Mapping Aboveground Biomass and Soil Organic Carbon Density in India—A Geospatial-Analytic Framework for Integrating Multi-year Remote Sensing, Large Field Surveys, and Machine Learning

Advances in Earth observation technologies have allowed for the collection of big data for large-scale analyses to better understand the carbon cycle. This chapter, with three case studies, addresses a few methodological approaches for analyzing big data for two critical variables in the carbon cycle—Above Ground Biomass (AGB) and Soil Organic Carbon (SOC). The first case study presents an approach to improve AGB’s existing global products, which are known to have large regional uncertainties, using available ground measurements. The following case study explores the use of AGB and other environmental variables coupled with machine learning techniques to predict the spatial variability of SOC over forest regions in central India. The final case study is a unique attempt to create an extensive database of nearly 40 million ground measurements of SOC and how it can be leveraged to better understand SOC stocks in India. With the help of these three case studies, we highlight the various ways data can be utilized to study the carbon cycle and where future opportunities for such studies lie.