The Asiatic black bear (Ursus thibetanus), a large and vulnerable mammal in northern Pakistan, faces increasing threats from habitat loss, tourism, human disturbance, and human–bear interactions. However, spatially explicit information on its potential habitat remains limited. In this study, 126 new Asiatic black bear presence records, collected through bear sign surveys and camera trapping in the Kalam Valley, were combined with an ensemble modelling approach incorporating regression-based, machine-learning, and maximum entropy models to map habitat suitability and identify key areas for conservation planning in northern Pakistan. The resulting ensemble habitat suitability map indicated relatively higher suitability in the northern and north-central portions of the study area, whereas southern areas consistently exhibited lower relative suitability. The binary suitability map classified 1061.22 km2 as suitable habitat, mainly located between 400 and 3500 m above sea level. Variable importance analysis identified the presence of rivers as the most important predictor of habitat suitability, followed by distance to settlements and the vector ruggedness measure. All models showed high values of the AUC, TSS, and Boyce index. Our results provide a spatially explicit and practical basis for habitat protection, field validation, connectivity assessment, and conflict-sensitive conservation planning for the Asiatic black bear in northern Pakistan.
Mapping Habitat Suitability of the Asiatic Black Bear (Ursus thibetanus) in the Kalam Valley, Pakistan: An Ensemble Modelling Approach
Luciano Bosso
2026
Abstract
The Asiatic black bear (Ursus thibetanus), a large and vulnerable mammal in northern Pakistan, faces increasing threats from habitat loss, tourism, human disturbance, and human–bear interactions. However, spatially explicit information on its potential habitat remains limited. In this study, 126 new Asiatic black bear presence records, collected through bear sign surveys and camera trapping in the Kalam Valley, were combined with an ensemble modelling approach incorporating regression-based, machine-learning, and maximum entropy models to map habitat suitability and identify key areas for conservation planning in northern Pakistan. The resulting ensemble habitat suitability map indicated relatively higher suitability in the northern and north-central portions of the study area, whereas southern areas consistently exhibited lower relative suitability. The binary suitability map classified 1061.22 km2 as suitable habitat, mainly located between 400 and 3500 m above sea level. Variable importance analysis identified the presence of rivers as the most important predictor of habitat suitability, followed by distance to settlements and the vector ruggedness measure. All models showed high values of the AUC, TSS, and Boyce index. Our results provide a spatially explicit and practical basis for habitat protection, field validation, connectivity assessment, and conflict-sensitive conservation planning for the Asiatic black bear in northern Pakistan.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
