Recommendations for photo‐identification methods used in capture‐recapture models with cetaceans

Capture‐recapture methods are frequently employed to estimate abundance of cetaceans using photographic techniques and a variety of statistical models. However, there are many unresolved issues regarding the selection and manipulation of images that can potentially impose bias on resulting estimates. To examine the potential impact of these issues we circulated a test data set of dorsal fin images from bottlenose dolphins to several independent research groups. Photo‐identification methods were generally similar, but the selection, scoring, and matching of images varied greatly amongst groups. Based on these results we make the following recommendations. Researchers should: (1) determine the degree of marking, or level of distinctiveness, and use images of sufficient quality to recognize animals of that level of distinctiveness; (2) ensure that markings are sufficiently distinct to eliminate the potential for “twins” to occur; (3) stratify data sets by distinctiveness and generate a series of abundance estimates to investigate the influence of including animals of varying degrees of markings; and (4) strive to examine and incorporate variability among analysts into capture‐recapture estimation. In this paper we summarize these potential sources of bias and provide recommendations for best practices for using natural markings in a capture‐recapture framework.     

MEASUREMENT OF PHOTOGRAPHIC QUALITY AND INDIVIDUAL DISTINCTIVENESS FOR THE PHOTOGRAPHIC IDENTIFICATION OF HUMPBACK WHALES, MEGAPTERA NOVAEANGLIAE

Accurate identification of humpback whales from photographic identification data depends on the quality of the photographs and the distinctiveness of the flukes. Criteria for evaluating photographic quality and individual distinctiveness were developed involving judgments about overall quality or distinctiveness and about specific aspects of each. These criteria were tested for the level of agreement among judges. The distinctiveness scheme was tested for the independence of distinctiveness judgments and photographic quality. Our results show that judges could agree when evaluating specific and overall aspects of photographic quality and individual distinctiveness. The level of agreement varied for different pairs of judges, and less adept judges were identified. Ability to agree on evaluations of photographic quality was independent of the experience of the judges. Overall photographic quality and overall distinctiveness were successfully predicted from more specific variables, but the agreement between judges for these was not significantly greater than the agreement for the overall measures judged directly. There was no correlation between individual distinctiveness and photographic quality for four of the five judges, but the power of this rest may be low. Analyses of photographic identification data frequently require evaluations of photographic quality and individual distinctiveness. To obtain reliable results from such analyses, evaluation schemes and judges should be tested to ensure reliable and consistent evaluations.     

AN OCEAN-BASIN-WIDE MARK-RECAPTURE STUDY OF THE NORTH ATLANTIC HUMPBACK WHALE (MEGAPTERA NOVAEANGLIAE)

Although much is known about the humpback whale, Megaptera novaeangliae, regional studies have been unable to answer several questions that are central to the conservation and management of this endangered species. To resolve uncertainties about population size, as well as the spatial and genetic structure of the humpback whale population in the North Atlantic, we conducted a two-year ocean-basin-wide photographic and biopsy study in 1992-1993. Photographic and skin-biopsy sampling was conducted of animals in feeding and breeding areas throughout most of the range of this species in the North Atlantic, from the West Indies breeding grounds through all known feeding areas as far north as arctic Norway. A standardized sampling protocol was designed to maximize sample sizes while attempting to ensure equal probability of sampling, so that estimates of abundance would be as accurate and as precise as possible. During 666 d at sea aboard 28 vessels, 4,207 tail fluke photographs and 2,326 skin biopsies were collected. Molecular analyses of all biopsies included determination of sex, genotype using six microsatellite loci, and mitochondrial control region sequence. The photographs and microsatellite loci were used to identify 2,998 and 2,015 individual whales, respectively. Previously published results from this study have addressed spatial distribution, migration, and genetic relationships. Here, we present new estimates of total abundance in this ocean using photographic data, as well as overall and sex-specific estimates using biopsy data. We identify several potential sampling biases using only breeding-area samples and report a consistent mark-recapture estimate of oceanwide abundance derived from photographic identification, using both breeding and feeding-area data, of 10,600 (95% confidence interval 9,300-12,100). We also report a comparable, but less precise, biopsy-based estimate of 10,400 (95% confidence interval of 8,000-13,600). These estimates are significantly larger and more precise than estimates made for the 1980s, potentially reflecting population growth. In contrast, significantly lower and less consistent estimates were obtained using between-feeding-area or between-breeding-area sampling. Reasons for the lower estimates using the results of sampling in the same areas in subsequent years are discussed. Overall, the results of this ocean-basin-wide study demonstrate that an oceanwide approach to population assessment of baleen whales is practicable and results in a more comprehensive understanding of population abundance and biology than can be gained from smaller-scale efforts.     

