Whale Tail Identification: How to Recognize Whales by Their Flukes

Whale tail flukes are the most reliable body part for identifying individual whales and, in many cases, determining the species. The underside of a humpback whale tail fluke carries a pigment pattern and trailing edge shape that is as unique to that animal as a fingerprint is to a human. For researchers, students, and life-science professionals, learning to read these patterns opens the door to photo-identification studies, population monitoring, and behavioral research. This article explains the anatomical basis of fluke identification, the key features to examine, the practical workflow for matching photographs, and the limitations of the method.

Why Tail Flukes Are the Primary Identification Feature

Whales surface to breathe through their blowhole, and the moment of exhalation is often the only opportunity for manual recognition in the field. The dorsal fin and lateral body can help identify the species, but these features are less distinctive for individual animals. The tail fluke, by contrast, offers a large surface area with stable markings that persist across years.

The fluke is the two-lobed structure at the end of the tail, technically called the caudal fin. It is composed of dense connective tissue, not bone, and it moves vertically to propel the whale through the water. The fluke stroke frequency changes with behavior, as documented in bowhead whale studies where fluking rate was significantly faster during feeding compared with non-feeding periods. This means the tail is in near-constant motion, giving observers repeated chances to capture a usable photograph.

The scientific value of fluke identification is substantial. A collaborative North Pacific humpback whale dataset compiled 27,956 unique individuals from 157,350 encounters over a 20-year period, built from 39 separate photo-identification catalogs. This scale of work is possible only because individual whales can be recognized reliably from their flukes across seasons and across ocean basins.

Anatomy of the Whale Tail and Fluke

The whale tail consists of several parts that matter for identification. The peduncle is the tapered muscular section connecting the body to the flukes. The flukes themselves are the two broad lobes, called the left and right flukes, separated by a central notch. The trailing edge is the rear margin of each fluke, and the leading edge is the front margin where the fluke meets the peduncle.

The fluke is supported internally by connective tissue and, in some species, by small bones that are remnants of the pelvis. The skeletal structure of the tail is part of a broader evolutionary pattern. Tunas, lamnid sharks, modern whales, and derived ichthyosaurs converged on the thunniform body plan, which includes a lunate caudal fin, compressed peduncle, and peduncle joint. These features contribute to posterior concentration of tail-stem oscillation, tail stem stabilization, peduncle joint flexibility, and caudal fin angle fixation. Understanding this anatomy helps explain why fluke shape varies among species and why some species are easier to identify by tail than others.

The fluke surface is covered in skin that can carry scars, bite marks, barnacle attachments, and pigment patterns. These marks accumulate over the life of the animal and provide the natural tags used in photo-identification. The skin microbiome of the tail region differs from other body sites, with skin wounds showing environmental similarities and high levels of Vibrio species in one stranded melon-headed whale study. This context matters for researchers handling dead animals, because the tail surface can carry bacteria that are relevant to decomposition and sampling protocols.

Key Fluke Features for Species Identification

Different whale species have different fluke shapes, sizes, and marking patterns. Learning these differences is the first step in species-level identification from tail photographs.

Humpback Whale Flukes

Humpback whales have the most distinctive flukes of any large whale. The trailing edge is deeply scalloped or serrated, and the underside carries a black and white pigment pattern that is unique to each individual. The fluke is broad and can span up to one-third of the body length. The white pigmentation on the ventral surface varies from completely white to completely black, with every intermediate pattern possible.

The humpback fluke is the basis of the largest photo-identification efforts in the world. The North Pacific dataset mentioned earlier relies entirely on fluke photographs, and image recognition algorithms using machine learning can detect matches between individuals with an estimated 97 to 99 percent accuracy. This high accuracy is possible because humpback fluke patterns are so variable and so stable over time.

Sperm Whale Flukes

Sperm whales have flukes that are triangular in shape with a smooth trailing edge and a deep central notch. The flukes are relatively small compared with the massive body, and they are held high in the air during a dive, a behavior called fluking up. The underside is typically uniform in color, usually gray or brown, without the bold pigment patterns seen in humpbacks.

