Monitoring vegetation degradation using remote sensing and machine learning over India – a multi-sensor, multi-temporal and multi-scale approach

Abstract

Vegetation cover degradation is often a complex phenomenon, exhibiting strong correlation with climatic variation and anthropogenic actions. Conservation of biodiversity is important because millions of people are directly and indirectly dependent on vegetation (forest and crop) and its associated secondary products. United Nations Sustainable Development Goals (SDGs) propose to quantify the proportion of vegetation as a proportion of total land area of all countries. Satellite images form as one of the main sources of accurate information to capture the fine seasonal changes so that long-term vegetation degradation can be assessed accurately. In the present study, Multi-Sensor, Multi-Temporal and Multi-Scale (MMM) approach was used to estimate vulnerability of vegetation degradation. Open source Cloud computing system Google Earth Engine (GEE) was used to systematically monitor vegetation degradation and evaluate the potential of multiple satellite data with variable spatial resolutions. Hotspots were demarcated using machine learning techniques to identify the greening and the browning effect of vegetation using coarse resolution Normalized Difference Vegetation Index (NDVI) of MODIS. Rainfall datasets of Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS) for the period 2000–2022 were also used to find rainfall anomaly in the region. Furthermore, hotspot areas were identified using high-resolution datasets in major vegetation degradation areas based on long-term vegetation and rainfall analysis to understand and verify the cause of change whether anthropogenic or climatic in nature. This study is important for several State/Central Government user departments, Universities, and NGOs to lay out managerial plans for the protection of vegetation/forests in India.