Estimating Alpine ibex Capra ibex abundance from photographic sampling

Monitoring procedures for Alpine ibex Capra ibex are limited in habitats with reduced visibility and when physical capture and marking of the animals is not intended. Photographic sampling, involving using camera‐trap data and identifying ibex from natural markings, was adopted with capture‐recapture models to estimate the abundance of ibex in Austria. The software CAPTURE's model produced an average capture probability of 0.44 with an estimate of 34–51 ibex and a mean population size of 38 ibex. This first study showed the applicability of photographic capture‐recapture techniques to estimate the abundance of ibex based on their natural markings.     

The effect of attractant lures in camera trapping: a case study of population estimates for the Iberian lynx (Lynx pardinus)

Capture–recapture analysis of camera trap data is a conventional method to estimate the abundance of free-ranging wild felids. Due to notorious low detection rates of felids, it is important to increase the detection probability during sampling. In this study, we report the effectiveness of attractants as a tool for improving the efficiency of camera trap sampling in abundance estimation of Iberian lynx. We developed a grid system of camera stations in which stations with and without attractant lures were spatially alternated across known Iberian lynx habitat. Of the ten individuals identified, five were detected at stations with no attractant (blind sets), and nine, at the lured stations. Thirty-eight percent of blind set station’s independent captures and 10?% of lured station’s independent captures resulted in photographs unsuitable for correct individual identification. The total capture probability at lured stations was higher than that obtained at blind set stations. The estimates obtained with blind set cameras underestimated the number of lynxes compared to lured cameras. In our study, it appears that the use of lures increased the efficiency of trail camera captures and, therefore, the accuracy of capture–recapture analysis. The observed failure to detect known individuals at blind set camera stations may violate capture–recapture assumptions and bias abundance estimates.     

标记重捕法在中华白海豚种群动态研究的应用及建议

早期的中华白海豚考察主要依赖样线调查法了解其资源分布,而近期研究更多采纳标记重捕法获取种群动态信息。在辨识个体的基础上,后者能够获取多种种群参数开展种群生存力分析。本文回顾在我国海域开展的中华白海豚种群动态研究进展及各地区种群标记重捕数据的累积情况;通过数据模拟评估努力值如何影响种群大小统计的误差和偏差;综合阐述野外考察方案设计、标志筛选和数据处理对数据分析的潜在影响;强调模型拟合优度检验和模型选择的重要性;最后,针对比较不同时期或不同方法获取的种群信息时的常见误解提出我们的意见。本文旨在帮助完善我国中华白海豚的后期资源监测工作。     

Application of photographic capture–recapture modelling to estimate demographic parameters for male Asian elephants

In addition to the threats of habitat loss and degradation, adult males of the Asian elephant Elephas maximus also face greater threats from ivory poaching and conflict with humans. To understand the impact of these threats, conservationists need robust estimates of abundance and vital rates specifically for the adult male segment of elephant populations. By integrating the identification of individual male elephants in a population from distinct morphology and natural markings, with modern capture–recapture (CR) sampling designs, it is possible to estimate various demographic parameters that are otherwise difficult to obtain from this long-lived and wide-ranging megaherbivore. In this study, we developed systematic individual identification protocols and integrated them into CR sampling designs to obtain capture histories and thereby estimate the abundance of adult bull elephants in a globally important population in southern India. We validated these estimates against those obtained from an independent method combining line-transect density estimates with age–sex composition data for elephants. The sampled population was open to gains and losses between sampling occasions. The abundance of adult males in the 176 km² study area was (SÊ ) = 134(14.20) and they comprised 14% (±1%) of the total elephant population. Time-specific abundance estimates for each sampling occasion showed a distinct increase in adult male numbers over the sampling period, explained by seasonal patterns of local migration. CR-based estimates for adult male abundance closely matched estimates from distance-based methods. Thus, while providing abundance data of comparable rigour and precision, photographic CR methods permit estimation of demographic parameters for the Asian elephant that are both urgently needed and difficult to obtain.     

