Skin Marks in Critically Endangered Taiwanese Humpback Dolphins (Sousa chinensis taiwanensis)

Simple Summary The population of Indo-Pacific humpback dolphin (Sousa chinensis) endemic to the inshore and estuarine waters of western Taiwan was identified as a subspecies (Sousa chinensis ssp. taiwanensis) and designated as “Critically Endangered” by the International Union for Conservation of Nature (IUCN). This population is at a high risk of extirpation, faces being impacted by some major anthropogenic threats, and numbers fewer than 65 individuals. Through a combination of long-term observations and photographic monitoring of identifiable individuals and stranding responses, accurate identification and classification of human-induced injuries and skin lesions is an effective approach and is particularly important to the management of the declining population. The results showed that the prevalence of skin marks is higher or comparable to the previous studies on the coastal cetacean populations across the globe. As the total population comprises dozens of individuals, radical measures to conserve the population and reduce environmental hazards are considered indispensable. Abstract As long-lived apex predators, Indo-Pacific humpback dolphins (Sousa chinensis) are key indicators of marine coastal environmental health. Skin marks can be observed on dorsal body surfaces that are visible during mark–recapture studies that rely on photo-identification (photo-ID) methods. Skin mark prevalence may be an indicator of environmental or anthropogenic stressors in the ecosystem, which may lead to individual and/or population-level health concerns. The prevalence of marks of anthropogenic origin was assessed in the Critically Endangered S. chinensis taiwanensis population along the coasts of central Taiwan. Fifty, twenty-eight, and thirty-four individuals were identified in 2018, 2019, and 2021, respectively. At least one category of injuries was observed in 47 of 57 distinctive individuals (82%), and adults showed a higher prevalence of deep injuries than the other coloration stages. At least one category of skin lesion was observed in 33 of 57 distinctive individuals (58%), and high prevalence of skin lesions was found in mature individuals. Given the difficulty in taking direct observations, skin mark prevalence is proposed as a proxy for estimating habitat health and anthropogenic stressors upon S. chinensis taiwanensis. The moderate-to-high prevalence of skin marks in this study was designated as a warning of risks. This study provides important updated information for the assessment of the health and survival of this population. More effective management measures are urgently needed to reverse the observed population decline.


Introduction
Indo-Pacific humpback dolphins (Sousa chinensis) dwell in tropical and subtropical close shore waters belonging to the Indian Ocean and Pacific Oceans from northern Australia, the southern parts of China, extending to southeastern Asia, and around the borders of the Indian Ocean, extending to southern Africa [1]. S. chinensis is red-listed by the In this study, with the aid of photo-ID, skin marks were visually assessed and quantified for S. c. taiwanensis off central Taiwan. The coastal waters of central Taiwan are one of most important habitats for this population, but there is an increasing amount of human activity in this area (reviewed in [5]). The objectives of our work were to (1) use photo-ID to classify and quantify the types of skin marks observed on S. c. taiwanensis off central Taiwan from 2018 to 2019 and in 2021 using descriptions from previous studies and (2) estimate the overall skin mark prevalence among S. c. taiwanensis off central Taiwan and the risks they encounter. Mark presence may provide insight into the condition of individual humpback dolphins. Thus, deriving updated baseline estimates for the prevalence of the skin marks can be useful for tracking population health.

