Abstract
This study developed and evaluated a visual health assessment (VHA) tool to preliminarily assess skin anomalies in Tamanend’s bottlenose dolphins (Tursiops erebennus) using photo-identification data collected during a 2022 capture-mark-recapture (CMR) study in Manteo, North Carolina. Traditional hands-on dolphin health assessments, while comprehensive, are limited in scale and may impose stress on animals. To address this, a non-invasive, photo-based method was designed to detect external signs of morbidity such as lesions, scars, and organism attachment. The VHA tool standardized evaluation across 20 pre-defined body segments and incorporated environmental metadata (e.g., salinity, temperature, depth) to contextualize skin conditions. Photographs of 242 individuals were reviewed, with 167 resighted dolphins meeting inclusion criteria. Of these, 91 (54.5%) exhibited presumptive indicators of freshwater disease (FWD), most frequently light skin discoloration. Skin anomalies due to wounds or scars were widespread (95.8%), often attributable to conspecific interaction. Predictive modeling revealed an inverse relationship between the likelihood of FWD indicators and salinity, temperature, and depth, with a significant interaction between salinity and temperature. Despite inter-rater variability in lesion classification—likely due to inconsistent photo quality and lack of full-body visibility—the tool proved functional for observers with limited prior experience. These findings highlight the utility of integrating visual health assessment tools with photo-ID surveys to monitor individual and population health non-invasively, especially in estuarine habitats where environmental stressors like freshwater influx may be increasing due to climate change and human activity. Future efforts should focus on improving image acquisition protocols, validating the tool's reliability with expert reviewers, and tracking individuals longitudinally to assess health progression over time.