Genetic fingerprinting proves cross‐correlated automatic photo‐identification of individuals as highly efficient in large capture–mark–recapture studies

Capture–mark–recapture (CMR) approaches are the backbone of many studies in population ecology to gain insight on the life cycle, migration, habitat use, and demography of target species. The reliable and repeatable recognition of an individual throughout its lifetime is the basic requirement of a CMR study. Although invasive techniques are available to mark individuals permanently, noninvasive methods for individual recognition mainly rest on photographic identification of external body markings, which are unique at the individual level. The re‐identification of an individual based on comparing shape patterns of photographs by eye is commonly used. Automated processes for photographic re‐identification have been recently established, but their performance in large datasets (i.e., > 1000 individuals) has rarely been tested thoroughly. Here, we evaluated the performance of the program AMPHIDENT, an automatic algorithm to identify individuals on the basis of ventral spot patterns in the great crested newt (Triturus cristatus) versus the genotypic fingerprint of individuals based on highly polymorphic microsatellite loci using GENECAP. Between 2008 and 2010, we captured, sampled and photographed adult newts and calculated for 1648 samples/photographs recapture rates for both approaches. Recapture rates differed slightly with 8.34% for GENECAP and 9.83% for AMPHIDENT. With an estimated rate of 2% false rejections (FRR) and 0.00% false acceptances (FAR), AMPHIDENT proved to be a highly reliable algorithm for CMR studies of large datasets. We conclude that the application of automatic recognition software of individual photographs can be a rather powerful and reliable tool in noninvasive CMR studies for a large number of individuals. Because the cross‐correlation of standardized shape patterns is generally applicable to any pattern that provides enough information, this algorithm is capable of becoming a single application with broad use in CMR studies for many species.     

How common is common? Rapidly assessing population size and structure of the frog Mantidactylus betsileanus at a site in east‐central Madagascar

Population‐level data are urgently needed for amphibians in light of the ongoing amphibian extinction crisis. Studies focused on population dynamics are not only important for rare species but also for common species which shape ecosystems to a greater degree than those that are rare. Some of the greatest global amphibian species diversity is found in Madagascar, yet there are few studies on the ecology of frog species on the island. We carried out a mark‐recapture study on the widespread frog Mantidactylus betsileanus (Mantellidae: Mantellinae: Mantidactylus) at two adjacent rainforest sites in east‐central Madagascar to assess its population size and structure. To do so, we validated and implemented an individual identification protocol using photographs of the ventral patterns of frogs and identified individuals with photographic‐matching software. Using this rapid, non‐invasive survey method, we were able to estimate a density of 26 and 28 frogs per 100 m² at each of the two sites sampled. Our results show the rainforests near the village of Andasibe, Madagascar support remarkably high amphibian abundance, helping illustrate the significant ecological role of frogs in this ecosystem. Further, individual frog markings allowed us to develop more precise estimates than traditional survey methods. This study provides a blueprint to augment existing population studies or develop new monitoring programs in Madagascar and beyond.     

Humpback whale abundance in the North Pacific estimated by photographic capture‐recapture with bias correction from simulation studies

We estimated the abundance of humpback whales in the North Pacific by capture‐recapture methods using over 18,000 fluke identification photographs collected in 2004–2006. Our best estimate of abundance was 21,808 (CV = 0.04). We estimated the biases in this value using a simulation model. Births and deaths, which violate the assumption of a closed population, resulted in a bias of +5.2%, exclusion of calves in samples resulted in a bias of ?10.5%, failure to achieve random geographic sampling resulted in a bias of ?0.4%, and missed matches resulted in a bias of +9.3%. Known sex‐biased sampling favoring males in breeding areas did not add significant bias if both sexes are proportionately sampled in the feeding areas. Our best estimate of abundance was 21,063 after accounting for a net bias of +3.5%. This estimate is likely to be lower than the true abundance due to two additional sources of bias: individual heterogeneity in the probability of being sampled (unquantified) and the likely existence of an unknown and unsampled breeding area (?8.7%). Results confirm that the overall humpback whale population in the North Pacific has continued to increase and is now greater than some prior estimates of prewhaling abundance.     

