EFFECTIVENESS AND RELIABILITY OF PHOTOGRAPHIC IDENTIFICATION METHODS FOR IDENTIFYING INDIVIDUALS OF A CRYPTICALLY PATTERNED TOAD

Various marking methods are used for amphibian population studies, some of which have been debated tier their invasive nature. Although non-invasive methods exist, they are often employed on vibrant species with distinct coloration and patterning. Here, we evaluate the Photographic Identification Method (PIM) as a marking method for identifying individuals of the Schneider's Toad (Rhinelki diptycha), which has cryptic dorsal patterning, and compare two PIM techniques: visual and computer assisted. We surveyed toads in Pilar, Paraguay, photographed dorsal patterns of each captured toad in situ and marked individuals with Visible Implant Elastomer (VIE) between their inter-digital webbing for cross-validation of recaptured individuals. Of 109 captured toads, we used 37 recaptures to test the accuracy of PIM and VIE methods. Volunteers matched photographs of recaptured toads with a set of all first-capture photographs to test visual PIM, and we recorded their accuracy and the time to reach a match decision. To test computer-assisted PIM, we used the program Wild-ID, which matched recapture and first-capture photographs. We cross-referenced suggested match pairs using photograph codes corresponding to individual specimens to test accuracy and recorded the time to reach a match decision. Computer-assisted PIM was the most accurate (100%) and fastest method (on average, 11.5 x faster than visual PIM), but visual PIM (86.7 degrees M and VIE (89.2%) were also highly accurate for identifying individual toads. Despite the cryptic pattern of these toads, our results suggest that non-invasive and cost-effective methods such as PIM can be used for population studies of dull-patterned taxa.