Sperm whale identification relies more on the trailing edge notches and scars than on pigment. The fluke edges accumulate nicks and tears over time, and these are used as natural marks. Aerial photogrammetry using uncrewed aerial vehicles has been used to infer sex and developmental stage of sperm whales based on body length, but individual identification still depends on fluke and dorsal fin marks.

Right Whales and Bowhead Whales

Right whales and bowhead whales have wide, paddle-shaped flukes with smooth trailing edges. They lack a dorsal fin, and their flukes are often covered with callosities, which are rough patches of thickened skin that host whale lice. These callosities are more prominent on the head than on the flukes, but fluke markings still contribute to identification.

Bowhead whale feeding behavior has been studied using biologging tags that record fluke stroke frequency. The fluke stroke rate was significantly faster during feeding at 0.191 Hz compared with 0.172 Hz during non-feeding, and the rate was depth-dependent, higher at shallow depths and lower at deeper depths. This research shows that fluke movement patterns carry behavioral information, beyond identification information.

Beaked Whales

Beaked whales have small flukes relative to body size, and they are rarely seen at the surface. The goose-beaked whale, also called Cuvier's beaked whale, has been studied extensively because of its stranding events associated with naval sonar. Beaked whale flukes are not commonly used for photo-identification because the animals are difficult to observe, but the tail plays a critical role in their diving behavior.

Beaked whales use a unique strong gait called B-strokes during ascents from deep dives. During sonar exposure, B-strokes were used during descent and ascent phases of both deep and shallow dives, and the onset occurred during all sonar exposure periods with levels above 100 dB re 1 microPa lasting more than three minutes. This change in swimming gait is an oxygen-saving strategy that starts earlier in exposed dives at 16 minutes compared with 33 minutes in unexposed dives. Understanding these tail movements is essential for interpreting behavioral responses to acoustic disturbance.

At a Glance: Fluke Features by Species

Photo-Identification Workflow

Photo-identification is the standard method for using flukes to recognize individual whales. The workflow involves capturing images, organizing them into catalogs, and matching new photographs against existing records. This process has been refined over decades and now includes automated matching tools.

Step 1: Capture Quality Images

The quality of the photograph determines whether identification is possible. The ideal fluke photograph is taken from directly behind the whale as it dives, with the ventral surface of the fluke facing the camera. The image must be in sharp focus, well lit, and large enough in the frame to show the trailing edge and pigment pattern in detail.

For humpback whales, the ventral surface is the standard view because the pigment pattern is most visible there. For sperm whales, both dorsal and ventral surfaces can be useful because the edge marks are visible from either side. The photographer should take multiple shots in rapid succession because the fluke is in motion and only a fraction of images will be usable.

Step 2: Organize Images by Species and Region

Images should be sorted by species first, then by geographic region and date. This sorting reduces the number of comparisons needed during matching. Regional catalogs are common because whales often show site fidelity to feeding or breeding areas. A comparison of humpback whale tail fluke catalogs from the Sultanate of Oman with Madagascar and the East African mainland demonstrates that regional catalogs can be compared to study migration patterns and population connectivity.

Step 3: Extract Identification Features

The identification features to extract from each image include the overall fluke shape, the trailing edge contour, the pigment pattern on the ventral surface, and any distinctive scars or marks. For humpback whales, the black and white pattern is the primary feature. For other species, the trailing edge nicks and tears are more important.

The features should be recorded in a standardized format. Many catalogs use a coding system that describes the amount of white on the fluke, the shape of the trailing edge, and the location of distinctive marks. This coding allows rapid sorting of images into groups that share similar features.

Step 4: Match Against the Catalog

Matching can be done manually by comparing images side by side, or with automated tools. Manual matching requires experience and careful attention to detail. Automated matching uses image recognition algorithms that compare the new image against all images in the catalog and return the most likely matches.

The North Pacific humpback whale dataset used a machine learning algorithm that rapidly detects matches between individuals with an estimated 97 to 99 percent accuracy. A separate study identified over one thousand individual wild humpback whales using fluke photos, and another used convolutional neural networks to classify images from the Gulf of Maine humpback whale catalog. These tools do not replace human verification, but they dramatically reduce the time needed to find candidate matches.

