In 2015, unprecedented and sudden floods paralysed Chennai with over 18 lakh people being displaced. Following this, at the behest of the office of the Principal Scientific Advisor to government of India, research institutions, chief among them the National Centre for Coastal Research (NCCR), Chennai, and IITs, got together to build a flood warning system customised for use in Chennai. Carrying the acronym C-FLOWS, which stands for Chennai FLOod Warning System, the six-module ensemble can predict flooding due to heavy rainfall, sea-level rise and increase in water levels of the three rivers — Cooum, Adyar and Kosasthalaiyar — that traverse the city.
Ward-level data
“The State government shared data such as ward boundaries, population details, infrastructure available across Tamil Nadu, which have been used in the warning system,” says M. V. Ramana Murthy, Director, NCCR. The topography data was obtained from the Indian Remote Sensing programme. “Next we plan to develop such a system for Mumbai city and Cochin area,” he adds.
Knowing the elevation at different spots, the system can predict the way the area would flood based on different scenarios that have been simulated. Inputs were taken from the India Meteorological Department (IMD) on forecast and National Centre for Medium Range Weather Forecasting (NCMRWF), which gives the prediction for rain 10 days in advance. Similarly, INCOIS gives the hydrodynamic congestion such as storm surge and tide etc. Using this, the system can, two weeks ahead of the event, simulate the scenario. “The system can predict what would happen at the level of individual wards,” says Tune Usha, scientist with the NCCR and one of the key persons behind the development of C-FLOWS. “We can now apply it to the Greater Chennai Corporation area of 462 sq. km,” she adds.
Thus, early warnings could be issued. “The flood itself cannot be avoided, but it can be managed and the disaster mitigated,” says Dr Usha.
The spatial database of the city which contains all administrative layers, details of ward boundaries, infrastructure, details of elevations and surfaces, soil, land use, geomorphology and census datasets are contained in the first module. There are many precomputed scenarios of flooding based on the forecast rainfall and tidal conditions, in a library. From this the most appropriate one may be selected for superimposing on the geomorphology. The second module has to do with this precisely. Simultaneously, models can also be run in real time based on actual datasets. Over this can be overlaid details of storm water drains and drainage fractions to enable pre-flood planning operations.
The 3D visualisation module makes possible a realistic visualisation of the flooding in various places such as infrastructure, buildings, roads, ward boundaries etc, by superimposing the model outputs on the layers of data.
Convergence
Data from IMD, NCMRWF, INCOIS and Tamil Nadu State government are brought together in an online hub along with the field data and the remote sensing data to observe the situation in real time. This is the fourth module.
Authorised personnel can use a mobile app, which has been developed alongside, to collect data from flooded areas such as geotagged photographs and add to the database. This fifth module helps capture the ground reality and provide primary information for decision makers to plan relief and mitigation operations.
The last aspect is a decision support system: this is an online GIS query portal which can answer questions on quantum of flooding in specific localities, flood proximity, for example, it can say which roads are inundated and suggest routes for planning relief operations and so on.
According to Dr. K Satyagopal, Principal Secretary and Commissioner, Revenue Administration and Disaster management, C-FLOWS will be integrated with TN-SMART, a portal being developed by his department.
