Individual identification and confirmation of nest site fidelity in painted stork ( <i>Mycteria leucocephala</i> ) using deep transfer learning

Abstract Accurate individual identification is vital in field studies. Since traditional marking techniques, though effective, can be intrusive and obtaining permits can be difficult, identification using natural markings has gained popularity across various taxa as a non-invasive alternative. Here, we report on a painted stork (Mycteria leucocephala) with a distinctive neck injury mark, observed at the Delhi Zoo, India, over four consecutive breeding seasons (2022–2025). To verify its identity and assess nest-site fidelity, we employed a non-invasive approach that combined traditional (morphometrics, scale-invariant feature transform (SIFT)) and advanced techniques (deep transfer learning (DTL)). High-resolution photographs were used to extract linear measurements and assess the repeatability of features by SIFT, while a DTL classifier further validated the individual’s identity with 98% accuracy, confirming the classification decision through an explainable artificial intelligence technique (Gradient-weighted Class Activation Mapping). Image-based morphometric measurements were particularly reliable for longer morphological features and classification based on wing feathers’ colour patterns, confirming that the scar-marked stork observed over four consecutive years is indeed the same individual. The repeated sightings of the scar-marked stork on the same site support evidence of nest-site fidelity. Our findings highlight the potential of pattern-based recognition and DTL as powerful, non-intrusive tools for long-term monitoring of colonial waterbirds.