Step 5: Verify Matches Manually

Every automated match must be verified by a human observer. The observer compares the candidate images side by side, checking the trailing edge contour, the pigment pattern, and the position of scars. A match is confirmed only when multiple features align. Misidentification is a known risk in photo-identification, and using all available natural marks reduces this risk.

Records and Measurements for Fluke Identification

Maintaining accurate records is essential for any photo-identification program. The following data should be recorded for each image:

The encounter history of each individual should be maintained in a separate record. The North Pacific dataset recorded each individual in an average of 5.6 sampling periods, with an annual average of 87 percent of whales encountered in more than one season. This encounter history is the foundation for studies of movement, survival, and population dynamics.

Common Failure Patterns in Fluke Identification

Several recurring problems reduce the accuracy of fluke identification. Recognizing these failure patterns helps researchers avoid them.

Poor Image Quality

Images that are out of focus, underexposed, or taken at an angle that obscures the trailing edge cannot be matched reliably. The solution is to take many images and to discard any that do not meet minimum quality standards. A common rule is to keep only images where the fluke fills at least a certain portion of the frame and the trailing edge is clearly visible.

Confusing Species

Beginners often confuse the flukes of different species, especially when the photograph is taken at a distance or at an angle. Humpback flukes are the most distinctive, but sperm whale flukes can be mistaken for those of other toothed whales. The solution is to learn the typical fluke shape of each species before attempting individual identification.

Overlooking Small Marks

Small nicks and scars on the trailing edge are often the only features that distinguish one individual from another in species with uniform pigmentation. These marks can be missed in low-resolution images or when the observer focuses only on the pigment pattern. The solution is to examine the trailing edge carefully at high magnification.

Relying on a Single Feature

Using only the pigment pattern or only the trailing edge shape increases the risk of misidentification. The most reliable matches use all available natural marks, including scars, barnacle attachments, and edge nicks. A study on photo-identification of individual humpback whales using all available natural marks examined the implications for misidentification and automated algorithm matching technology.

Catalog Drift

Over time, a catalog can accumulate errors from misidentified matches, duplicate entries, or outdated images. Regular audits of the catalog are necessary to maintain data quality. The solution is to have independent observers recheck a sample of matches and to document any corrections.

Limitations of Fluke Identification

Fluke identification is a powerful tool, but it has clear limitations that researchers must acknowledge.

Species Coverage

Fluke identification works best for species that regularly raise their flukes when diving and that have distinctive markings. Humpback whales are ideal subjects. Species that rarely fluke up, such as some beaked whales, are poor candidates for fluke-based identification. For these species, other methods such as genetic sampling or acoustic monitoring may be more appropriate.

Geographic Coverage

Photo-identification requires observers to be in the right place at the right time. Many whale habitats are remote, and survey effort is uneven. Species distribution models for baleen whales in the Irish Exclusive Economic Zone used long-term citizen science data to fill information gaps, showing that community contributions can improve coverage. However, large areas of the ocean remain unsurveyed.

Temporal Stability of Marks

Fluke markings are generally stable over time, but they can change. Scars heal, new nicks appear, and pigment patterns can shift slightly. The longer the interval between photographs, the more difficult the match. Catalogs that span decades must account for these changes.

Individual Variation

Some individuals have flukes that are nearly identical to others, especially in species with limited pigment variation. In these cases, even high-quality photographs may not be sufficient for a confident match. The solution is to use additional identification features, such as dorsal fin shape or body scars, when available.

Welfare and Safety Context for Field Observation

Observing whales for fluke identification requires attention to both human safety and whale welfare. Vessel approach distances are regulated in many jurisdictions, and researchers must be familiar with local laws. The goal is to obtain usable photographs without disturbing the animals.

Whale behavior changes in response to disturbance. Beaked whales exposed to naval sonar showed altered swimming gaits, with B-strokes occurring during descent and ascent phases of both deep and shallow dives. The onset of B-strokes occurred during all sonar exposure periods with levels above 100 dB re 1 microPa lasting more than three minutes. This response is an oxygen-saving strategy that, if prolonged, could lead to physiological changes contributing to gas bubble formation and growth that could lead to stranding.

