The Use of Photo-Identification in Dolphin Research
The study of ecology, behavior and life history parameters of any species necessitates the need to identify the individuals within a population. Individual identification of study animals broadens our understanding of such things as population size, migratory routes, site fidelity, preferred habitat, life spans and reproductive histories. In addition, behavioral
studies are dependent on identifying individuals within a focus population in order to interpret social interactions and associations of animals, quantify rates of behavior and gain an overall understanding of social structure within groups.
Studies of many cetacean species took a great leap forward with the introduction of photo-identification techniques in the 1970s. Roger Payne was the first to document the ability to distinguish individual right whales by taking and comparing photographs of the callosity patterns found on their heads. The unique saddle pattern coloration and distinct shapes of dorsal fins proved useful to Michael Bigg for identification of specific animals in the study of Orca populations. The husband and wife research team, Bernd and Melany Wursig, gave further validity to the use of photo-identification by determining that individual bottlenose dolphins could be identified through the comparison of photographs of their dorsal fins, most of which displayed curves, notches, nicks and tears (Wells, 2000). Most cetaceans display individually specific coloration patterns or tattered and uniquely curved edges of flukes and dorsal fins as well as scars which accumulate over their lifetimes through interaction with other cetaceans, predators and the environment (figure 1). These markings often provide a convenient means by which scientists can distinguish one cetacean from another.
Figure 1. Distinct dorsal fins of individual bottlenose dolphins displaying unique permanent characteristics used in their identification. (Photographs © Dolphin Research Center).
Since the Wursigs’ groundbreaking use of photo-identification techniques with bottlenose dolphins, a number of independent studies around the world have incorporated this methodology as a tool toward gaining incredible insights into the lives of these animals. A long-term study of bottlenose dolphins in Galveston Bay, Texas from 1990-2001 employed photo-identification as a tool which helped determine that individual bottlenose dolphins in the area showed strong site fidelity with seasonal fluctuations in their use of different habitats. The study also concluded that the number of resident animals during that time ranged from 28-37, though additional animals were also found to pass through the area (Brager, et al., 1994, Maze & Wursig 1999, Irwin & Wursig 2004). In Shark Bay, Australia, photo-identification techniques helped determine that several foraging techniques were unique to only a certain number of individual bottlenose dolphins at specific sites. One of the most interesting examples included five particular animals that were the only ones documented to carry sponges and use them during feeding (Smolker, et al. 1997). Additional studies in Shark Bay, using photo-identification to distinguish individuals, have raised awareness of preferred male associations and spurred subsequent investigations into alliance strategies that contribute to their mating success (Krutzen, et al., 2003). Knowing the identity of each of the players in a situation is essential to identifying the frequencies of associations as well as the number of behavioral occurrences amongst specific individuals and within a population. This information assists in better understanding the social structure of these animals.
Bottlenose dolphins have also been studied off the coast of Sarasota, Florida since the 1970s. More than 120 resident animals have been identified. Photo-identification has been essential to many studies of the species in the area and particularly helpful when focusing on the affects of boat disturbance on dolphins. Identifying particular individuals allowed researchers to compare the behavior of an animal during situations involving commercial and private boat approaches as well as during undisturbed periods. Findings from this study indicated that boat activity caused significant changes in behavior and physiological response in the dolphins and supported the need for better management plans of boating activity in areas of importance to the animals (Nowacek, et al., 2001; Mote Marine Lab).
As shown above, photo identification provides a means of obtaining information on the habitat use of individuals and populations by allowing researchers to recognize specific animals in their environment. This non-invasive technique eliminates the need for invasive methods such as tags or other artificial man made markings like fin-notching or freeze branding to identify them. Tagging, branding and other artificial marking methods can cause injury to the animal or alter its behavior and is often labor intensive. Additionally, artificial marks tend to fade faster than natural markings and tags are often shed, rubbed or pulled off, rendering them useless. Natural markings such as individual contours of fins/flukes and nicks and notches within them, coloration patterns, scars, healed lesions and rare deformities are now becoming commonplace characteristics with which to identify individuals within many different types of cetaceans, though the ease of doing so varies with species. Unlike tagging and marking, photo identification can be unobtrusive and generally results in little disturbance to the behavior of most cetaceans.
