Early warning system for flood prediction in the river basins of India

Early Warning System for Flood Prediction in the River Basins of India

Objectives:

Design, develop and deploy an Early Warning System for Flood Prediction (EWS-FP) on HPC platform
Develop Sediment Transport model
Develop Integrated Reservoir Operation tools
Design geospatial portal for information dissemination on flood prediction

Key Deliverables:

Early Warning System for Flood Prediction- 2-days flood forecast (water level, inundation extent, flow, village level % inundation)
Integrated framework for flood modelling at River Basin Level adaptable for the entire country

End Users:

Central Water Commission (CWC)
Odisha State Water Resources Department (OSWRD)
National and State Disaster Management Authorities (NDMA, SDMA)
National Disaster Response Force (NDRF)
District Administration

Collaborators:

Central Water Commission (CWC), Delhi
Indian Institute of Science (IISc), Bengaluru

About the Project:

India is highly vulnerable to floods, which has large scale economic as well as social impacts. To address the issue and lessen the burden of the disaster management agencies, Centre for Development of Advanced Computing (C-DAC), Pune, is executing a project viz., ‘Early Warning System for Flood Prediction for River Basins of India’ under the National Supercomputing Mission of MeitY and DST, Govt. of India.

Under this project, three important aspects of flood management are being handled. Flood Prediction and Early Warning, Integrated Reservoir Operations and Sediment Transport Model.

A free and open source software tool for 2D hydrodynamic modelling is being used for prediction modelling and simulation. The model is designed such that it is both scalable and flexible and without much changes, except input data, and can be implemented in any river basin of India. The simulation runs for predicting floods are being carried out since year 2020. Every year daily flood predictions have been carried out for the monsoon season (June to October) for Mahanadi Basin. The model is massively parallelised and NSM HPC resources are being used for carrying out these daily simulation runs. The results have been shared with State Water Resources Department and Central Water Commission for validations. Since 2022 monsoon season, Tapi River Basin simulations have also been started.

The daily outputs include a 2-day flood forecast in the form of village-level percentage inundation information and predicted inundation spread and water level information. After proper validation and calibration exercises the model may be implemented in other flood affected river basins of the country. Both Odisha State Water Resources Department and Central Water Commission Bhubaneshwar have been part of this project and as such their continuous support has been fruitful for the project.

Figure: Portal for visualization of daily forecast (Inset – Report showing villages that are flooded 40% or more by area)

Status:

Currently the water level, inundation spread and flow forecast for the next 24 hours is being shared with the Odisha State Water Resources Department (OSWRD) on a daily-basis. The accuracy and lead time of the forecast is appreciated by the department
Geospatial Portal (SimInu) being developed on open source technology presents an opportunity to develop indigenous early warning dissemination systems for the country