Toward an Intelligent Decision Support System for Environmental Application Using Earth Digital Twin

Digital Twin (DT) is an innovative technology representing virtually a physical object. DT applications include real-time monitoring, planning, optimization, maintenance, remote access, etc. DT allows gathering massive and heterogeneous data from the physical environment to understand the link between past events, monitor the present, and predict the future. The implementation of D.T. is increasing in various domains like the industrial, automotive, and agriculture. This paper describes an overview of Earth Digital Twin (EDT). Firstly, we discuss the challenges of the implementation. Secondly, we present a generic EDT architecture for environmental applications with a real case study, particularly based on simulation scenarios. EDT is mainly employed to 1) utilize massive and heterogeneous data, 2) implement a suitable machine learning model structure, and 3) develop a platform to help policy and support decision-making.