Aim: Deforestation is rapidly altering Southeast Asian landscapes, resulting in some of the highest rates of habitat loss worldwide. Among the many species facing declines in this region, clouded leopards rank notably for their ambassadorial potential 1640 | MACDONALD et AL.

Rates of biodiversity loss in Southeast Asia are among the highest in the world, and the Indo-Burma and South-Central China Biodiversity Hotspots rank among the world’s most threatened. Developing robust multi-species conservation models is critical for stemming biodiversity loss both here and globally. We used a large and geographically extensive remote-camera survey and multi-scale, multivariate optimization species distribution modelling to investigate the factors driving biodiversity across these two adjoining biodiversity hotspots. Four major findings emerged from the work. (i) We identified clear spatial patterns of species richness, with two main biodiverse centres in the Thai-Malay Peninsula and in the mountainous region of Southwest China. (ii) Carnivores in particular, and large ungulates to a lesser degree, were the strongest indicators of species richness. (iii) Climate had the largest effect on biodiversity, followed by protected status and human footprint. (iv) Gap analysis between the biodiversity model and the current system of protected areas revealed that the majority of areas supporting the highest predicted biodiversity are not protected. Our results highlighted several key locations that should be prioritized for expanding the protected area network to maximize conservation effectiveness. We demonstrated the importance of switching from single-species to multi-species approaches to highlight areas of high priority for biodiversity conservation. In addition, since these areas mostly occur over multiple countries, we also advocate for a paradigmatic focus on transboundary conservation planning.

Conservation of Indochinese primates is hampered by a lack of knowledge of species ecology, habitat preferences and, locally, distribution. Predictive distribution and habitat suitability models, using predictors known to affect the distribution of similar species elsewhere, may, therefore be of great benefit to conservationists within the region. Yellow-cheeked crested gibbon Nomascus gabriellae is an IUCN-listed endangered primate distributed east of the Mekong River in Cambodia, southern Vietnam, and possibly southern Lao PDR. Within Cambodia, yellow-cheeked crested gibbon are naturally restricted to evergreen forest fragments within a landscape matrix of deciduous dipterocarp and semievergreen forests. During the 2008 dry season, auditory surveys for yellow-cheeked crested gibbon were conducted within Phnom Prich Wildlife Sanctuary, Mondulkiri province, Cambodia. Predictive distribution models, in which variables were included at the scale at which they best explained gibbon occurrence (multi-grain models), were constructed to examine the species' habitat associations and tolerance of habitat fragmentation within the wildlife sanctuary. Gibbon occupancy ( j) was higher in evergreen (0.43 AE 0.26-0.62) than in semi-evergreen forest (0.21 AE 0.09-0.4), with gibbon presence constrained by a critical amount of evergreen forest within 5 km radius of listening posts. Three patches of optimal habitat within Phnom Prich Wildlife Sanctuary were identified. These, and connecting habitats, should be the target of increased enforcement effort to limit hunting and habitat conversion. Similar multi-grain models are likely to be valuable for conservationists within mosaic habitats as they can facilitate identification of the minimum suitable fragment size for species of conservation concern.

No abstract

Monitoring ungulate populations is an essential part of wildlife management with ungulates performing essential ecosystem roles including structuring populations of large carnivores. A number of ungulate species in Southeast Asia are also globally threatened and are therefore important conservation targets in their own right. We estimated large (. 15 kg) ungulate densities in two protected areas, i.e. Mondulkiri Protected Forest and Phnom Prich Wildlife Sanctuary, in eastern Cambodia using distance-based line transect sampling. During the 2009/2010 and 2010/2011 dry seasons, we surveyed 110 line transects (randomly distributed across 3,406 km 2 ) for a total of 1,310 km. We used DISTANCE 6.0 to model detection functions from observations of banteng Bos javanicus, wild pig Sus scrofa and red muntjac Muntiacus muntjak generating estimates of group density, cluster size and individual density. Estimated densities 6 SE were 1.1 6 0.2 individual banteng/km 2 , 1.4 6 0.4 individual wild pig/km 2 and 2.2 6 0.2 individual red muntjac/km 2 giving an overall density of approximately 4.7 large ungulates/km 2 . Although wild pig and red muntjac densities were within the range of estimates reported from ecologically similar protected areas in tropical Asia, overall large ungulate density is much lower than the intrinsic carrying capacity of deciduous dipterocarp forest. This appears largely to be due to the scarcity of large deer (i.e. hog deer Axis porcinus, sambur Cervus unicolor and Eld's deer Cervus eldii) as a result of extensive historic hunting. Current large ungulate densities appear too low to support a viable tiger Panthera tigris population in the long term, and ungulate recovery, driven by strong protected area management, needs to be achieved before tiger populations can be restored.