Pterocarpus marsupium Roxb., locally known as Bijaysal, is a high-value tropical deciduous tree with a restricted and fragmented distribution in Nepal. We integrated 129 spatially occurring records with climatic, anthropogenic, edaphic, and topographic variables in an ensemble species distribution modelling framework (biomod2) to project current and future (2050s and 2090s) habitat suitability under four shared socioeconomic pathways (SSP) scenarios. Minimum temperature of the coldest month and precipitation seasonality were the dominant climatic drivers, followed by soil organic carbon, bulk density, and human footprint. The current suitable habitat is limited (11,154 km²; 7.6 % of Nepal) and confined to the western and central lowlands. Future projections indicate habitat expansion ranging from 40 % (SSP1-2.6 2090s) to 313 % (SSP2-4.5 2050s) with consistent southeastward centroid shifts (46–127 km), reflecting both northward expansion into the mid‑hills and eastward reorganization within the lowlands. However, these projections should be interpreted with caution, as actual migration may be limited by seed dispersal mechanisms and soil suitability in high-elevation zones. The unimodal response to human footprint challenges the conventional protected-area paradigm and indicates that well-managed community forests can serve as critical climate refugia. These findings suggest that well‑managed community forests can complement protected areas by maintaining intermediate disturbance and provide a spatially explicit roadmap for assisted colonization into the mid‑elevation zones.

Fire is a major ecological driver affecting vegetation growth species distributions, and wildlife habitats, emphasizing its ecological importance’s. This study analysis a vegetation health and fire dynamic in Langtang National Park (LNP) Nepal using MODIS derived fire occurrences data, Normalized Difference Vegetation Index (NDVI), Vegetation Condition Index (VCI), Climate Hazard Center Infrared precipitation with stations (CHRIPS) data, Land Surface Temperature (LST) data to examine the vegetation health and fire dynamic over 20 years (200-2020).The Mann-Kendall (MK) trend tests showed a significant increase in NDVI maximum values ( p < 0.05 ), while fire occurrences showed no consistent temporal trend ( p > 0.05 ). While (r = 0.62) a good positive correlation was found between maximum temperature and burn area, increased temperature relates to increase in fire activity and larger burn areas. Similarly VCI maximum highly predicated NDVI values (p = 0.0001), with moderate though significant fires impacting vegetation health. Grassland and needle leaf forests are the major land cover with relatively high fire frequency. Likely due to flammable biomass and seasonal dryness, while broad leaved closed forests displayed better fire resilience. The findings underline the need for integrated fire management strategies involving satellite monitoring, risk zonation, climate responsive planning, and community engagement to reduce fire risks in ecological sensitive area and strengthen ecosystem resilience in the Himalaya.