Cetaceans often surface only briefly to breathe, limiting time of sighting them to mere seconds. This quick glimpse of a moving target may not be long enough for a person to visually process and recognize the individual animal. Photographs can capture distinguishing features of the animal for later analysis and comparison. Animals with more subtle characteristics, which may be missed in a brief sighting in the field, may be identified more readily under scrutiny of a photo in the lab. In a study published by Ken Norris, ten times the number of animals were correctly identified by photo than visually in the field. (Norris 1985, Whitehead, et al., 2000)
Many challenges present themselves when trying to obtain useful photos for identification purposes. Weather and lighting conditions contribute to the ability to sight and obtain useful photos of subjects. Long hours of searching for animals, which may never be found, can induce fatigue and hinder the ability of researchers to sight animals. If an animal is sighted, it can be difficult to maneuver into position in order to obtain an angle for a photo that will enable easy identification and comparison. Obtaining the correct orientation of the fin or fluke in the camera frame, both vertically and horizontally, can be detrimental to the usefulness of the photo for identification. Deciding on specific individuals to focus on and follow from among a large group of animals can cause difficulty.
Obtaining images of all animals encountered is challenging, if not impossible, due to the large number of animals or the quick, active or even shy nature of certain species. This can be problematic to studies of the social organization of a population of animals, since the interaction of all players is important to determine what is occurring and the roles of the individuals involved. Surveys focused on entire populations of animals require researchers to identify and keep track of all animals in the population. The need to obtain images of a large number of individuals, especially over vast areas, almost inherently means fewer sightings and images taken of each individual due to the wide focus of the study. The brief glimpses of each individual make surveys less advantageous for the study of the species’ behavior and social structure. This is in contrast to focal follows where an individual may be the intended target of the study in order to gain specific information on that particular animal’s social interactions, bonds and life history over an extended period. Photo identification used in visual surveys can sometimes lead to the identification of an animal’s gender, either by the persistent presence of a calf next to it, indicating the larger animal as a female, or through the rare ability to sight genital markings. Photos of the ventral genital markings, however, can generally present difficulties in the identification of an animal, as the dorsal fin is no longer in the image, requiring additional photographs or documented information to verify the identity of the individual. In focal studies, the focus is narrow and may result in a larger number of photos and information on specific individuals rather than all the animals encountered in an area. The restricted view of focal follows makes them undesirable for obtaining a broad understanding of the entire population. Therefore, the focus of the research may determine what and how many images are taken and catalogued during a particular study and how those are applied. (Mann, J., 2000).
While natural markings are useful to distinguish many cetaceans, steps must be taken to monitor whether they are permanent or superficial and therefore useful or not for identification. Changes may occur to distinctive scars and characteristics over time, the appearance of new markings may overlap older ones, and scratches or lesions may heal and fade making identification of an individual more challenging if these alterations are not regularly monitored. Observing changes in distinguishing characteristics can reduce error in the identification process. Animals with very obvious permanent markings, fin/fluke contours, coloration or rare deformities are more easily identifiable than other animals, such as calves, which often have few observable distinct characteristic markings. Care must be taken to ensure that clearly marked individuals do not end up over-represented compared to those animals whose markings are subtler and less readily recognized. Lack of attention to this detail may result in biases in population counts (Scott, et al., 1990).
