Rapid development of individual identification and presence systems for a critically endangered antelope, the Mountain bongo

Abstract

AbstractMonitoring of species, particularly remnant populations requiring urgent conservation is often hampered by the lack of reliable tools for individual identification (using images or their spoor). Here, we develop rapid monitoring tools for individual animals of the Mountain bongo (Tragelaphus eurycerus isaaci), a critically endangered subspecies of the bongo only found in Kenya. We developed and tested an individual identification system using camera trap footage, as well as a quantitative tool to identify bongo spoor in the field, both useable by naïve observers. We implemented an information content approach to assess the importance of different visual elements in 61 individual bongos to optimise our identification system. We tested the reliability of the system with 15 naïve observers. We conclude that an optimal identification system should rely on three main visual features (stripe pattern, facial markings and horns appearance). We show that reliability amongst observers is high (κ = 0.64). We also developed a field scheme to identify footprint and spoor sign. Measurements of bongo footprints were compared with those of waterbuck (Kobus ellypsiprimnus), a syntopic antelope. Confusion occurs between spoor and footprints of both species. We find that differences in the aspect ratio of bongo and waterbuck footprints can identify the two species, 1.22 (±0.08) for bongo and 1.49 (±0.10) for waterbuck. The acquisition of reliable tools ensures monitoring activities are less dependent on individual expertise, which will allow consistent monitoring of bongo remnant populations in the future. The methods we used to develop these monitoring tools can help managers and field workers in the study of this and similar rare species where monitoring is a challenge.