Materials and Methods
In 2020, the Ocean Affairs Council in Taiwan announced a "Major Wildlife Habitat of Indo-Pacific Humpback Dolphin" as an area within 1-3 nautical miles offshore from the west coast of Taiwan, from 24 • 42 N to 23 • 26 N. The major wildlife habitat involves four counties: Miaoli, Taichung, Changhua, and Yunlin. This study focused on two counties located within the central habitat area: Taichung County (24 • 58 N~24 • 11 N) and Changhua County (24 • 11 N~23 • 51 N) ( Figure 1). The study area included waters from the shoreline to shallow waters with depths < 15 m. Boat-based photo-ID surveys were conducted in 2018 (May to September), 2019 (April to September), and 2021 (April to November) by traveling at an average speed of 6 to 9 knots (13.89 km/h), with an average of 2 to 3 knots (4.63 km/h) when approaching animals using 10~12-meter-long CT-1 fishing boats. Dolphins were photographed using digital single-lens reflex cameras (Canon, Olympus, Pentax, or Nikon with 70-300 mm zoom lenses or 400 mm prime lenses). Every image was cataloged and cut into shape using PhotoImpact 11software, and individuals were further identified from those images by their distinctive scars and markings.   Identified individuals were grouped into five coloration stages based on changes in body color [21]: gray and unspotted (calf), gray with some white spotting (mottled), white with lots of black spotting (speckled), white with less black spotting (spotted adult), and white with a pinkish tint and no spotting (unspotted adult) ( Figure 2). In this study, markings that possibly resulted from conspecifics or sharks were excluded from the analysis. Injuries were classified into five categories: dorsal fin/fluke mutilation, narrow-spaced linear marks, wide-spaced linear marks, back indentation, and others (Table 1). Skin lesions were classified into seven categories: nodules, hypertrophic scars, barnacles, pale, black patch, red patch, and orange/yellow patch (Table 2).     [10,22,23] Narrow-spaced linear marks Serial linear marks, including both shallow and deep marks, especially those appearing along the spine of the peduncle net entanglement or inherent deformities [23,24] Wide-spaced linear marks Serial linear marks that extend from one side of the peduncle to the other side, with a long and equal interval between each mark net entanglement or ship strike [23,24] cut from net entanglement or other object [10,22,23] Narrow-spaced linear marks Serial linear marks, including both shallow and deep marks, especially those appearing along the spine of the peduncle Animals 2023, 13, x FOR PEER REVIEW 6 of 17 Table 1. Definitions of injury (arrows and circles) categories and potential causes.

Injury Categories Description Examples Potential Causes References
Dorsal fin/fluke mutilation Missing the apex of the dorsal fin/fluke, with a sharp edge or a linear cut down the dorsal fin cut from net entanglement or other object [10,22,23] Narrow-spaced linear marks Serial linear marks, including both shallow and deep marks, especially those appearing along the spine of the peduncle net entanglement or inherent deformities [23,24] Wide-spaced linear marks Serial linear marks that extend from one side of the peduncle to the other side, with a long and equal interval between each mark net entanglement or ship strike [23,24] net entanglement or inherent deformities [23,24] Wide-spaced linear marks Serial linear marks that extend from one side of the peduncle to the other side, with a long and equal interval between each mark Animals 2023, 13, x FOR PEER REVIEW 6 of 17 Table 1. Definitions of injury (arrows and circles) categories and potential causes.

Injury Categories Description Examples Potential Causes References
Dorsal fin/fluke mutilation Missing the apex of the dorsal fin/fluke, with a sharp edge or a linear cut down the dorsal fin cut from net entanglement or other object [10,22,23] Narrow-spaced linear marks Serial linear marks, including both shallow and deep marks, especially those appearing along the spine of the peduncle net entanglement or inherent deformities [23,24] Wide-spaced linear marks Serial linear marks that extend from one side of the peduncle to the other side, with a long and equal interval between each mark net entanglement or ship strike [23,24]  Back indentation V-shaped indentation located along the spine; size varies net entanglement or ship strike [10,23,25] Others Marks seemingly unrelated to natural causes and that cannot be assigned to the other scar categories cut from net entanglement or other object [26] cut from net entanglement or other object [26] Animals 2023, 13, 608

Nodules
Oval or circular cutaneous elevations of the skin fungal or bacterial infection [27] Hypertrophic Scars Raised lesions with oval shapes infection or immune deficiency [22] Barnacles One or multiple hanging barnacles, especially seen on dorsal fins and flukes swimming slowly due to health issues [28] fungal or bacterial infection [27] Hypertrophic Scars Raised lesions with oval shapes

Nodules
Oval or circular cutaneous elevations of the skin fungal or bacterial infection [27] Hypertrophic Scars Raised lesions with oval shapes infection or immune deficiency [22] Barnacles One or multiple hanging barnacles, especially seen on dorsal fins and flukes swimming slowly due to health issues [28] infection or immune deficiency [22] Barnacles One or multiple hanging barnacles, especially seen on dorsal fins and flukes Animals 2023, 13, x FOR PEER REVIEW 8 of 17 Table 2. Definitions of skin lesion (arrows and circles) categories and potential causes.