Abundance and Survival Rates of the Hawai’i Island Associated Spinner Dolphin (Stenella longirostris) Stock

Reliable population estimates are critical to implement effective management strategies. The Hawai’i Island spinner dolphin (Stenella longirostris) is a genetically distinct stock that displays a rigid daily behavioural pattern, foraging offshore at night and resting in sheltered bays during the day. Consequently, they are exposed to frequent human interactions and disturbance. We estimated population parameters of this spinner dolphin stock using a systematic sampling design and capture–recapture models. From September 2010 to August 2011, boat-based photo-identification surveys were undertaken monthly over 132 days (>1,150 hours of effort; >100,000 dorsal fin images) in the four main resting bays along the Kona Coast, Hawai’i Island. All images were graded according to photographic quality and distinctiveness. Over 32,000 images were included in the analyses, from which 607 distinctive individuals were catalogued and 214 were highly distinctive. Two independent estimates of the proportion of highly distinctive individuals in the population were not significantly different (p = 0.68). Individual heterogeneity and time variation in capture probabilities were strongly indicated for these data; therefore capture–recapture models allowing for these variations were used. The estimated annual apparent survival rate (product of true survival and permanent emigration) was 0.97 SE±0.05. Open and closed capture–recapture models for the highly distinctive individuals photographed at least once each month produced similar abundance estimates. An estimate of 221±4.3 SE highly distinctive spinner dolphins, resulted in a total abundance of 631±60.1 SE, (95% CI 524–761) spinner dolphins in the Hawai’i Island stock, which is lower than previous estimates. When this abundance estimate is considered alongside the rigid daily behavioural pattern, genetic distinctiveness, and the ease of human access to spinner dolphins in their preferred resting habitats, this Hawai’i Island stock is likely more vulnerable to negative impacts from human disturbance than previously believed.     

Validation of mark‐recapture population estimates for invasive common carp,Cyprinus carpio,in Lake Crescent,Tasmania

A mark‐recapture study based on the Petersen method was implemented in 1998 to estimate the abundance of the invasive common carp, Cyprinus carpio L., in Lake Crescent, Tasmania. Multiple gear types were employed to minimise capture bias, with multiple capture and recapture events providing an opportunity to compute and compare Petersen and Schnabel estimates. A single Petersen estimate on recapture data and two Schnabel estimates – one each on mark (forward‐Schnabel estimate) and recapture (reverse‐Schnabel estimate) data – were conducted. An independent long‐term double tag study facilitated estimation of the annual natural mortality. Subsequent fish‐down of the population suggests that, in all likelihood, the carp have been eradicated from the lake, providing an unprecedented opportunity to verify the forward population estimates carried out in 1998. Results suggest that all three estimates were close to the true population size, with the reverse‐Schnabel estimate being the most accurate and within 1% of the true population in this relatively large lake (～2365 ha). Greater accuracy of the reverse‐Schnabel approach can be attributed to either minimised fish behavioural (i.e. gear susceptibility or avoidance) or computational bias associated with the forward‐Schnabel and Petersen approaches, respectively. While the original estimates served as a guide in eradication of carp from the lake, the ultimate validation provides a reliable framework for abundance estimation of this invasive fish in relatively large water bodies elsewhere.     

