Individual identification of endangered amphibians using deep learning and smartphone images: case study of the Japanese giant salamander (Andrias japonicus).

Information obtained via individual identification is invaluable for ecology and conservation. Physical tags, such as PIT tags and GPS, have been used for individual identification; however, these methods could impact on animal behavior and survival rates, and the tags may become lost. Although non-invasive methods that do not affect the target species (such as manual photoidentification) are available, these techniques utilize stripes and spots that are unique to the individual, which requires training, and applying them to large datasets is challenging. Many studies that have applied deep learning for identification have focused on species-level identification, but few have addressed individual-level identification. In this study, we developed an image-based identification method based on deep learning that uses the head spot pattern of the Japanese giant salamander (Andrias japonicus), an endemic and endangered species in Japan. We trained and evaluated a dataset collected over two days from 11 individuals in captivity, which included 7075 images taken by a smartphone camera. Individuals were photographed three times a day at approximately 11:00 (morning), 15:00 (afternoon), and 18:00 (evening). As a result, individual identification by our method, which used the EfficientNetV2 achieved 99.86% accuracy, kappa coefficient of 0.99, and an F1 score of 0.99. Performance was lower for the evening  model than for the morning and afternoon models, which were trained and evaluated using photographs taken at the corresponding time of the day. The proposed method does not require direct contact with the target species, and the effect on the animals is minimal; moreover, individual-level information can be obtained under natural conditions. In the future, smartphone images can be applied to citizen science surveys and individual-level big data collection, which is difficult using current methods.

Effects of sensor ear tags with twin pin fixing system on health and well-being of cattle.

INTRODUCTION: A sensor ear tag (SET) containing Global Positioning System (GPS), accelerometer, Radio-Frequency Identification (RFID), and Bluetooth technologies was tested for wearing comfort and compliance with animal welfare requirements in cattle in a free stall barn and on summer pasture in Switzerland. The SET was equipped with a long-lasting battery via solar panel and used a «twin pin¼ fixing system. Right ears of 12 newborns and 26 adolescent animals were tagged with the SET. While left ears were tagged simultaneously with official ear tags in newborns, the adolescent animals already carried the official ear tags. The newborns stayed in a free stall barn during the entire experiment, while adolescent animals were housed in a free stall barn and on pasture during summer. All animals developed crusts beginning on day 7 after tagging with the SET. Pain reactions were observed occasionally in the first two weeks. Ear growth in newborns during 11 months of observation did not differ between ears with SET and official ear tags. Cortisol concentration in saliva of newborns decreased in the first week after tagging which is physiological for this age group. In older animals cortisol concentrations in saliva were not affected. We registered 19 incidences in 11 animals with the SET, that required veterinary or staff intervention. Two animals lost the SET with ear injury. Scars due to tag migration were observed in ears of all newborns after the 9th month of observation. In conclusion, SET with a weight of 32 g that need a twin pin fixation in cows do not seem to induce systemic or local inflammations more frequently compared to official ear tags; however, the higher risk of accidental injuries and migration in ear cartilage would not meet Swiss welfare standards and the attachment to the ear needs to be improved for general use.

INTRODUCTION: Une marque auriculaire à capteur (SET) contenant les technologies Global Positioning System (GPS), accéléromètre, identification par radiofréquence (RFID) et Bluetooth a été testée en termes de confort et de conformité aux exigences de bien-être animal chez des bovins dans une étable à stabulation libre et sur des pâturages d'estivage en Suisse. Le SET était équipé d'une batterie longue durée via un panneau solaire et utilisait un système de fixation «twin pin¼. Les oreilles droites de 12 veaux et de 26 génisses ont été équipées avec le SET. Les oreilles gauches ont été marquées en même temps avec les marques officielles chez les nouveau-nés alors que les génisses portaient déjà ces marques officielles. Les nouveau-nés ont été détenu dans une étable à stabulation libre, avec accès à une aire de sortie et aux paturages voisins pendant toute la durée de l'expérience tandis que les génisses ont été logés dans une étable à stabulation libre et en pâturage pendant l'été. Tous les animaux ont développé des croûtes à partir du 7e jour après le marquage avec le SET. Des réactions douloureuses ont occasionnellement été observées au cours des deux premières semaines. La croissance des oreilles des nouveau-nés au cours des 11 mois d'observation n'a pas différé entre les oreilles marquées par le SET et les oreilles marquées de manière standard. La concentration de cortisol dans la salive des nouveau-nés a diminué au cours de la première semaine successive au marquage, ce qui est physiologique pour ce groupe d'âge. Chez les animaux plus âgés, les concentrations de cortisol dans la salive n'ont pas été affectées. Nous avons enregistré 19 incidents chez 11 animaux avec le SET, qui ont nécessité l'intervention d'un vétérinaire ou d'un membre du personnel. Deux animaux ont perdu le SET avec blessure à l'oreille. Des cicatrices dues à la migration des marques ont été observées sur les oreilles de tous les nouveau-nés après le 9e mois d'observation. En conclusion, les SET d'un poids de 32 g qui nécessitent une fixation par deux tiges chez les bovins ne semblent pas induire d'inflammations systémiques ou locales plus fréquemment que les marques auriculaires officielles. Cependant le risque plus élevé de blessures accidentelles et de migration dans le cartilage de l'oreille ne correspondrait pas aux normes suisses en matière de bien-être et la fixation à l'oreille doit être améliorée pour une utilisation généralisée.

