Metabarcoding reveals diet diversity in an ungulate community in Thailand

The diverse large mammal communities found in Asian dry forests and savannas should segregate based on their diet selection. We examined the diet composition of sympatric ungulate species using metabarcoding to determine whether their diet was segregated and whether obvious attributes (i.e., body size, phylogeny, ecology) explained the structure. We collected fecal samples from eight ungulate species in Huai Kha Khaeng Wildlife Sanctuary in the western forest complex of Thailand. The fecal collections occurred around a plot where all woody species were codified within a genetic barcode library, and this library was supplemented with samples from plant species known to be consumed by these species. Of 273 plant species tested, at least 93 were found within the fecal samples. Over half of the identified species were not previously known by experts as forage species. All ungulate species showed a strong consumption of grasses and forbs. For the three species with sufficient sample size (sambar, banteng, and guar), there were seasonal differences in their diet, with each showing increased occurrence of woody plants during the dry season. The pattern of forage consumption did not follow obvious paradigms of body size or taxonomy, with significant diet differences found in two similar‐sized bovids (gaur, banteng), while the diet of sambar was more similar to bovids than to the other deer species. Asian ungulates differ in their forage consumption and metabarcoding should allow for testing of diet shifts in response to seasonal rains and fires which dominate the phenology of Asian dry forests and savannas.