Abstract ID: 3.12967 | Accepted as Talk | Requested as: Talk | TBA | TBA

Remote Sensing (RS) and Geographic Information Systems (GIS) have become essential tools for modelling species distributions, but their accuracy is often limited by the spatial resolution of available environmental datasets and the ecological complexity of mountainous landscapes. In the Western Himalaya, Taxus species valued for their medicinal properties are predominantly modelled using climatic and topographic variables. However, discrepancies often arise between predicted and actual distributions due to unresolved sub-grid ecological processes, spatial heterogeneity, and the exclusion of critical microhabitat factors from large-scale datasets. Sub-grid ecological processes refer to fine-scale environmental interactions, such as soil moisture variations, understory competition, and localized human disturbances, which are often averaged out in coarse-resolution RS-GIS models. Spatial heterogeneity, particularly in complex mountain systems, results from the high variability in temperature, precipitation, and soil conditions across short distances. For instance, in Uttarakhand’s Kumaon region, Taxus populations are often found in mixed oak-rhododendron forests with deep, moist soils, whereas in Garhwal, they are more restricted to isolated patches within conifer-dominated forests, affected by aspect-driven microclimatic differences. Similarly, in Himachal Pradesh, Taxus distribution varies significantly between the moist, shaded valleys of Kullu and the drier slopes of Chamba, challenging uniform model predictions. This study examines whether RS-GIS models adequately capture these variations by comparing model predictions with field-measured Taxus distribution. The study assesses the role of microhabitat conditions—such as canopy cover, soil depth, and disturbance levels—often absent from spatial models. Preliminary observations suggest that while RS-GIS models provide broad-scale predictions, their reliability diminishes when applied to fine-scale habitat assessments. Addressing these limitations requires integrating high-resolution ecological data and fostering collaboration between modelers and field researchers to improve the accuracy of species distribution models. This research underscores the need for interdisciplinary approaches to enhance habitat modelling and conservation planning in the Western Himalaya.