Jared R. Towers

Life history parameterization is foundational to population monitoring, particularly for long‐lived species under targeted management schema. In the eastern North Pacific, multidecadal photographic identification studies have facilitated the estimation of life history parameters for several distinct populations of fish‐eating “Resident” killer whales ( Orcinus orca ater ). However, to date, comparable estimates have not been generated for the increasingly abundant West Coast Transient (WCT) population of mammal‐eating “Bigg's” killer whales ( Orcinus orca rectipinnus ), despite their designation as “Threatened” under the Canadian Species at Risk Act. We used over 50 years of photographic identification data to estimate life history and reproductive parameters for WCT for the first time and compared our estimates to those previously published for sympatric Resident killer whale populations. Several aspects of WCT life history were consistent with Resident killer whales: mean age of first successful recruitment in WCT was 15.2 years (SE = 0.4), while average WCT calving interval was 4.88 years (SE = 0.13), yielding average annual fecundity of 0.20 calves per mature WCT female per year. However, WCT showed potential evidence of decreasing age at first recruitment (regression slope = −0.07, p = 0.08), were the only population examined to exhibit decreasing calving intervals over the course of their study period (1972–2024), and had the lowest calf mortality of any killer whale population assessed in this study. While shared evolutionary lineage conserves many aspects of life history among killer whales, these potential differences may be associated with ongoing WCT population growth over the last 5 decades and could reflect ecological conditions, including increasing prey abundance, within their core range.

West Coast Transient (WCT) killer whales ( Orcinus orca rectipinnus ) hunt marine mammals along the west coast of North America from California to southeastern Alaska. However, little information is available on their seasonal use of nearshore areas and the relative importance of different parts of their range. Using sighting data from 2016 to 2023, we assessed their seasonal distribution and abundance in the inside waters of Vancouver Island (the Salish Sea and North Island Waters). WCT killer whales were seen an average of 250 days throughout each year. They were most frequently observed in the Salish Sea from March to September but had fewer sightings in July of even years when pink salmon returns were low. In contrast, occurrences of WCT killer whales in the North Island Waters peaked in September but were significantly lower overall. Peak occurrences in both areas corresponded with the seasonal presence of seals and sea lions, their primary prey. Overall, ~70% of the coastal subset of the WCT population occurred annually in a region that accounts for a relatively small portion of their range. This high concentration highlights the importance of the inside waters for their survival and emphasizes the need to assess other key areas within their range.

The distinctive Subantarctic (Type D) killer whale (DKW) ( Orcinus orca ) is a highly divergent and apparently inbred form previously known only from three mass strandings and several dozen at‐sea sightings, primarily from Patagonian toothfish ( Dissostichus eleginoides ) longline vessels in the southern Indian Ocean and from tourism vessels in the Scotia Sea. Its range appears to be circumpolar throughout the Subantarctic, but very little is known about total numbers and overall movements. We summarize photo‐identification data from 352 of 406 sightings collected between 44° S and 62° S from 2003 to 2024. Among 207 photo‐identified individuals there were two distinct clusters of 78 and 29 toothfish depredators, one in the southern Indian Ocean and the other off southern Chile, respectively. An additional 100 DKW were photo‐identified throughout the wider Subantarctic. Discovery curves indicate that DKW may be more abundant than some less divergent or inbred killer whale populations. Depredating DKW showed seasonal and interannual site fidelity, but 10 depredating individuals identified in the southern Indian Ocean were also resighted in the southern South Atlantic, 4428 km away. Probable re‐sightings of two other DKW occurred in the Scotia Sea over distances of 1070 and 41 km, respectively. These latitudinally narrow, long‐distance inter‐ocean movements are unique among killer whales. During the study period, the entirely pelagic distribution of encounters appears to have shifted significantly southward.

