Mapping threatened Thai bovids provides opportunities for improved conservation outcomes in Asia

Wild bovids provide important ecosystem functions throughout their ranges. Five wild bovids remain in Thailand: gaur (Bos gaurus), banteng (Bos javanicus), wild water buffalo (Bubalus arnee), mainland serow (Capricornis sumatraensis) and Chinese goral (Naemorhedus griseus). However, their populations and habitats have declined substantially and become fragmented. Here, we identify potentially suitable habitat for these threatened bovids using ecological niche models and quantify how much suitable area remains within protected areas. We combined species occurrence data with environmental variables and used spatially-restricted Biotic-Abiotic-Mobility frameworks with species-specific and single large accessible areas. We used ensembles from eight algorithms for generating maps and out-of-sample predictions to validate model performance against new data. Gaur, banteng, and buffalo models performed well throughout the entire distribution (≥62%) and in Thailand (≥80%). Mainland serow and Chinese goral performed poorly for the entire distribution and in Thailand, though a 5 km movement buffer markedly improved model performance for serow. Particularly large suitable areas were in Thailand and India for gaur, Cambodia and Thailand for banteng, and India for buffalo. Over 50% of overall suitable habitat is located outside protected areas, with just 9% for buffalo in Thai protected areas, highlighting area for potential habitat management and conflict mitigation.

Methods
We collected the raw species occurrence data from several different sources. The data were collected using various field study methodologies, including observation of animal signs (e.g. footprint and dung) during forest patrols, direct observation during wildlife surveys, camera trapping, and radio-collar signals. The data coverage by country can be found in Table 1 in the main text, and for details of the data collection sites, see the supplementary materials Table S2 and S10.

We built ecological niche models (using ENMTML package in R) to identify habitat suitability and calculate how much potential suitable habitat for these species remains across their species distribution. The R code and map results are provided.