濒危物种
物种丰富度
栖息地
生态学
地理
地方性
濒危物种
物种分布
生物多样性
近危物种
栖息地破坏
全球生物多样性
保护区
物种多样性
常见种
伞种
山麓
指示物种
栖息地保护
环境生态位模型
空间生态学
IUCN红色名录
植被(病理学)
关键栖息地
消光(光学矿物学)
自然保护区
作者
Wilbert A. Aureo,Cristino Tiburan,Eduardo C. Calzeta,IRENEO L. JR. LIT,Yingwei Yan,David Mark taylor,Rico Ancog
标识
DOI:10.3897/oneeco.11.e184595
摘要
Identifying suitable habitats and High Conservation Value Areas (HCVAs) for threatened species is critical for informing site-based management in biodiversity-rich, but vulnerable landscapes. In this study, Species Distribution Modelling (SDM) and species richness analysis were applied to identify areas of high conservation value within Rajah Sikatuna Protected Landscape (RSPL). Using the MaxEnt algorithm, we modelled habitat suitability for thirteen threatened species across major taxonomic groups, incorporating topographic and bioclimatic variables. Models showed high performance (mean Area Under Curve (AUC) = 0.905, True Skill Statistic (TSS) = 0.717 and Variance Inflation Factor (VIF) < 4), with terrain ruggedness, slope, mean annual temperature and annual precipitation identified as primary drivers of model structure. While statistical performance was robust, model outputs were strictly framed as spatial hypotheses to account for the data-limited nature of the focal endemics and the use of bioclimatic layers as proxy gradients. The resulting potential suitability maps revealed heterogeneous, species-specific distributions, ranging from localised and fragmented habitats for Dipterocarpus grandiflorus , Gallicolumba crinigera and Shorea guiso , to broader ranges for Edolisoma mindanense , Platymantis guentheri and Voacanga megacarpa . Stacked binary models produced a species richness map that highlighted northern and north-western RSPL as biodiversity hotspots, while southern areas were relatively depauperate, reflecting probable degradation. Threshold-based analyses to delineate HCVAs showed that the CRFD (Cumulative Richness Frequency Distribution) inflection point (~ 6 species) best balanced spatial extent and ecological sensitivity, capturing 90% of species occurrences. In contrast, conservative thresholds (e.g. 95 th percentile, Mean + 1 SD) isolated only a few core hotspots, but excluded moderately rich areas that may serve as corridors or buffer zones. This integrated modelling approach offers valuable insights, enabling the prioritisation of potential critical habitats and identification of targets for ecological restoration.
科研通智能强力驱动
Strongly Powered by AbleSci AI