There has been an increasing interest in the scientific community toward modeling data as Knowledge Graphs, for effective exploration, analysis, and mining of actionable insights. Given the rapid rate of generation of humongous amounts of EO data, there is a dire need for the development of novel and efficient methods to effectively explore and leverage such huge repositories of remote sensing data. The depiction of remote sensing data as knowledge graphs has tremendous potential to create value for effective analysis and extraction of the otherwise implicit spatial and contextual information. The graph model for data representation is known to be amenable for seamless integration with other data sources. This chapter focuses on the transformation of information-rich heterogeneous EO data into Knowledge Graphs. It proposes and demonstrates an end-to-end modular workflow for Geospatial Knowledge Graph Construction from EO data for enhanced remote sensing scene understanding. It discusses the need for formalization of domain knowledge through the development of upper-level and ancillary ontologies. The chapter comprehensively describes the different processes—Remote Sensing Scene Ingestion, Classification, Geometric Shape Extraction and Resource Description Framework (RDF) based Serialization involved in the construction of Knowledge Graphs for Remote Sensing Scene Understanding. The chapter also illustrates use cases and application scenarios in the remote sensing domain that can potentially benefit by leveraging Knowledge Graphs for enhanced situational awareness. The chapter concludes with a brief discussion on the envisaged future directions for Geospatial Knowledge Graphs.