Skin Lesion Categories Description Examples Potential Causes References
Nodules Oval or circular cutaneous elevations of the skin fungal or bacterial infection [27] Hypertrophic Scars Raised lesions with oval shapes infection or immune deficiency [22] Barnacles One or multiple hanging barnacles, especially seen on dorsal fins and flukes swimming slowly due to health issues [28] swimming slowly due to health issues [28] Pale White in color and irregular in shape; diffuse edges and of any size

Skin Lesion Categories Description Examples Potential Causes References
Black patch Uniform in color and irregular in shape; diffuse edges Pale White in color and irregular in shape; diffuse edges and of any size fungal infection [9,22] Black patch Uniform in color and irregular in shape; diffuse edges pox virus infection [9,22,25] Red patch Pink patch with multiple small red spots scattered around; irregular in shape bacterial or viral infection [27] Orange/yellow patch Orange or yellow coloration with diffuse edges attached diatom [22,27,29] pox virus infection [9,22,25] Red patch Pink patch with multiple small red spots scattered around; irregular in shape Pale White in color and irregular in shape; diffuse edges and of any size fungal infection [9,22] Black patch Uniform in color and irregular in shape; diffuse edges pox virus infection [9,22,25] Red patch Pink patch with multiple small red spots scattered around; irregular in shape bacterial or viral infection [27] Orange/yellow patch Orange or yellow coloration with diffuse edges attached diatom [22,27,29] bacterial or viral infection [27] Orange/yellow patch Orange or yellow coloration with diffuse edges Pale White in color and irregular in shape; diffuse edges and of any size fungal infection [9,22] Black patch Uniform in color and irregular in shape; diffuse edges pox virus infection [9,22,25] Red patch Pink patch with multiple small red spots scattered around; irregular in shape bacterial or viral infection [27] Orange/yellow patch Orange or yellow coloration with diffuse edges attached diatom [22,27,29] attached diatom [22,27,29] Animals 2023, 13, 608 9 of 16

Results
In the three years (2018-2019 and 2021), 87 boat surveys were performed (43,20, and 24 surveys, respectively), and 73,073 good-quality images were analyzed. In total, 50 individuals were identified in 2018, 28 individuals were identified in 2019, and 34 individuals were identified in 2021. In total, 57 distinctive individuals were identified. Among all 57 distinctive individuals, 17.5% (n = 10) were identified as calves, 29.8% (n = 17) were identified in the mottled stage, 28% (n = 16) were identified in the speckled stage, 21% (n = 12) were identified as spotted adults, and 3.5% (n = 2) were identified as unspotted adults. Spotted adults and unspotted adults were pooled into the adult stage for further analysis. Among the 57 distinctive individuals, 17 were resighted in all three years, including four mottled, nine speckled, and four spotted adults. Images of these 17 individuals were further analyzed for an interannual comparison.

Results
In the three years (2018-2019 and 2021), 87 boat surveys were performed (43,20, and 24 surveys, respectively), and 73,073 good-quality images were analyzed. In total, 50 individuals were identified in 2018, 28 individuals were identified in 2019, and 34 individuals were identified in 2021. In total, 57 distinctive individuals were identified. Among all 57 distinctive individuals, 17.5% (n = 10) were identified as calves, 29.8% (n = 17) were identified in the mottled stage, 28% (n = 16) were identified in the speckled stage, 21% (n = 12) were identified as spotted adults, and 3.5% (n = 2) were identified as unspotted adults. Spotted adults and unspotted adults were pooled into the adult stage for further analysis. Among the 57 distinctive individuals, 17 were resighted in all three years, including four mottled, nine speckled, and four spotted adults. Images of these 17 individuals were further analyzed for an interannual comparison.

