The number of bowhead whales passing within viewing range of the ice-based census at Point Barrow, Alaska, during spring migrations from 1978-1993 is estimated from visual census data. The trend in these annual numbers yields an estimated annual rate of increase of 4.9% (95% CI 1.4-8.5%). It is estimated that 7249 whales (95% CI 5294-9204 based on the most conservative of two alternative variance estimates) passed Point Barrow with viewing range in 1993. Since this estimate is considerably higher than past estimates the authors examined factors unique to the 1993 season. The dominant factor was the distribution of the whales offshore from the observation site. In previous years, up to half of the whales were outside viewing range. Acoustic data indicate that 90.7% of the population passed within viewing range in 1993. A 1993 population estimate that accounts for the whales that passed beyond viewing range is 8000 (95% CI 6900-9200). An estimate of 1978-1993 rate of increase that takes in to account whales that passed beyond viewing range was previously derived: 3.1% per year (95% CI 1.4-4.7%). When this estimate is combined with data on pregnancy rates and other life history parameters, population size, population age structure and historic patches via a population dynamics model, the post-model distribution of 1978-1993 rate of increase has a median of 2.3% and a 95% probability interval from 0.9-3.4%. -from Authors
Article
Acoustic location analysis was completed on 231 hrs of tapes resulting in 4636 locations. Locations were converted into tracks and from these, numbers of vocalizing whales and the distribution of their closest point of approach relative to the visual observation perch were estimated. Data were corrected to take into account effects of array geometry, distance to a calling whale and whales that swim in pairs. There were 4726 vocalizing bowheads detected in 1986 between 17 April to 2 June during the 841 hr of acoustic observation. This estimate is further corrected to 6132 whales when time periods during the season with no acoustic monitoring are taken into account. -from Authors