Composite Population–Road–Vegetation Index for Assessing Anthropogenic Vulnerability of Protected Areas in the Western Himalaya, India

Abstract

Protected areas are the principal instrument of in-situ biodiversity conservation, yet the intensity of human pressure on them varies widely from one reserve to the next, and no simple, transferable procedure exists for ranking that pressure across a large, administratively fragmented landscape. This study proposes such a ranking for the southern flank of the Indian Himalaya — the Union Territory of Jammu & Kashmir and Ladakh together with the states of Uttarakhand, Himachal Pradesh, and Punjab — a region that contains 69 named protected areas but has not previously been assessed as a single analytical unit. District-level population density and road density were each rescaled to a common 0–1 range and combined into a composite anthropogenic-pressure indicator, while sixteen-day MODIS MOD13Q1 NDVI composites were mosaicked, masked to the protected-area boundaries, and reduced to a per-reserve mean vegetation score. Two independent scoring schemes — a rank-based average and an exponential weighting function — were used to combine the pressure indicator with the vegetation score, and the two schemes were compared for internal consistency. Both approaches agree that the Talra Wildlife Sanctuary segment in Uttarkashi district, Uttarakhand, and several sanctuaries in the Kullu and Mandi districts of Himachal Pradesh obtain the most favourable combined scores (least vulnerable), while the Manali Wildlife Sanctuary segment in Kangra district is consistently the lowest-scoring (most vulnerable) reserve in the pixel-inclusive analysis; when open-water pixels are excluded from the vegetation calculation, Bir Gurdialpura in Patiala district, Punjab, replaces Manali as the lowest-scoring site. Across the region, mean NDVI is positively rather than negatively rank-correlated with district-level population–road pressure (Spearman ρ ≈ 0.53 across 36 districts), because the lowest NDVI values occur in high-altitude, sparsely populated trans-Himalayan reserves; low NDVI in this dataset therefore appears to reflect elevation and aridity more than anthropogenic degradation. Because the analysis is based on a single vegetation-index composite rather than a multi-year time series, these patterns describe present-day spatial variation in reserve condition rather than a temporal trend, and are presented accordingly. The resulting ranked table offers state forest and wildlife departments a transparent and repeatable basis for prioritising monitoring and enforcement resources among competing reserves.