Prevalence of Skin Lesions
At least one category of skin lesion was observed in 33 of 57 distinctive individuals (58%). The prevalence of skin lesions was 62% (31/50) in 2018, 39% (11/28) in 2019, and 41% (14/34) in 2021. Among the 57 distinctive individuals, the prevalence of the six categories of skin lesions was as follows: nodules (51%), orange/yellow patch (44%), hypertrophic scars (30%), barnacles (12%), white patch (2%), and black patch (2%). Nodules were the most prevalent skin lesion in each year ( Figure 5 and Table 4). A pale lesion was observed in only one individual in three years, and a black patch only appeared in another individual sighted in both 2018 and 2019. Orange/yellow patches and barnacles were not observed in 2019. The prevalence of skin lesions of the 57 distinctive individuals in each age class was as follows: calf (10%, 1/10), mottled (24%, 4/17), speckled (88%, 14/16), and adult stage (100%, 14/14). The adult stage showed a higher prevalence of nodules, orange/yellow patches, hypertrophic scars, and red patches ( Figure 6). Among the 17 resighted individuals, no skin lesions were noticed in three individuals. In addition, at least one category of skin lesion was found in seven individuals in all three years, while other individuals showed different conditions of lesion remission and relapse (Figure 7).

Prevalence of Skin Lesions
At least one category of skin lesion was observed in 33 of 57 distinctive individuals (58%). The prevalence of skin lesions was 62% (31/50) in 2018, 39% (11/28) in 2019, and 41% (14/34) in 2021. Among the 57 distinctive individuals, the prevalence of the six categories of skin lesions was as follows: nodules (51%), orange/yellow patch (44%), hypertrophic scars (30%), barnacles (12%), white patch (2%), and black patch (2%). Nodules were the most prevalent skin lesion in each year ( Figure 5 and Table 4). A pale lesion was observed in only one individual in three years, and a black patch only appeared in another individual sighted in both 2018 and 2019. Orange/yellow patches and barnacles were not observed in 2019. The prevalence of skin lesions of the 57 distinctive individuals in each age class was as follows: calf (10%, 1/10), mottled (24%, 4/17), speckled (88%, 14/16), and adult stage (100%, 14/14). The adult stage showed a higher prevalence of nodules, orange/yellow patches, hypertrophic scars, and red patches ( Figure 6). Among the 17 resighted individuals, no skin lesions were noticed in three individuals. In addition, at least one category of skin lesion was found in seven individuals in all three years, while other individuals showed different conditions of lesion remission and relapse (Figure 7).

Discussion
There was no consistency in the age reported for dolphins among previous studie on skin scars, and age-class documentation from photo-ID data was usually limited to adult or calf distinctions, which are of limited use in terms of information for ailment affecting sub-adults [11]. Calves of S. chinensis are dark gray or black at birth, and thei coloration lightens through a mottled and speckled gray as they age, becoming juvenile and sub-adults, while adults are spotted and unspotted light gray to white [21]. Since mos species of dolphins do not dramatically change coloration with age, the presumed age class information in S. chinensis provides a good chance to understand the epidemiologica factors influencing the prevalence of skin marks and different mark types in coastal dol phins.
S. chinensis, similar to certain coastal cetacean species, is predominantly a predato and is hence subjected to indirect hazards such as fisheries bycatch, living space destruc tion due to land reclamation and pollution, reduced water discharge into sea-like wate diversions, and noise pollution underwater [2]. These factors contribute to the skin lesion of S. chinensis. Differences in the salinity and temperature of the water are assumed to b important factors in the skin lesions in S. chinensis. However, this could not explain why there is a difference in the prevalence of skin conditions among the different coloration stages. Among other dolphins, similar skin conditions have been supposed to be associ ated with abnormally high levels of residual contaminants in fat tissues [12]. Compro mised immune function associated with environmental pollutants has been considered [30,31]. Documented evidence stated that cetaceans in the Taiwanese region have compar atively lower levels of polychlorinated biphenyls (PCBs), mercury (Hg), and polybromin ated diphenyl ethers (PBDEs) [32][33][34], but later it was proven that these water bodies ar contaminated with heavy metals such as silver (Ag) and Cadmium (Cd), which are poten tial health threats [35][36][37]. The higher prevalence of skin lesions in this study provide important supporting evidence for the bioaccumulation of pollutants. A greate