Survival estimation of a cryptic antelope via photographic capture–recapture

Adult survival is a primary determinant of abundance and dynamics of large herbivore populations. For species that are inconspicuous, however, accurate survival estimation depends on accommodating low detection probability. For species with individually recognizable markings, photographic capture–recapture (CR) provides an approach to estimate population parameters while accounting for imperfect detection. I investigated the use of photographic CR for a cryptic large herbivore, the nyala, in a region of Hluhluwe–iMfolozi Park, South Africa. I conducted photographic sampling based on the closed robust design, with 5–6 daily sampling occasions nested within three week‐long sampling periods, which delineated one dry and one wet season. Detection differed between sexes: encounter probability of female adults depended on whether individuals fell into high‐encounter (seasonal range: 0.61–0.71) or low‐encounter (seasonal range: 0.29–0.40) groups, whereas male adults had a constant encounter probability of 0.39 per day. For both sexes, monthly survival probability was ≥0.93 and did not differ appreciably between seasons or sexes. Given the role of survival in population dynamics, photographic CR has the potential to provide survival estimates for cryptic large herbivores that lack such information.     

Striped dolphins and short‐beaked common dolphins in the Gulf of Corinth,Greece: Abundance estimates from dorsal fin photographs

Capture‐recapture methods relying on dorsal fin natural markings have never been applied successfully to striped dolphins, Stenella coeruleoalba, and were rarely used to assess abundance of short‐beaked common dolphins, Delphinus delphis. We used digital photo‐identification to obtain abundance estimates of striped and common dolphins living in mixed groups in the Gulf of Corinth, Greece. The proportion of either species was calculated based on the relative number of photographs of adult animals showing relevant portions of their body during conspicuous surfacings. Striped dolphins and common dolphins averaged 95.0% and 3.2% of all individuals, respectively. Animals showing intermediate pigmentation accounted for another 1.8%. Striped dolphin numbers were relatively high, with a point estimate of 835 animals (95% CI = 631–1,106). Common dolphins numbers were low (point estimate 28 animals; 95% CI = 11–73) and individuals were scattered within striped dolphin groups, indicating that this common dolphin population may be nonviable. Within a semiclosed Gulf exposed to considerable anthropogenic impact, the future of both dolphin species is of concern due to their suspected geographic isolation and restricted extent of occurrence. Information provided here can be used to inform timely conservation efforts.     

Comparison of photo‐matching algorithms commonly used for photographic capture–recapture studies

Photographic capture–recapture is a valuable tool for obtaining demographic information on wildlife populations due to its noninvasive nature and cost‐effectiveness. Recently, several computer‐aided photo‐matching algorithms have been developed to more efficiently match images of unique individuals in databases with thousands of images. However, the identification accuracy of these algorithms can severely bias estimates of vital rates and population size. Therefore, it is important to understand the performance and limitations of state‐of‐the‐art photo‐matching algorithms prior to implementation in capture–recapture studies involving possibly thousands of images. Here, we compared the performance of four photo‐matching algorithms; Wild‐ID, I3S Pattern+, APHIS, and AmphIdent using multiple amphibian databases of varying image quality. We measured the performance of each algorithm and evaluated the performance in relation to database size and the number of matching images in the database. We found that algorithm performance differed greatly by algorithm and image database, with recognition rates ranging from 100% to 22.6% when limiting the review to the 10 highest ranking images. We found that recognition rate degraded marginally with increased database size and could be improved considerably with a higher number of matching images in the database. In our study, the pixel‐based algorithm of AmphIdent exhibited superior recognition rates compared to the other approaches. We recommend carefully evaluating algorithm performance prior to using it to match a complete database. By choosing a suitable matching algorithm, databases of sizes that are unfeasible to match “by eye” can be easily translated to accurate individual capture histories necessary for robust demographic estimates.     

Killer whale abundance and predicted narwhal consumption in the Canadian Arctic

Range expansions and increases in the frequency of killer whale (Orcinus orca) sightings have been documented in the eastern Canadian Arctic, presumably the result of climate change‐related sea‐ice declines. However, the effects of increased predator occurrence on this marine ecosystem remain largely unknown. We explore the consequences of climate change‐related range expansions by a top predator by estimating killer whale abundance and their possible consumptive effects on narwhal (Monodon monoceros) in the Canadian Arctic. Individual killer whales can be identified using characteristics such as acquired scars and variation in the shape and size of their dorsal fins. Capture–mark–recapture analysis of 63 individually identifiable killer whales photographed between 2009 and 2018 suggests a population size of 163 ± 27. This number of killer whales could consume >1,000 narwhal during their seasonal residency in Arctic waters. The effects of such mortality at the ecosystem level are uncertain, but trophic cascades caused by top predators, including killer whales, have been documented elsewhere. These findings illustrate the magnitude of ecosystem‐level modifications that can occur with climate change‐related shifts in predator distributions.     

