Abstract:We present a novel method for the non‐invasive and automatic assessment of harbour seal body condition, using unoccupied aerial vehicles (UAVs/drones), and systematically apply the method to wild seals over 2 years, achieving a near‐population‐level assessment for pups on land (82.5% measured). With reference to previous mark recapture data, we estimate mean expected probability of over‐winter survival. This work marks a significant step forward for the non‐invasive assessment of body condition in pinnipeds and could provide daily estimates of body mass for thousands of individuals, acting as an early warning for deteriorating environmental conditions. Body mass is a fundamental indicator of animal health closely linked to survival and reproductive success. Systematic assessment of body mass for a large proportion of a population can allow early detection of changes likely to impact population growth, facilitating responsive management and a mechanistic understanding of ecological trends. One challenge with integrating body mass assessment into monitoring is sampling enough animals to detect trends and account for individual variation. Harbour seals (Phoca vitulina) are philopatric marine mammals responsive to regional environmental changes, resulting in their use as an indicator species. We present a novel method for the non‐invasive and semi‐automatic assessment of harbour seal body condition, using unoccupied aerial vehicles (UAVs/drones). Morphological parameters are automatically measured in georeferenced images and used to estimate volume, which is then translated to estimated mass. Remote observations of known individuals are utilized to calibrate the method. We achieve a high level of accuracy (mean absolute error of 4.5 kg or 10.5% for all seals and 3.2 kg or 12.7% for pups‐of‐the‐year). We systematically apply the method to wild seals during the Spring pupping season and Autumn over 2 years, achieving a near‐population‐level assessment for pups on land (82.5% measured). With reference to previous mark‐recapture work linking Autumn pup weights to survival, we estimate mean expected probability of over‐winter survival (mean = 0.89, standard deviation = 0.08). This work marks a significant step forward for the non‐invasive assessment of body condition in pinnipeds and could provide daily estimates of body mass for thousands of individuals. It can act as an early warning for deteriorating environmental conditions and be utilized as an integrative tool for wildlife monitoring. It also enables estimation of yearly variation in demographic rates which can be utilized in parameterizing models of population growth with relevance for conservation and evolutionary biology.

Evaluating the performance of semiautomated photographic identification programs for leopard seals

Image-Based Leopard Seal Recognition: Approaches and Challenges in Current Automated Systems

Comparison of Two Individual Identification Algorithms for Snow Leopards after Automated Detection

Optimizing the automated recognition of individual animals to support population monitoring

Species classification of Antarctic pack‐ice seals using very high‐resolution imagery

Investigating the viability of photo-identification as an objective tool to study endangered sea turtle populations

Where to spot: individual identification of leopard cats (Prionailurus bengalensis euptilurus) in South Korea

The effectiveness of the semi‐automated Identifin software to accurately fast‐track dorsal fin photographic identifications of sharks

Automatic Identification of Individual African Leopards in Unlabeled Camera Trap Images

Performance of different automatic photographic identification software for larvae and adults of the European fire salamander

SealID: Saimaa Ringed Seal Re-Identification Dataset

Similarity learning networks uniquely identify individuals of four marine and terrestrial species

Multispecies facial detection for individual identification of wildlife: a case study across ursids

Occurrence, residency, and habitat characterization of leopard seals in Chile

Approaching a population‐level assessment of body size in pinnipeds using drones, an early warning of environmental degradation

Site fidelity of harbor seals in Casco Bay, ME, USA using facial recognition technology: a pilot study

Combining feature aggregation and geometric similarity for re-identification of patterned animals

Semi-automated detection of ungulates using UAV imagery and reflective spectrometry

NORPPA: NOvel Ringed seal re-identification by Pelage Pattern Aggregation

Exploiting facial side similarities to improve AI-driven sea turtle photo-identification systems

Machine learning to detect marine animals in UAV imagery: effect of morphology, spacing, behaviour and habitat