Assessing Climate-Induced Vegetation Dynamics in the Kullu Valley Watershed, Western Himalayas, through Remote Sensing and Machine Learning: Statistical Analysis spanning 2000–2020

Abstract The dynamics of vegetation are critical for identifying climate change trends. The current study aims to examine the spatio-temporal changes in vegetation cover in the Kullu valley (Western Himalaya) during the decadal 2000, 2010, and 2020. The objective of the study includes the computation of normalized difference vegetation index (NDVI) vegetation spectral indices, the extraction of various classes of vegetation, and statistical analysis of the sequential Mann Kendall test and Mann-Kendall (MK) test on historical metrological data from the study site (specifically precipitation, relative humidity, and temperature). Power data access viewer (NASA) datasets have been used in the statistical analysis of climatological data. The primary feature classes in the study are forest cover, snow, river, and grassland/scrubland. The result indicated that the region's grassland cover declined by 120.57 km2 and its forest cover decreased by 40.6 km2 between 2000 and 2020. The results demonstrate that climatic variables like slopes and increased minimum temperatures by two meters (A), minimum temperatures (B), maximum temperatures (C), relative humidity (D), and annual mean precipitation (E) from 2000-2020 are the main factors limiting vegetation growth. The determined NDVI displays significant variations across the study area. The annual maximum temperature was falling. The study's objectives are to: 1) analyse the spatiotemporal variation of vegetation cover, 2) identify its primary drivers, and 3) examine statistical trends in long-term metrological data. The result of the research presented will be useful in properly managing and monitoring the forest ecosystem.