For researchers using boats, the approach should be slow and steady, avoiding sudden speed changes or course alterations. The whale should be allowed to control the interaction. If the whale changes its diving behavior, the vessel should increase its distance. The welfare of the animal takes priority over obtaining a photograph.

For researchers handling dead stranded animals, the tail surface can carry bacteria relevant to health and safety. A study of a stranded melon-headed whale found Vibrio species across all body sites, with skin wounds showing high levels of Vibrio at 47.84 percent, Pseudoalteromonas at 17.84 percent, and Psychrobacter at 12.36 percent. Seven Vibrio species were identified across all niches, including V. alginolyticus, V. parahaemolyticus, and V. cholerae in non-respiratory samples. Protective gloves and appropriate hygiene protocols are essential when examining flukes of dead animals.

Professional Escalation Criteria

Knowing when to escalate an identification question to a specialist is part of professional practice. The following situations warrant consultation with an experienced photo-identification researcher or a regional catalog curator:

• The photograph is of a species that is rare or unusual for the region, and the observer is not confident in the species identification.
• The automated matching algorithm returns a match with low confidence, and the observer cannot verify the match manually.
• The fluke shows marks that could indicate injury, entanglement, or disease, and the observer needs guidance on reporting.
• The photograph is intended for inclusion in a formal research catalog, and the observer is unsure whether the image meets quality standards.
• The identification is for a species with known conservation concerns, and the observation may inform management decisions.

Regional catalogs and research networks can provide expert review. The collaborative North Pacific humpback whale dataset is an example of a living and accessible resource for collaborative, basin-wide studies. Researchers should contact the relevant catalog curator before submitting images for inclusion.

Frequently Asked Questions

Why are humpback whale flukes used for identification more than other species?

Humpback whales have the most distinctive flukes of any large whale. The ventral surface carries a black and white pigment pattern that is unique to each individual, and the trailing edge is deeply scalloped. These features are stable over time and are visible when the whale raises its flukes during a dive. The North Pacific humpback whale dataset documented 27,956 unique individuals from 157,350 encounters, demonstrating the practical value of fluke identification for this species.

Can whale flukes be used to identify the species of a whale?

Yes, fluke shape and markings can help identify the species. Humpback flukes are broad with a scalloped trailing edge and bold pigment patterns. Sperm whale flukes are triangular with a smooth trailing edge and uniform color. Right whale and bowhead whale flukes are wide and paddle-shaped. Beaked whale flukes are small relative to body size. However, species identification is usually easier using other features such as body size, dorsal fin presence and shape, and blow shape.

How accurate is automated fluke matching with machine learning?

The image recognition algorithm used in the North Pacific humpback whale dataset was capable of rapidly detecting matches between individuals with an estimated 97 to 99 percent accuracy. A separate study using convolutional neural networks on the Gulf of Maine humpback whale catalog reported successful image classification. Automated matching does not replace human verification, but it reduces the time needed to find candidate matches.

What is the best angle for photographing a whale fluke?

The best angle is directly behind the whale as it dives, with the ventral surface of the fluke facing the camera. This angle shows the pigment pattern and the trailing edge clearly. The photograph should be in sharp focus, well lit, and large enough in the frame to show detail. Multiple shots should be taken because the fluke is in motion.

How long do fluke markings remain stable on an individual whale?

Fluke markings are generally stable over many years, which is why photo-identification is possible. Scars heal, new nicks appear, and pigment patterns can shift slightly, but the overall pattern remains recognizable. The longer the interval between photographs, the more difficult the match, and observers should account for changes when comparing images from different years.

What should I do if I photograph a whale fluke with signs of injury or entanglement?

Document the marks carefully with multiple photographs from different angles. Record the date, time, and location. Report the observation to the relevant regional stranding network or research organization. Entanglement scars and fresh injuries may indicate an animal in distress that requires monitoring or intervention.

Are there safety concerns when photographing whale flukes from a boat?