Discussion
There was no consistency in the age reported for dolphins among previous studies on skin scars, and age-class documentation from photo-ID data was usually limited to adult or calf distinctions, which are of limited use in terms of information for ailments affecting subadults [11]. Calves of S. chinensis are dark gray or black at birth, and their coloration lightens through a mottled and speckled gray as they age, becoming juveniles and sub-adults, while adults are spotted and unspotted light gray to white [21]. Since most species of dolphins do not dramatically change coloration with age, the presumed age-class information in S. chinensis provides a good chance to understand the epidemiological factors influencing the prevalence of skin marks and different mark types in coastal dolphins.
S. chinensis, similar to certain coastal cetacean species, is predominantly a predator and is hence subjected to indirect hazards such as fisheries bycatch, living space destruction due to land reclamation and pollution, reduced water discharge into sea-like water diversions, and noise pollution underwater [2]. These factors contribute to the skin lesions of S. chinensis. Differences in the salinity and temperature of the water are assumed to be important factors in the skin lesions in S. chinensis. However, this could not explain why there is a difference in the prevalence of skin conditions among the different coloration stages. Among other dolphins, similar skin conditions have been supposed to be associated with abnormally high levels of residual contaminants in fat tissues [12]. Compromised immune function associated with environmental pollutants has been considered [30,31]. Documented evidence stated that cetaceans in the Taiwanese region have comparatively lower levels of polychlorinated biphenyls (PCBs), mercury (Hg), and polybrominated diphenyl ethers (PBDEs) [32][33][34], but later it was proven that these water bodies are contaminated with heavy metals such as silver (Ag) and Cadmium (Cd), which are potential health threats [35][36][37]. The higher prevalence of skin lesions in this study provides important supporting evidence for the bioaccumulation of pollutants. A greater prevalence of skin lesions was found in mature animals, and a moderate-to-low prevalence of skin lesions was found among immature animals. Similar findings were reported in a previous study conducted between 2006 and 2010 on the whole population [18]. Furthermore, the prevalence of skin lesions in this study was higher than that in the previous study (from calf to unspotted: 7.7%, 32%, 37.9%, 92.3%, and 75% in [18] vs. 10%, 24%, 88%, 100%, and 100% in this study), indicating potentially significant health risks in this population. The prevalence is comparable to the previous studies on coastal cetacean populations across the globe (reviewed in [18]). It is asserted that there is a need for monitoring programs for pollutant concentrations in S. c. taiwanensis and its habitats, which is imperative to preserve the species and the coastal ecosystem. In addition, the monitoring of pathogens and their epidemiological patterns associated with stress and immunosuppression in dolphins serves as an indicator of ocean health [38]. As the total population comprises dozens of individuals, radical measures to conserve the population and reduce environmental hazards are considered indispensable.
In this study, nodules were highly prevalent (51%) and presented as circumscribed and swollen skin lumps. Orange films were the next most commonly prevalent (44%) lesion in this study. The prevalences of nodules and orange films in the previous study [18] were 15.5% and 11.3%, respectively. The known potentially etiological agents of skin nodules in odontocetes include fungi (Lacazia loboi, Fusarium spp., Paracoccidioides brasiliensis, and Trichophyton spp.), the bacteria Streptococcus iniae, and papillomaviruses [39]. Bottlenose dolphins (Tursiops truncatus) infected by L. loboi from the Indian River Lagoon, Florida, demonstrated marked impairment in adaptive immunity that can be potentially related to prolonged exposure to these environmental stressors. On the other hand, the salinity and temperature fluctuations of water also play their role in susceptibility to infection [12]. Further studies are necessary to investigate the etiology and to assess the impact of cutaneous nodules on the long-term health of affected S. c. taiwanensis. The orange films were a direct manifestation of a diatom, which is true in the case of other cetacean species [40]. Diatom attachment, by itself, is not a skin ailment that is an infectious condition [11], but due to the accumulation of diatoms there is slower skin regeneration following any insult and there is also a compromised swimming speed that negatively influences the physical condition and performance of the affected animal. This is an important factor in the decline of this critically endangered population. Pale lesions are either circular or amorphous, with rounded edges and a white-to-matte appearance, which indicates viral infection or inflammation. A black patch is similar in appearance to the poxvirus infection lesions described in previous studies [13,15]. Whether pale or black patches are significant to dolphin health, the current study showed that these two lesions were found much less (one individual for each) compared to the previous study conducted between 2006 and 2010 (10 and 3 individuals, respectively) [18].
