Floods are one of the most persistent and destructive natural disasters on Earth, wreaking havoc across landscapes and devastating lives and livelihoods. According to the National Disaster Management Authority of India, out of the total geographical area of 329 million hectares in India, more than 40 million hectares are flood-prone. It also notes that on average, every year, 1600 lives are lost and damages worth Rs.1805 crores are caused to crops, houses and public utilities due to floods.
These raging waters are particularly problematic in deltaic regions, where rivers meet another body of water, like the sea. Deltas are places where the interaction of inland waters and coastal tides amplifies the risk of catastrophic flooding. Predicting floods accurately in such regions is incredibly tough because they aren't just one thing causing the problem – it's a complex mix of rain, tides, and waves all colliding at once.
To address the challenge, researchers from the University of Notre Dame, USA, Indian Institute of Technology (IIT) Kharagpur, India Meteorological Department (IMD), and the National Institute of Hydrology (India) have developed an innovative computer model for predicting floods in deltaic regions.
Their study centres around the Brahmani-Baitarani River delta in Odisha, India, where the two rivers meet to form a delta before pouring out into the Bay of Bengal. The region is notoriously susceptible to floods and cyclones. The team developed an integrated method, where, instead of relying on separate models that only look at the river or the ocean separately, they've linked them together. This allows the models to talk to each other and understand how a storm surge in the Bay of Bengal pushes water up the rivers while heavy rainfall inland swells the rivers at the same time.
The team used three computer models: ADCIRC, SWAN, and HEC-RAS 2D. ADCIRC (Advanced Circulation Model) takes on the task of simulating how water moves in the ocean, accounting for tides and storm surges. SWAN (Simulating Waves Nearshore) focuses on the waves and calculates their height and direction. These two work together, with ADCIRC sending information on water levels to SWAN and SWAN returning data on wave forces. Finally, HEC-RAS 2D (Hydrologic Engineering Center's River Analysis System) simulates river flooding, taking into account rainfall and the flow of water downstream. The key to the new model is in the way these three models are connected. The ocean surge data from ADCIRC+SWAN becomes the coastal boundary condition for the HEC-RAS model, informing it how much the ocean is pushing into the rivers.
The team tested and refined their model using data from past cyclones. They focused on the two most recent major cyclones: Fani (2019) and Yaas (2021). By feeding the model with weather data from these events and validating it against satellite imagery, the scientists were able to fine-tune it. The model was given wind and pressure information during the cyclones from sources like the Climate Forecast System Version 2 (CFSv2) and European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis 5 (ERA5).They then used satellite images from Sentinel-1 to verify how well the model predicted the actual flooded areas. The results were impressive, with the model accurately predicting water levels and flood extents.
The research offers a significant step forward in our ability to predict and prepare for coastal flooding. By combining different models into a unified framework, scientists are providing a clearer picture of the complex forces that drive these devastating events. Coastal and deltaic communities stand to gain immensely from these accurate models as they pave the way for real-time flood warnings, enabling faster response and more intelligent resource allocation during cyclonic events.
This research article was written with the help of generative AI and edited by an editor at Research Matters.