Yes, vessel approach distances are regulated in many jurisdictions, and researchers must follow local laws. The approach should be slow and steady, and the whale should be allowed to control the interaction. If the whale changes its diving behavior, the vessel should increase its distance. Whale welfare takes priority over obtaining a photograph.

Can fluke identification be used for species that rarely raise their tails?

Fluke identification is less useful for species that rarely raise their tails, such as some beaked whales. For these species, other methods such as genetic sampling from biopsy darts, acoustic monitoring, or dorsal fin identification may be more appropriate. The choice of method depends on the species and the research question.

Related Articles

• plant disease identification
• qPCR Machine: Features That Affect Quantification and Reproducibility
• qPCR Machine: Features That Affect Quantification and Reproducibility
• qPCR Machine: Features That Affect Quantification and Reproducibility
• Batch Effects in RNA-seq: How to Recognize and Plan Around Them

References and Further Reading

• NCBI Literature Resources. National Center for Biotechnology Information.
• PubMed. National Library of Medicine.
• A collaborative and near-comprehensive North Pacific humpback whale photo-ID dataset.. Scientific reports, 2023.
• Vibrio spp.: a potential critical pathogen for mammals with implications beyond marine aquaculture.. BMC microbiology, 2025.
• A multi modal fusion coal gangue recognition method based on IBWO-CNN-LSTM.. Scientific reports, 2024.
• Genetic and morphological approaches distinguish the three sibling species of the Anisakis simplex species complex, with a species designation as Anisakis berlandi n. sp. for A. simplex sp. C (Nematoda: Anisakidae).. The Journal of parasitology, 2014.
• Skeletal convergence in thunniform sharks, ichthyosaurs, whales, and tunas, and its possible ecological links through the marine ecosystem evolution.. Scientific reports, 2023.
• Global contamination of coplanar polybrominated/chlorinated biphenyls (Co-PXBs) in the market fishes from Japan.. Chemosphere, 2008.
• Identification of two allelic forms of ovine CD4 exhibiting a Ser183/Pro183 polymorphism in the coding sequence of domain 3.. Veterinary immunology and immunopathology, 2006.
• Source levels and estimated yellowfin tuna (Thunnus albacares) detection ranges for dolphin jaw pops, breaches, and tail slaps.. The Journal of the Acoustical Society of America, 2000.
• Bowhead whale foraging dives are defined by speed and body orientation.. 2026.
• Inferring sperm whale (Physeter macrocephalus) sex and developmental stage using aerial photogrammetry.. 2026.
• Assessment of goose-beaked whale responses to mid-frequency active sonar using a hierarchical hidden Markov model.. 2026.
• Naval sonar induces an anaerobic swimming gait in beaked whales.. 2025.
• Shark species identification from bite marks on a Short-finned Pilot Whale, Globicephala macrorhynchus. Ocean and Coastal Research, 2022.
• Distribution models of baleen whale species in the Irish Exclusive Economic Zone to inform management and conservation.. Marine Environmental Research, 2024.
• Deep-Whale: Integration of Deep-Learning Networks for Identification, Classification and Moment Pattern Analysis of Whales Using Multi-Temporal Spaceborne Remote Sensing Data. IEEE International Geoscience and Remote Sensing Symposium, 2025.
• Whale Species Classifier Using Dorsal Fin And Lateral Body Characteristic. International Conference on Computational Collective Intelligence, 2023.
• Photo-identification of individual humpback whales (Megaptera novaeangliae) using all available natural marks: Implications for misidentification and automated algorithm matching technology. Journal of Cetacean Research and Management, 2020.
• Identification of over One Thousand Individual Wild Humpback Whales using Fluke Photos. Proceedings of the International Joint Conference on Computer Vision Imaging and Computer Graphics Theory and Applications, 2022.
• A note on the comparison of humpback whale tail fluke catalogues from the Sultanate of Oman with Madagascar and the East African mainland. Journal of Cetacean Research and Management, 2010.
• Image classification with convolutional neural networks using gulf of maine humpback whale catalog. Electronics Switzerland, 2020.

This article is educational and does not replace institutional policy, professional advice, or applicable safety and regulatory requirements.