Detection and analysis of body scars using photo‐identification data of cetaceans can provide insights into life history, behavior, and exposure to threats, but such analyses typically require manual examination of large image datasets. To explore whether machine learning can assist this process, we developed a pipeline for automatically detecting and counting conspecific scarring events on killer whales. The approach combines an nnU‐Net segmentation model, which generates binary masks containing individual scar lines, with hand‐crafted feature representations and HDBSCAN clustering to group scars into events and estimate the number of events per image. Across a dataset in which images contained between 1 and 10 annotated scarring events (median = 2), the best‐performing configuration produced a conservative estimator with a mean absolute error (MAE) of 0.82 events per image and exactly recovered the annotated number of events in just over 50% of cases. These results demonstrate that machine learning‐assisted analysis can extract biologically meaningful information on scarring burden from standard photo‐identification images of killer whales. While accurate recovery of absolute scarring burden from single images is achievable with useful reliability, detecting small inter‐annual changes remains more challenging under heterogeneous field conditions. Rather than replacing expert annotation, the present system is best viewed as a semiautomated framework that can extend existing photo‐identification workflows toward longitudinal scarring analysis and broader population‐level assessment. It therefore provides a practical foundation on which more comprehensive detectors of cetacean scarring events can be built.

Photo‐identification underpins individual‐based inference in numerous ecological studies, but scaling it to decades‐long archives remains limited by expert time. Deep learning can accelerate matching, yet most pipelines treat photographs as independent observations and therefore ignore a key aspect of the data collection method: individuals are recorded in structured encounters and often exhibit persistent, non‐random associations. We present a model agnostic, encounter‐level identification procedure that incorporates social context as a deployable probabilistic component. Given per‐image classifier posteriors, we perform log‐linear fusion of three information sources: (i) image‐based probabilities, (ii) global sighting priors (class frequency), and (iii) an encounter‐conditioned context term derived from historical co‐occurrence (log lift). The method operates as lightweight post‐processing and requires no retraining or architectural changes to the image model. Using a longitudinal photo‐identification dataset as a case study (West Coast Transient Bigg's killer whales), we evaluate (a) expert‐assisted settings in which a small number of individuals present in an encounter are known without image‐level labels, and (b) fully automated settings that initialize context from the model's own high‐confidence predictions. On a strict temporal holdout (newest 10%), encounter‐context fusion reduces top‐1 error by ~14%–25% with expert‐assisted seeding; a fully automated variant yields up to ~24% fewer misidentifications once sufficient training history exists, improving Macro‐F1 by +0.088 to +0.104, with minimal computational overhead. Placebo and seed‐corruption controls confirm that gains depend on meaningful co‐occurrence structure and collapse when encounter context is destroyed. By turning encounter structure into a reusable probabilistic component, this work bridges established methods for analyzing animal societies with practical, scalable photo‐identification pipelines. The approach is applicable to any system where individuals are repeatedly observed in groups (e.g., cetaceans, primates, ungulates, camera‐trap bursts) and provides a transparent mechanism to incorporate social context into automated identification.

Altruistic behaviors such as prey sharing are prosocial acts that can instigate and perpetuate various forms of reciprocity. Subsequent relationship dynamics provide a foundation for the evolution of societal norms and associated encephalization in social taxa, like primates and dolphins. Some cultures within these families benefit from interactions with other mammal species but accounts of any wild animals attempting to provision humans are extremely rare. In this article, we present 34 cases of both sexes and all age classes of killer whales (Orcinus orca) offering prey and other items to people who were on boats (n = 21), in the water (n = 11), and on shore (n = 2) in four oceans. A total of 18 species were offered-six fishes, five mammals, three invertebrates, two birds, one reptile, and one seaweed. In almost every case the whales awaited a human response before subsequently reacting. The occurrence of these events suggests a limited cost to exploratory behaviors in some populations of this species. We suggest these apparently nonrandom cases may be representative of interspecific generalized altruism. This behavior may represent some of the first accounts of any wild predator intentionally using prey and other items to directly explore human behavior and thus may highlight the evolutionary convergence of intellect between highest order primates and dolphins. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

Scars obtained from interactions with conspecifics may be caused by both playful and aggressive activities, making them useful when studying cetacean behaviour. This study investigates the effects of age and sex on conspecific scar acquisition and healing in three genetically distinct populations of killer whales (Orcinus orca) each with unique diets and social structures. The sample consisted of 50 of the most commonly photo-identified individuals from all sex and age classes in each of the Bigg’s, Northern Resident, and Crozet killer whale populations. The number of new scars annually acquired by an individual as well as how long it took them to disappear were extracted from annual photo-identification images of these individuals taken between the years of 2008 and 2021. Scar acquisition was analysed using a generalized additive model while scar healing was assessed using Kaplan-Meier survival curves. Results showed an inverse relationship between scar acquisition and age, as well as an effect of sex with males being more scarred than females amongst all age classes. No significant differences in scar acquisition between populations was found. Scar re-pigmentation was faster in Northern Residents compared to Crozet and Bigg’s individuals and varied amongst age classes, with scars on calves and juveniles disappearing more quickly than those on adults. These population- and age-based differences in healing may be due to scar severity, while results around scar acquisition suggest that the nature of physical interactions between sex and age classes in this species are homogenous despite cultural and genetic differences that have evolved between populations.

