You are here

The 2015 Chennai floods – A rapid assessment

In December 2015, Chennai witnessed massive floods killing about 400 people and displacing thousands. Now, researchers from Interdisciplinary Centre for Water Research (ICWaR) at the Indian Institute of Science, Bangalore, in collaboration with researchers from Indian Institutes of Technology (IITs), Madras and Bombay have assessed the flood related information in the public domain and data from secondary sources and carried out quick first level analyses to develop an understanding about possible reasons for the floods. They observed that the massive floods were a result of a combination of factors; high rainfall intensity, overflowing rivers, global climate drivers, unplanned urbanization, inadequate drainage system and upstream reservoir releases.

Apart from deaths, the floods also damaged more than 100,000 structures and caused economic losses ranging from 2 lakhs to 20 lakhs. “While the loss was humongous, the positive thing that came out of our study was that the health issues were kept in control”, says Prof. Mujumdar, Chairman, ICWaR and one of the lead members of the study.

The researchers assessed the hourly rainfall intensities from the IMD weather stations at Nungambakkam and Chembarambakkam for the 1st and 2nd December 2015. An intensity duration frequency (IDF) plot of this data showed that the 24-hour rainfall recorded in Nungambakkam (within the city) on December 1, 2015 was more like a 25-year storm, whereas the rainfall that occurred in Chembarambakkam (outskirts of Chennai city) could be higher than a 100-year storm. The rainfall in Chembarambakkam was more than the city could handle and therefore brought in huge influx of water into the city.

“We also did a return period analysis to forecast the possibility of such intensity of rainfall. The results showed that the 2015 extreme rainfall event in Chennai was found to be rare with a return period close to 100 years in the observed record excluding this event”, mentions Prof. Mujumdar.

While the study does not make any conclusive inference on the effect of El Nino on the Bay of Bengal, it does observe that El Nino in 2015 was the strongest and Bay of Bengal was warming up during the same time, leading to excess precipitation during these days. “Fortunately, the tidal levels were not high during this time, otherwise a worse situation would have occurred”, expresses Prof. Mujumdar.

The study further noted that the reservoir releases from the Adyar, Chembarambakkam and Poondi reservoirs brought in additional waters into the city. Together, the flood at its peak when it entered the city has been estimated to be about 3,800 m3/s (1,34,195 cusecs), whereas the flood carrying capacity of the Adyar river is only about 2,038 m3/s (72,000 cusecs). Though the reservoir releases were higher than usual, the flooding cannot be attributed only to this, as the rivers were also full due to heavy rainfall in November. More importantly, the reservoirs do not act as flood control systems, but are only drinking water reservoirs.

The micro drainage system (storm water drainage system) in the city is unable to drain the high intensity rainfall that occurred during the event, pointed the study. “The storm-water drains (SWD) were designed bearing in mind certain rainfall intensity. However, over the years, the rainfall intensities may have increased. In addition, the solid waste dumping in the drains has increased, causing blockage. This contributes to local flooding and exacerbates the situation”, adds Prof. Mujumdar.   

While this report has attempted to compile the information available, gaps do exist in it to make any conclusive inferences and provide information for making informed decisions by the concerned authorities. “Urban floods management is not at the forefront as it should be. Developing urban flood models is a necessity and we are trying to develop them though which we will be able to understand the impact of rainfall intensity and response of the surface”, explains Prof. Mujumdar. “Data becomes very critical in developing such forecast models. Hence, maps of SWD, high resolution rainfall, groundwater levels and terrains are essential to be maintained. Also, ‘wiring of the cities’ with sensors at appropriate places to collect data would be a useful investment”, he adds.

“All of this together backed by appropriate reservoir operation policies, functioning weather stations would help in improving disaster preparedness and management,” concludes Prof. Mujumdar.

The team comprised of Profs. Balaji Narasimhan, B.S. Murty (from IIT Madras), Subimal Ghosh, Arpita Mondal (IIT Bombay) and P P Mujumdar (IISc Bangalore).

The report and supplementary material may be accessed from  


Contact Information:

Prof. P. P. Mujumdar is the Chairman of ICWaR and a Professor in Civil Engineering Department at IISc, Bangalore.

He can be contacted at: