Mapping black panthers: Macroecological modeling of melanism in leopards (Panthera pardus)

• Published in: PloS one Vol. 12; no. 4; p. e0170378
• Main Authors: da Silva, Lucas G, Kawanishi, Kae, Henschel, Philipp, Kittle, Andrew, Sanei, Arezoo, Reebin, Alexander, Miquelle, Dale, Stein, Andrew B, Watson, Anjali, Kekule, Laurence Bruce, Machado, Ricardo B, Eizirik, Eduardo
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 05.04.2017; Public Library of Science (PLoS)
• Subjects: Adaptiveness; Algorithms; Analysis; Animal Distribution - physiology; Animals; Biogeography; Biological Evolution; Biology and Life Sciences; Camouflage; Color; Coloration; Conservation of Natural Resources - methods; Ecology; Ecology and Environmental Sciences; Ecosystem; Endangered & extinct species; Genotype & phenotype; Geographical distribution; Habitats; Hypotheses; Leopards; Macroecology; Mapping; Melanism; Melanism (Evolutionary adaptation); Modelling; Natural selection; New records; Panthera - physiology; Panthera pardus; People and Places; Phenotype; Phenotypes; Phylogeography; Physiological aspects; Selection, Genetic - physiology; Spatial distribution; Studies; Thermoregulation; Vegetation type; Wildlife conservation
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0170378

The geographic distribution and habitat association of most mammalian polymorphic phenotypes are still poorly known, hampering assessments of their adaptive significance. Even in the case of the black panther, an iconic melanistic variant of the leopard (Panthera pardus), no map exists describing its distribution. We constructed a large database of verified records sampled across the species' range, and used it to map the geographic occurrence of melanism. We then estimated the potential distribution of melanistic and non-melanistic leopards using niche-modeling algorithms. The overall frequency of melanism was ca. 11%, with a significantly non-random spatial distribution. Distinct habitat types presented significantly different frequencies of melanism, which increased in Asian moist forests and approached zero across most open/dry biomes. Niche modeling indicated that the potential distributions of the two phenotypes were distinct, with significant differences in habitat suitability and rejection of niche equivalency between them. We conclude that melanism in leopards is strongly affected by natural selection, likely driven by efficacy of camouflage and/or thermoregulation in different habitats, along with an effect of moisture that goes beyond its influence on vegetation type. Our results support classical hypotheses of adaptive coloration in animals (e.g. Gloger's rule), and open up new avenues for in-depth evolutionary analyses of melanism in mammals.