Vegetation trend analysis and change quantification based on time series satellite data for Northeast India

Mapping vegetation change and dynamics is an essential input for natural resource management and ecosystem-related policy design. Trend analysis on vegetation indices (VI) derived from satellite images is a common way of mapping vegetation change due to its ease of use, generalizability, and acceptable accuracy for large-scale applications. To better quantify the vegetation change, the current study analyzes the long-term spatio-temporal variability of vegetation and trend breaks in vegetation vigour for the Northeast region of India based on the normalized difference vegetation index (NDVI) derived from moderate resolution imaging spectroradiometer (MODIS) data as well as NDVI derived from Global Inventory Modelling and Mapping Studies (GIMMS). This chapter summarizes approaches to detecting vegetation change and methods for observing spatio-temporal trends based on long-term NDVI records. Standard anomaly-based methods designed to overcome assumptions of long-term linearity in time series analysis are also discussed. Finally, a novel approach for identifying breaks in vegetation trends is presented to quantify the year during which the trend break took place by decomposing the time series information on vegetation vigour into its seasonal and non-seasonal components based on bi-monthly data from 1982 to 2020. The results demonstrate nonlinear changes in the trend of natural vegetation. A gradual decline in positive anomaly was observed between 2006 and 2008 in Arunachal Pradesh. Positive anomalies were noticed in 2003 for the states of Meghalaya and Assam. Our findings indicated vegetation frequency of negative anomalies ranged from 53.6% to 27.5%, particularly in the regions of Arunachal Pradesh and Assam. The breaks in trends of vegetation were mainly identified in the years 2005, 2010, 2014, and 2019.