Deep S. Banerjee

Following the completion of MSc in Physics, Deep began his career as Guest Lecturer in Physics at SS College (Govt. Sponsored) followed by Assistant Professor in Department of Physics at SKFGI Engineering College in India. Subsequently, Deep joined the Indian National Centre for Ocean Information Services (INCOIS), an autonomous research institution under the Government of India, as Project Scientist in Data Assimilation and Ocean Modelling division.

At INCOIS, Deep’s primary focus was on developing and implementing a state-of-the-art Data Assimilation system for the Indian Ocean Region, using ROMS (Regional Ocean Modelling System) – LETKF (Local Ensemble Transform Kalman Filter) framework for Ocean state forecast for Indian Ocean domain. Concurrently, he continued to enhance the INCOIS-Global Ocean Data Assimilation System based on the Modular Ocean Model (MOM)-3DVar. This effort significantly improved real-time and near real-time ocean reanalysis production, leading to enhanced predictions of tropical cyclones in the Indian Ocean domain.

Later, Deep joined the Euro-Mediterranean Center on Climate Change (CMCC), Italy as Research Associate in the Ocean Modelling and Data Assimilation division. He contributed to the enhancement of the global ocean data assimilation system, CGLORS, funded by Copernicus Marine Service (CMEMS), with a focus on the NEMO-OceanVar framework. Subsequently, he engaged in working with Community Earth System Model (CESM) to introduce a Sea Ice Data Assimilation scheme in NEMO-CICE coupled framework to improve the ocean and sea ice states for a better seasonal climate prediction.

Currently, Deep is a Modelling Scientist at Plymouth Marine Laboratory, where his work brings artificial intelligence directly into the core of marine ecosystem forecasting. Under the NECCTON (European Union), National Centre for Earth Observation (NCEO) and UK Research and Innovations (UKRI) funded projects, he developed and successfully demonstrated the use of AI-generated nutrient products assimilated into the UK Met Office AMM7 biogeochemical forecasting system. This work led to significantly improved phytoplankton forecasts and subsequent publications. At the same time, Deep has contributed a high-resolution digital twin system for harmful algal bloom early warning, combining ecosystem modelling with glider assimilation, now completed and operationally demonstrated. Alongside, he is leading ecosystem parameter estimation in an ESA-funded project. At the same time, his work on hybrid AI–ecosystem models has already been demonstrated as proof of concept, with the next step being their transition into operational forecasting systems in collaboration with forecasting agencies.

Deep has published in several peer reviewed international journals, reviewed for Biogeosciences and the Journal of Geophysical Research (AGU), and frequently presents at international conferences. His work consistently advances the development of next-generation hybrid AI- biogeochemical and ocean forecasting systems, connecting academic innovation with practical operational use. Deep also regularly reviews manuscripts for journals including Nature Communications, Scientific Reports, Earth System Dynamics (EGU), Biogeosciences (EGU), and Journal of Geophysical Research (AGU).

Key Projects

• The Climate and Marine Production (CAMP) funded by ESA
• New Copernicus Capability of Trophic Ocean Networks (NECCTON) funded by EU
• Synchronising Earth Observation and Modelling Frameworks Towards a Digital Twin Ocean (SyncED-Ocean) funded by UKRI
• 114-GLORAN CMEMS LOT 2 Research programme funded by Copernicus Marine Services, EU
• CMCC_C3S2_370 project, funded by Copernicus Marine Services, EU
• High-resolution Operational Ocean Forecast and reanalysis System (HOOFS) funded by the Ministry of Earth Sciences, Govt. of India.

Selected Publications