Biologging and Biotelemetry: Tools for Understanding the Lives and Environments of Marine Animals.

Addressing important questions in animal ecology, physiology, and environmental science often requires in situ information from wild animals. This difficulty is being overcome by biologging and biotelemetry, or the use of miniaturized animal-borne sensors. Although early studies recorded only simple parameters of animal movement, advanced devices and analytical methods can now provide rich information on individual and group behavior, internal states, and the surrounding environment of free-ranging animals, especially those in marine systems. We summarize the history of technologies used to track marine animals. We then identify seven major research categories of marine biologging and biotelemetry and explain significant achievements, as well as future opportunities. Big data approaches via international collaborations will be key to tackling global environmental issues (e.g., climate change impacts), and curiosity about the secret lives of marine animals will also remain a major driver of biologging and biotelemetry studies.

Survival, growth and tag retention of juvenile European eel (Anguilla anguilla L.) with implanted 12 mm passive integrated transponder tags and acoustic tags.

To evaluate the efficiency of tagging juvenile European eels with implanted 12 mm passive integrated transponder (PIT) tags or Eel/Lamprey acoustic transmitters (ELATs), the authors studied tag retention, survival and growth of eels (7-25 g). Experimental eels were obtained from an eel farm, tagged and then released in a series of shallow dug-out ponds with a surface area of c. 200 m2 . Tagged and control eels were distributed evenly, with 50 tagged and 50 control eels in each of four ponds, giving a total of 200 tagged and 200 control eels mixed. After 76 days, the ponds were drained, and eels were sampled and measured. A total of 344 eels (86%) were recaptured, indicating high survival. Tag retention was 99% as only one of the recaptured PIT-tagged eels had lost the tag and none of the ELAT tagged. The results demonstrated that tagging juvenile eels >16 cm with these small tags is indeed feasible. The growth of tagged and control fish was differentiated but generally low in length and negative in mass but did not differ between the three groups.

Wound lesions caused by ear tagging in unweaned calves: assessing the prevalence of wound lesions and identifying risk factors.

Identification of cattle by ear tagging is legally required to ensure traceability. However, studies indicate that ear tagging causes pain-associated physiological and behavioural responses. The wound healing process and prevalence of wound lesions in calves remain mostly unknown. Therefore, this study sought to estimate the prevalence of wound lesions and identify associated risk factors by assessing ear tagging management in unweaned dairy calves. We conducted one field study with single visits to estimate the prevalence of wound lesions and associated risk factors (Study 1, 42 farms, 802 calves) and one follow-up study with repeated visits to assess farmers' view on ear tag management, the relationship between calf health and wound healing, and the development of wound lesions over time (Study 2, five farms, 42 calves). Study 1 comprised a short interview with the farmer (four questions regarding ear tagging). Ear tag position (on or between ridges) and wound lesions were evaluated using a three-level scoring system (1 = no blood, scab, or pus discharge; 2 = incrustation or scab and slight blood or pus discharge; and 3 = heavy purulent discharge, tissue deformation, or both). In Study 2, farmers were interviewed about ear tagging (30 questions), and 10 calves from each farm were assessed on the day of ear tagging and 1, 3, and 6 weeks after tag insertion. Calf health, ear tag position, and wound characteristics were assessed during all visits. Both studies were analysed descriptively, and odds ratios (ORs) for wound lesions in Study 1 were calculated using logistic regression. Of the ears assessed in Study 1, 31.1% showed clinical signs classified as category 2. Score 3 was less common and was found for 6.7% of all ears. Although the highest incidence of wound lesions was found in calves aged 2-4 weeks, wound lesions were also found in calves aged >10 weeks (18.5%). Identified risk factors for wound lesions were small farm size, calf age, single housing, group size, placement of ear tags on ridges, and other ear's score. Individual farmers in Study 2 were able to place ear tags very accurately, although awareness about ear tag lesions appeared to be low among farmers. Sensitising farmers to this issue, implementing routine check-ups of ear tag wounds 2 weeks after insertion, and considering the identified risk factors may reduce animal welfare impairments associated with ear tagging.