Scientists conduct the initial analysis and comparison of photographs through negatives on a light table, contact sheets, and slides or via digital images on computer screens. Photographs are rated for quality and only those with the highest quality rating are used to assist in reducing errors in matching and classifying individuals. Trailing edges of notch patterns in dorsal fins/flukes are traced onto paper in small, several inch frames or, more often today, they are traced via computer programs designed to produce exact outlines of each dorsal fin/fluke. Dorsal edge marks are the primary means of identification along with curvature of the fin and any noticeable scars, scratches, lesions or deformities. These are compared manually or via computer software to catalogued photographs. Images that are matched are catalogued in the existing individual’s file, while those photographs unmatched and deemed new individuals have a new file established (Robinson, pers. comm.). Non-matches are considered new individuals. Pictures are usually inspected by several people, to prevent errors in identification, before being catalogued (Whitehead, et al., 2000). Difficulty can arise in matching right and left side photos of dorsal fins. Identical rake (scratch) marks may appear on both sides of a fin, due to altercations with other animals, making it easy to match some right and left shots. However, photos of animals with few distinct markings may result in an inability to match right and left dorsal fin images resulting in the identification of one animal as two different individuals, or a “false negative”. Taking images of only one side of each animal’s dorsal fin can reduce error and effort (Robinson, pers. comm.).
Photo quality is of extreme importance when sorting and identifying individual animals from pictures in the lab. Stevick, et. al., 2001, identified a significant relationship between the quality of photographs and the number of errors in identification of animals as well as the overrepresentation of certain individuals that had more distinctive features. False positives, meaning the identification of two different animals as the same one, and false negatives are often the result of discrepancies in photo quality. Stevick further suggested that maintaining high standards for the quality of photographs could reduce rate of error and eliminate bias (Stevick, et al., 2001). The experience level of staff members and the identification protocol used contributed greatly to the rate of error.
Long term studies of a cetacean population can result in the cataloguing of thousands of animals. As they grow, these databases become more laborious to maintain and sort through in order to identify new individuals. Digital photography and computer assisted dorsal fin matching programs are now coming to the forefront. These tools can reduce the time involved to sort, identify and catalogue individuals. Programs such as Finscan and Europhlukes can now trace fin/fluke contours, calculate dorsal ratios and compare all pre-existing catalogued photos. These software packages provide scientists with a smaller number of possible matches, greatly reducing the amount of time otherwise spent manually sifting through pictures (Kreho, et al., 1999). Matches produced by the software can give confidence limits for the nearest match, further assisting researchers in their decision. These programs can also help bring to light errors in previous classifications by identifying false negatives or false positives. These programs still need development to overcome some remaining challenges to the correct identification and classification of animals. If an animal’s dorsal fin or fluke profile changes dramatically, current computer software programs may not identify it, resulting in classifying the animal as a new individual. The unique differences between species require alternative models of software to be developed in order to produce effective matching programs. Some species display subtle coloration differences between individuals which currently cannot be picked up visually or by existing matching programs in the detail required for identification (Robinson, pers. comm.).
While there is much development yet to be done with existing computer matching software, the possibilities it provides for collaboration among scientists are almost endless. Prior to the creation of systems such as Europhlukes, scientific research on cetacean communities had been confined to “pockets” focused on specific regions. Collaborative databases will foster sharing of knowledge and piece together a potentially much bigger picture of how cetaceans use and fit into their environment. What is currently known about the ranging patterns of some cetaceans may quickly change as scientists in different countries begin comparing pictures of identified individuals. Understanding of the preferred habitats, site fidelity and intra- and inter-species associations of cetaceans, and, possibly, genetics may also broaden as researchers pool photos and are encouraged to share additional information together in matching databases, which can then be compared on a much larger scale and eventually worldwide. Worldwide collaboration could be especially important for long ranging species of baleen whales, sperm whales and orcas, to name a few. This type of cooperation within the scientific community could lead to better, more effective and expansive management of cetacean populations and the habitats they rely upon.
Photo identification studies have greatly contributed to the small amount of what we know of cetaceans today. Continual use of this technique will enhance and broaden our understanding of the needs of these unique animals and the role they play within the ecosystem. This understanding will allow for the establishment of management proposals and actions that can help cetaceans and other marine species against the ever-increasing threat of human disturbance.
References:
Araabi B.N., Kehtarnavaz N., McKinney T., Hillman G., Wursig B. 2000. A String Matching Computer-Assisted System for Dolphin Photo-identification. Annals of Biomedical Engineering, 28: 1269-1279.