A narrow linear mark might barely pose a threat to a dolphins' health and survival, although it still indicates dolphin-fishery interactions. In contrast, the other marks, such as mutilation, wide-spaced linear marks, and V-shaped indentations, have been associated with deep injuries attributed to human interactions and with different degrees of severity [24]. When excluding narrow-spaced linear marks, the prevalence of the severe injuries was 38% (19/50) in 2018, 39% (11/28) in 2019, and 38% (13/34) in 2021, and the prevalence of injuries of the 57 distinctive individuals in each coloration stage was calf (0%, 0/10), mottled stage (47%, 8/17), speckled stage (56%, 9/16), and adult stage (79%, 11/14). The prevalence in mottled, speckled, and adult individuals was comparable to the previous study on this population [20] and higher than other previous reports for small cetaceans [24,41,42]. Dolphins exhibit an extraordinary capacity to heal deep soft-tissue injuries such as shark bites and anthropogenic trauma, and the deep wounds in the dolphins may heal in a regenerative manner rather than repair [43]. However, these scars significantly alter the physiology and behavior of the dolphins and occasionally lead to death [42]. For example, a juvenile male S. c. taiwanensis was found stranded in south Taiwan on 21 January 2022, with injuries caused by past gillnet entanglement. The necropsy report showed that the acute bacterial infection and severe blunt force trauma were the direct causes of death, while the past gillnet entanglement injuries may be the contributory cause (WCY, personal communication). Furthermore, dolphins caught in driftnets may sustain injuries inflicted by fishermen, such as deep indentations or mutilations of stuck appendices (reviewed in [10]). This results in serious infections that may reduce swimming activities, causing higher energy expenditure and starvation and leading to death. Previous studies on T. truncatus and other species demonstrated a relationship between the incidence of injury and fishery activities and that the occurrence of skin injuries proportionally increased with higher fishery interactions [24,42]. The prevalence of deep injuries, which can be expected to reduce health and survivorship, in the current study was similar or lower than those in previous studies [20,44], although there were some differences in the survey effort, survey area, and injury classification method. In addition, no new deep injuries were found in the 17 individuals from 2018 to 2021. However, there are potential under-representations of the impacts since fatal injuries could not be observed in this study. For example, the severe injury classified as wide-spaced linear marks, which are likely the result of a turning propeller hit or net entanglement, were not observed in 2021. Furthermore, the preliminary investigation (data not shown) indicated that this population was at risk of predation from sharks, and it warrants further study on the correlation between anthropogenic injuries and shark bites. Our results can serve as a baseline for future analyses of injuries among the whole population. Further research should document the statistics of fatal injuries among stranded individuals and determine to what level the injuries can influence the health and survival of this critically endangered population.

Conclusions
It has been reported that the cetaceans habituating the Taiwanese waters may be affected by immunosuppression resulting from multiple anthropogenic stress factors, including fishery bycatch, the vessel movement rate, whale-watching activities, water and vessel noise, and land-based pollution [45]. The moderate-to-high prevalence of skin marks in this study is designated as a warning of the risks. Our results offer important recommendations, and further studies are warranted to examine if the situation is also found in the whole population of S. c. taiwanensis. Apart from our current efforts, consistently collecting photos and photo-ID data from free-ranging cetaceans would be highly recommended for other countries along the western Pacific region. Further studies investigating the skin marks of free-ranging cetaceans will help us to understand the crucial and potential threats to cetaceans, the ocean, and human health.  Institutional Review Board Statement: This research was conducted through a boat-based photo-ID survey, causing no harm or suffering to the animals. Therefore, there was no permit or ethics approval requested by the government.

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.