Deep learning increases the availability of organism photographs taken by citizens in citizen science programs.

Citizen science programs using organism photographs have become popular, but there are two problems related to photographs. One problem is the low quality of photographs. It is laborious to identify species in photographs taken outdoors because they are out of focus, partially invisible, or under different lighting conditions. The other is difficulty for non-experts to identify species. Organisms usually have interspecific similarity and intraspecific variation, which hinder species identification by non-experts. Deep learning solves these problems and increases the availability of organism photographs. We trained a deep convolutional neural network, Xception, to identify bee species using various quality of bee photographs that were taken by citizens. These bees belonged to two honey bee species and 10 bumble bee species with interspecific similarity and intraspecific variation. We investigated the accuracy of species identification by biologists and deep learning. The accuracy of species identification by Xception (83.4%) was much higher than that of biologists (53.7%). When we grouped bee photographs by different colors resulting from intraspecific variation in addition to species, the accuracy of species identification by Xception increased to 84.7%. The collaboration with deep learning and experts will increase the reliability of species identification and their use for scientific researches.

Observations on the use of passive integrated transponder (PIT) tags in teleosts and elasmobranchs at a public aquarium, 728 cases, 2007-2020.

Between 2007 and 2020 at New England Aquarium, Boston, MA, USA, we implanted passive integrated transponder (PIT) tags into 728 fish representing 105 teleost and elasmobranch species to identify animals as individuals. At the time of retrospective data analysis, mean longevity interval (median, range) after tag placement for animals that remained alive (n = 236) was 4.7 years (4.5, 0.3-13.8). Mean interval (median, range) between tag placement and death (n = 317) was 2.1 years (1.6, 0-11.2); and mean interval (median, range) between tag placement and transfer to other facilities (n = 175) was 2.5 years (3.1, 0.1-9.3). Possible adverse effects of tagging were extremely rare. Using the described methods, the equipment cost for every 10 PIT tag implantations was $2.83. PIT tag implantation in fishes is a safe and cost effective method to identify individuals, providing an opportunity to accumulate valuable data regarding individual longevity, welfare, basic demographics, and outcome of medical management. PIT tag implantation is recommended as a routine aspect of acquisition, quarantine, and medical management of fish under human care.

Impact evaluation of marking Salmo trutta with Alizarin Red S produced by different manufacturers.

Fish otolith marking with the alizarin dye is a commonly used tool in sustainable fishery management. However, the reported effects of this dye on fish health are rather controversial and are possibly linked to differences in the composition of different brands of Alizarin red S (ARS). Laboratory experiments designed to elucidate effects of different concentrations of theoretically the same ARS as indicated by the CAS (Chemical Abstracts Service) number on fish at different development stages were carried out. The acute toxicity of ARS to Salmo trutta was found to be concentration- and fish developmental stage-dependant. Our study results showed that S. trutta sensitivity to ARS varies depending on its developmental stages as follows: fry (50-days after hatching) > alevins (30-days after hatching) > alevins (1-day after hatching). One of the tested ARS brands (purchased from VWR International LLC (Matsonford Road, USA)) was found to be several times more toxic to fish than another (purchased from Sigma-Aldrich (St. Louis, USA)), although according to the certificates of analysis, the tested substances were identical. Survival and growth of the S. trutta fry, which was marked with different ARS brands and stocked in the same natural stream, was investigated for two consecutive years. The results obtained indicate remarkable differences (p < 0.05) in the effects produced by the tested ARS brands, thus confirming our laboratory findings. The performed elemental analysis of the tested ARS dyes revealed significant differences in chemical impurities that these dyes contain. This study has, for the first time, expressed concern about the probable long-term impact of some ARS brands on the marked fish and their potential to bias the results of the studies dealing with ARS-marked fish.

Use of a Fluorophore to Tag Arthropods for Mark-Release-Recapture Type Research.

We examined the feasibility of externally marking insects with the liquid fluorescent forensic theft deterrent, SmartWater (SmartWater CSI, LLC.). We sprayed captive Lygus hesperus (Knight) (Hemiptera: Miridae), Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae), and Hippodamia convergens Guérin-Méneville (Coleoptera: Coccinellidae) with SmartWater fluorophore, and then qualitatively examined them for fluorescence by visual inspection under ultraviolet (UV) light and quantitatively measured them with a multiwavelength microplate fluorometer. The results indicate that this product has enormous potential as a taggant for L. hesperus and B. tabaci. However, the marking efficiency for H. convergens was only adequate. The advantages and limitations of using SmartWater as a biological marker for arthropod mark-release-recapture research are discussed.

Behavioral Effects and Retention of Protein Immunomarkers on Plum Curculio Conotrachelus nenuphar (Coleoptera: Curculionidae).