Brager S., Wursig B., Acevedo A., Henningsen T. 1994. Association patterns of bottlenose dolphins (Tursiops truncatus) in Galveston Bay, Texas. Journal of Mammalogy 75 (2) 1994. 431-437.
Gope C., Kehtarnavaz N., Hillman G., Wursig B. 2005. An affine invariant curve matching method for photo-identification of marine mammals. Pattern Recognition, 38: 125-132.
Irwin L.J., Wursig B. 2004. A small resident community of bottlenose dolphins, Tursiops truncatus, in Texas: Monitoring recommendations Gulf of Mexico. Science [print] 22(1): 13-21.
Kreho A., Kehtarnavaz N., Araabi B., Hillman G., Wursig B., Weller D. 1999. Assisting Manual Dolphin Identification by Computer Extraction of Dorsal Ratio. Annals of Biomedical Engineering, 27: 830-838.
Krutzen M., Sherwin W.B., Connor R.C., Barre L.M., Van De Casteele T.,
Mann J., Brooks R. 2003. Contrasting relatedness patterns in bottlenose dolphins (Tursiops sp.) with different alliance strategies. Proceedings of the Royal Society Biological Sciences Series B [print] 270 (1514) 7: 497-502.
Mann J. 2000. Unraveling the Dynamics of Social Life. Pp. 45 – 64. In: Mann J., Connor R.C., Tyack P.L., Whitehead H. 2000. Cetacean Societies: Field Studies of Dolphins and Whales. University of Chicago Press. Chicago.
Maze K. S., Wursig B. 1999. Bottlenose dolphins of San Luis Pass, Texas: Occurrence patterns, site-fidelity, and habitat use. Aquatic Mammals [print] 25 (2) 91-103.
Mazzoil M., McCulloch S.D., Defran R.H., Murdoch M.E. 2004. Use of Digital Photography and Analysis of Dorsal Fins for Photo-Identification of Bottlenose Dolphins. Aquatic Mammals, 30(2): 209-219.
Nowacek S. M., Wells R.S., Solow A.R. 2001. Short-term effects of boat traffic on bottlenose dolphins, Tursiops truncatus, in Sarasota Bay, Florida. Marine Mammal Science [print] 17 (4): 673-688.
Owen E.C.G., Wells R.S., Hofmann S.2002. Ranging and association patterns of paired and unpaired adult male Atlantic bottlenose dolphins, Tursiops truncatus, in Sarasota, Florida, provide no evidence for alternative male strategies. Canadian Journal of Zoology [print] 80 (12): 2072-2089.
Scott M.D., Wells R.S., Irvine A.B., Mate B.R. 1990. Tagging and Marking Studies on Small Cetaceans. Pp. 489-514. In: Leatherwood S., Reeves R.R. 1990. The Bottlenose Dolphin. Academic Press, Inc. San Diego, CA.
Smolker R., Richards A., Connor R., Mann J., Berggren P. 1997. Sponge carrying by dolphins (Delphinidae, Tursiops sp.): A foraging specialization involving tool use? Ethology 103 (6): 454-465.
Stevick P.T., Palsboll P.J., Smith T.D., Bravington M.V., Hammond P.S. 2001. Errors in identification using natural markings: rates, sources and effects on capture-recapture estimates of abundance. Canadian Journal of Fishery and Aquatic Science 58: 1861-1870
Wells R.S. 2002. Identification Methods. Pp. 601-608. In: Perrin W.F., Wursig B., Thewissen J.G.M. 2002. Encyclopedia of Marine Mammals. Academic Press. San Diego, CA.
Whitehead H., Christal J., Tyack P. 2000. Studying Cetacean Social Structure in Space and Time. Pp. 65-87. In: Mann J., Connor R.C., Tyack P.L., Whitehead H. 2000. Cetacean Societies: Field Studies of Dolphins and Whales. University of Chicago Press. Chicago.
 Copyright © 2007 Dolphin Research Center
Information Can Be Used with Proper Citing
Dolphin Research Center, 58901 Overseas Highway, Grassy Key, FL 33050-6019
305-289-1121 www.dolphins.org
|
