Ecological overlap of sympatric sun bears and Asiatic black bears in tropical forest, Thailand

Asiatic black bears (Ursus thibetanus) and sun bears (Helarctos malayanus) are ecologically similar species with broadly overlapping distributions in mainland Southeast Asia. Using bear sign distinguished to species, I investigated habitat selection and feeding ecology of these bear species in 3 habitat types in Thung Yai Naresuan Wildlife Sanctuary, Thailand.;I obtained imprints of claw marks on trees climbed by captive Asiatic black bears and sun bears (n = 43, all sex/ages). Discriminant function analysis was used to develop criteria for distinguishing the 2 species based on spacing of claw marks. I then applied these criteria to differentiate species-specific tree climbing in the field. Adult black bears made wider claw mark prints than adult sun bears. Based on three width measurements, captive individuals could be separated to species with 95% accuracy. Similar accuracy was achieved on independent claw mark samples from Borneo (occupied only by sun bears) and China (where only black bears occur). However, black bear cubs <1 year old could not be distinguished from sun bear adults.;Strip transects (n = 71, total area = 31.2 ha) were distributed among semi-evergreen forest and mixed deciduous forest (both <1200m), and montane evergreen forest (>1200m). Every tree within transects was examined for claw marks, and other feeding signs were recorded. Claw marks were measured, and climbed trees were identified. Ages of claw marks were determined experimentally, as recent (<3 months), old (3-10 months), or very old (>10 months). Vegetation plots were established to quantify the density of fruit-tree species that bears climbed.;Claw marks on trees comprised >80% of bear signs in each habitat, indicating bears fed mostly on fruit. Insect feeding signs were most common in mixed deciduous forest, where tree density was lowest. Overall sign density (both bear sp.) was 3 times higher in evergreen forest types (34-38 signs/ha) as in deciduous forest (12 signs/ha). Habitat partitioning was minimal, occurring in 1 of 3 habitats. Both bear species had similar sign densities (1-3 fresh signs/ha) in lowland habitats, but in montane forest black bears predominated (14 signs/ha), and sun bear sign was scarce despite abundant fruit.;Sun bears and black bears were primarily frugivorous throughout the year and insects were of secondary importance. Signs of insect-feeding were mostly from sun bear. Of the 10 habitat variables I examined, including interspecific activity, only fruiting tree density was related to presence of feeding signs, for both bear species. I recorded evidence of feeding on 94 fleshy-fruited plant species. Use of available fruit trees by bears was disproportionate to tree abundance, suggesting selection for a mixed diet of different fruits. Niche overlap on fruit species was high (Pianka's index >0.75), and this overlap did not decrease during periods of fruit scarcity; thus, bears did not partition fruit resources.;One habitat (montane forest) was used almost exclusively by black bears, and one resource (insects) was used almost exclusively by sun bears. Such exclusivity may enable each species to maintain sufficient densities, and hence coexistence, despite strong shared preferences for most other habitats and resources.