Protein immunomarking can be used to track the dispersal of insects in the field or identify plant-insect interactions. By marking insects with known proteins and recapturing them, their movement or host use can be quantified with Enzyme Linked Immunosorbent Assay (ELISA). Before using this technique, retention and behavioral effects of these markers should be evaluated to ensure that the insect's natural behaviors are conserved. Here, we tested the effects of protein markers on the plum curculio, Conotrachelus nenuphar (Coleoptera: Curculionidae) using two different application methods. This weevil is native to North American and a pest of tree fruit and blueberry in the United States and causes damage resulting in near complete crop loss if left untreated. We tested the effects of marking adult C. nenuphar with two inexpensive food-based immunoprotein markers, bovine casein (cow's milk) and chicken albumin (egg whites) on climbing distance (total cm), lateral movement (total cm), and lateral movement speed (cm/s), as well as retention time of protein immunomarkers. Neither protein immunomarker affected C. nenuphar movement or climbing, although females climbed significantly greater distances than males. ELISA assays detected 37.5-56.2% of milk protein and 56.2-59.3% of egg on the insect 7 d after application depending on application method. Our findings indicate that food-based protein immunomarkers can be used in future studies to test C. nenuphar movement within host plants without impacting behavior. The use of protein immunomarking will allow studies that will lead to behaviorally based management tactics.

Development and validation of a 1 K sika deer (Cervus nippon) SNP Chip.

BACKGROUND: China is the birthplace of the deer family and the country with the most abundant deer resources. However, at present, China's deer industry faces the problem that pure sika deer and hybrid deer cannot be easily distinguished. Therefore, the development of a SNP identification chip is urgently required. RESULTS: In this study, 250 sika deer, 206 red deer, 23 first-generation hybrid deer (F1), 20 s-generation hybrid deer (F2), and 20 third-generation hybrid deer (F3) were resequenced. Using the chromosome-level sika deer genome as the reference sequence, mutation detection was performed on all individuals, and a total of 130,306,923 SNP loci were generated. After quality control filtering was performed, the remaining 31,140,900 loci were confirmed. From molecular-level and morphological analyses, the sika deer reference population and the red deer reference population were established. The Fst values of all SNPs in the two reference populations were calculated. According to customized algorithms and strict screening principles, 1000 red deer-specific SNP sites were finally selected for chip design, and 63 hybrid individuals were determined to contain red deer-specific SNP loci. The results showed that the gene content of red deer gradually decreased in subsequent hybrid generations, and this decrease roughly conformed to the law of statistical genetics. Reaction probes were designed according to the screening sites. All candidate sites met the requirements of the Illumina chip scoring system. The average score was 0.99, and the MAF was in the range of 0.3277 to 0.3621. Furthermore, 266 deer (125 sika deer, 39 red deer, 56 F1, 29 F2,17 F3) were randomly selected for 1 K SNP chip verification. The results showed that among the 1000 SNP sites, 995 probes were synthesized, 4 of which could not be typed, while 973 loci were polymorphic. PCA, random forest and ADMIXTURE results showed that the 1 K sika deer SNP chip was able to clearly distinguish sika deer, red deer, and hybrid deer and that this 1 K SNP chip technology may provide technical support for the protection and utilization of pure sika deer species resources. CONCLUSION: We successfully developed a low-density identification chip that can quickly and accurately distinguish sika deer from their hybrid offspring, thereby providing technical support for the protection and utilization of pure sika deer germplasm resources.

Mark-release-recapture of male Aedes aegypti (Diptera: Culicidae): Use of rhodamine B to estimate movement, mating and population parameters in preparation for an incompatible male program.

Rapid advances in biological and digital support systems are revolutionizing the population control of invasive disease vectors such as Aedes aegypti. Methods such as the sterile and incompatible insect techniques (SIT/IIT) rely on modified males to seek out and successfully mate with females, and in doing so outcompete the wild male population for mates. Currently, these interventions most frequently infer mating success through area-wide population surveillance and estimates of mating competitiveness are rare. Furthermore, little is known about male Ae. aegypti behaviour and biology in field settings. In preparation for a large, community scale IIT program, we undertook a series of mark- release-recapture experiments using rhodamine B to mark male Ae. aegypti sperm and measure mating interactions with females. We also developed a Spatial and Temporally Evolving Isotropic Kernel (STEIK) framework to assist researchers to estimate the movement of individuals through space and time. Results showed that ~40% of wild females captured daily were unmated, suggesting interventions will need to release males multiple times per week to be effective at suppressing Ae. aegypti populations. Males moved rapidly through the landscape, particularly when released during the night. Although males moved further than what is typically observed in females of the species, survival was considerably lower. These unique insights improve our understanding of mating interactions in wild Ae. aegypti populations and lay the foundation for robust suppression strategies in the future.