Abstract The forests of Borneo support some of the highest biodiversity in the world, yet have experienced among the world's highest rates of deforestation. Such rapid forest loss and associated fragmentation reduces the availability of suitable habitat for wildlife and creates dispersion barriers. Understanding the prevalence and impacts of this anthropogenic disturbance, and developing ways in which to mitigate such changes, is thus critical to the conservation of Borneo's wildlife. Here, we applied a path selection function with conditional logistic regression and used it to develop a resistance surface for a population of Sunda clouded leopards (Neofelis diardi) residing within a fragmented and human dominated landscape in Malaysian Borneo. We used cumulative resistant kernel and factorial least-cost path analysis to predict how connectivity may change in response to four future scenarios involving conversion of remaining unproductive forest to palm oil plantations, conversion of unproductive palm oil back to forest, and restoration of a riparian buffer zone along the river, and combination of the two forest restoration scenarios. We showed that Sunda clouded leopard movement is facilitated by forest canopy cover and resisted by non-forest vegetation, particularly recently cleared/planted and underproductive (flooded) plantation areas with low canopy closure. By combining resistant kernel and factorial least-cost path modelling we mapped core areas and the main linkages among them, and identified several key pinch points that may limit regional connectivity of the population. We predict that Sunda clouded leopard connectivity in the region can be greatly enhanced through the protection of privately owned forest patches and the reforestation of underproductive oil palm plantation areas, and creation of a forested buffer zone along the river. Conversely, we show that if the region's unprotected forests were to be converted to plantations then connectivity across the Kinabatangan floodplain would be significantly reduced.

中文翻译：

---

评估人类主导景观中巽他云豹连通性的景观变化情景

摘要 婆罗洲的森林支持着世界上一些最高的生物多样性，但却经历了世界上最高的森林砍伐率。如此迅速的森林丧失和相关的破碎化减少了野生动物合适栖息地的可用性，并造成了分散障碍。因此，了解这种人为干扰的普遍性和影响，并制定减轻此类变化的方法，对于婆罗洲野生动物的保护至关重要。在这里，我们应用了带有条件逻辑回归的路径选择函数，并使用它为居住在马来西亚婆罗洲的一个支离破碎和人类主导的景观中的巽他云豹（Neofelis diardi）种群开发了一个抵抗面。我们使用累积抗性内核和阶乘最小成本路径分析来预测连通性如何变化以响应四种未来情景，包括将剩余的非生产性森林转化为棕榈油种植园、将非生产性棕榈油转化为森林以及恢复河岸缓冲区沿河区，以及两种森林恢复方案的结合。我们表明，巽他云豹的运动受到森林冠层覆盖的促进，并受到非森林植被的抵抗，特别是最近被砍伐/种植和生产不足（水淹）的人工林区，冠层关闭率低。通过结合抗性核和因子最小成本路径建模，我们绘制了核心区域及其之间的主要联系，并确定了可能限制人口区域连通性的几个关键夹点。我们预测，通过保护私有森林斑块和重新造林低产油棕种植区，以及沿河建立森林缓冲区，可以大大增强该地区巽他云豹的连通性。相反，我们表明，如果该地区未受保护的森林被转变为人工林，那么京那巴当岸洪泛区的连通性将显着降低。