Humpback whales ( Megaptera novaeangliae) use southern British Columbia waters to feed, but the type and quantity of prey in many areas used for feeding is unknown. We conducted active acoustic prey mapping in 55 small grid-surveys in two regions off Vancouver Island. We quantitatively compared fish and zooplankton-dominated biomass in known feeding areas with and without foraging humpback whales, and qualitatively described the prey characteristics of the foraged areas. Surveys of the water column suggest that, on average, humpback whale foraging was associated more with increased zooplankton than fish biomass. Prey characteristics varied between the two regions (∼500 km apart), but there was no significant difference in mean backscatter strength in the actively foraged areas between the two regions. Frequency differencing discriminated between the dominant taxa in the water column, but potential epipelagic prey (<10 m) would have been omitted from analysis. However, average depth at the maximum acoustic prey detections was significantly deeper when whales were present (84 m) versus absent (60 m), suggesting predominantly subsurface foraging opportunities suitable to prey mapping.

Endangered Southern Resident killer whales (Orcinus orca) are fish‐eaters that preferentially prey on adult Chinook salmon (Oncorhynchus tshawytscha). Despite being salmon specialists, individuals from all three killer whale pods (J, K, L) have been observed harassing and killing porpoises (family Phocoenidae) without consuming them. Retrospectively, we identified and analyzed 78 episodes of Southern Resident killer whales harassing porpoises between 1962 and 2020, of which 28 resulted in the porpoise's death (“phocoenacide”). Fifty‐six episodes involved harbor porpoise (Phocoena phocoena), 13 involved Dall's porpoise (Phocoenoides dalli), and the porpoise species was unreported for nine episodes. Southern Resident killer whales often targeted young porpoises that were similar in size to adult Chinook salmon. Both sexes participated in porpoise harassment. Juveniles engaged in the behavior the most; however, their rates of engagement were not found to differ significantly from most other age classes. The behavior was passed through generations and social groupings, as it was first observed in L pod and spread to the other two pods. Killer whales are highly complex animals known to exhibit social learning and cultural transmission of learned behaviors, but the reason(s) for this behavior is unknown. Hypotheses include the social and developmental benefits of play, hunting practice, or displaced epimeletic behavior.

Researchers can investigate many aspects of animal ecology through noninvasive photo–identification. Photo–identification is becoming more efficient as matching individuals between photos is increasingly automated. However, the convolutional neural network models that have facilitated this change need many training images to generalize well. As a result, they have often been developed for individual species that meet this threshold. These single‐species methods might underperform, as they ignore potential similarities in identifying characteristics and the photo–identification process among species. In this paper, we introduce a multi‐species photo–identification model based on a state‐of‐the‐art method in human facial recognition, the ArcFace classification head. Our model uses two such heads to jointly classify species and identities, allowing species to share information and parameters within the network. As a demonstration, we trained this model with 50,796 images from 39 catalogues of 24 cetacean species, evaluating its predictive performance on 21,192 test images from the same catalogues. We further evaluated its predictive performance with two external catalogues entirely composed of identities that the model did not see during training. The model achieved a mean average precision (MAP) of 0.869 on the test set. Of these, 10 catalogues representing seven species achieved a MAP score over 0.95. For some species, there was notable variation in performance among catalogues, largely explained by variation in photo quality. Finally, the model appeared to generalize well, with the two external catalogues scoring similarly to their species' counterparts in the larger test set. From our cetacean application, we provide a list of recommendations for potential users of this model, focusing on those with cetacean photo–identification catalogues. For example, users with high quality images of animals identified by dorsal nicks and notches should expect near optimal performance. Users can expect decreasing performance for catalogues with higher proportions of indistinct individuals or poor quality photos. Finally, we note that this model is currently freely available as code in a GitHub repository and as a graphical user interface, with additional functionality for collaborative data management, via Happywhale.com.