Research Fellowship in Computing Science, AI, Data, Environment

The environmental monitoring and its efficient management highly dependent upon the integration of heterogeneous data sources, followed by advanced numerical approaches and artificial intelligence techniques. In alignment with the objectives of JUNON programme, this fellowship focused on design and development of operational Digital Twin (DT) for environmental monitoring dedicated to groundwater and air quality systems. The aim of this project is to propose a Digital Twin architecture which is capable of integrating diverse data from in-situ sensors, satellite observations, physics-based and data driven models and to map them in web platform. The proposed system allows historical visualization, prediction, forecasting, 2d spatial mapping, and model-oriented simulation through services and ensures extensibility and long-term maintainability. A major contribution of the fellowship was consolidation and extension of an initial Digital Twin architecture into fully functional web application, incorporates data manager, scheduler for periodic updates, modeling services and user-friendly web interface. The proposed architecture was validates using real-world data across Centre-Val de Loire region, demonstrating its capability to handle heterogeneous temporal and spatial resolutions, and data quality constraints. Along with the technical development, this fellowship also contributed towards methodological design for hybrid modeling policies, operational robustness, and Digital Twins role for natural resource management. The outcome of this work offers a solid foundation for future research publications, technology transfer, and continued collaboration within the JUNON programme.