Genotyping validates photo‐identification by the head scale pattern in a large population of the European adder (Vipera berus)

Capture‐mark‐recapture procedures are a basic tool in population studies and require that individual animals are correctly identified throughout their lifetime. A method that has become more and more popular uses photographic records of natural markings, such as pigmentation pattern and scalation configuration. As with any other marking tool, the validity of the photographic identification technique should be evaluated thoroughly. Here, we report on a large‐scale double‐marking study in which European adders (Vipera berus) were identified by both microsatellite genetic markers and by the pattern of head scalation. Samples that were successfully genotyped for all nine loci yielded 624 unique genotypes, which matched on a one‐to‐one basis with the individual assignments based on the head scalation pattern. Thus, adders considered as different individuals by their genotypes were also identified as different individuals by their head scalation pattern, and vice versa. Overall, variation in the numbers, shape, and arrangement of the head scales enabled us to distinguish among 3200+ photographed individual snakes. Adders that were repeatedly sequenced genetically over intervals of 2–3 years showed no indication whatsoever for a change in the head scale pattern. Photographic records of 900+ adders that were recaptured over periods of up to 12 years showed a very detailed and precise match of the head scale characteristics. These natural marks are thus robust over time and do not change during an individual's lifetime. With very low frequency (0.3%), we detected small changes in scalation that were readily discernible and could be attributed to physical injury or infection. Our study provides a conclusive validation for the use of photo‐identification by head scale patterns in the European adder.     

Combining capture–recapture data and known ages allows estimation of age‐dependent survival rates

In many animal populations, demographic parameters such as survival and recruitment vary markedly with age, as do parameters related to sampling, such as capture probability. Failing to account for such variation can result in biased estimates of population‐level rates. However, estimating age‐dependent survival rates can be challenging because ages of individuals are rarely known unless tagging is done at birth. For many species, it is possible to infer age based on size. In capture–recapture studies of such species, it is possible to use a growth model to infer the age at first capture of individuals. We show how to build estimates of age‐dependent survival into a capture–mark–recapture model based on data obtained in a capture–recapture study. We first show how estimates of age based on length increments closely match those based on definitive aging methods. In simulated analyses, we show that both individual ages and age‐dependent survival rates estimated from simulated data closely match true values. With our approach, we are able to estimate the age‐specific apparent survival rates of Murray and trout cod in the Murray River, Australia. Our model structure provides a flexible framework within which to investigate various aspects of how survival varies with age and will have extensions within a wide range of ecological studies of animals where age can be estimated based on size.     

Male-biased sex ratios in mature damselfly populations: real or artefact?

1. There is ongoing controversy about whether biased sex ratios in diploid insect populations are real or an artefact caused by different behaviours and/or different catchability of the sexes. This was tested by monitoring two field and three semi-natural populations of the damselfly Lestes sponsa. 2. Capture–mark–recapture data showed that population size estimates were about twice as large for males as for females at both field sites. Independent estimates of the sex ratios based on total numbers of males and females captured supported the male bias. 3. Males had higher recapture probabilities than females due to longer times between successive visits in females. Because the same pattern was found in the semi-natural populations, the longer intervals in females are no artefact due to their lower detectability. 4. Theoretical models show that the strong temporary emigration of females tends, if anything, to overestimate female population sizes and that the heterogeneity of recapture probabilities observed in males tends to underestimate male population sizes. Hence, behavioural differences between the sexes do not cause an artificially male-biased sex ratio. 5. Spatial data show that operational sex ratios are male biased at the pond but become female biased in the plots further away from the shoreline; however because of the decrease in densities away from the shoreline, this does not result in a global even sex ratio. 6. Spatial data, temporary emigration patterns, and independent estimates suggest strongly that the male-biased sex ratios in mature damselfly populations are real.