Raging rains: Why Chennai fell apart
On December 1, houses on the ground floor in Jafferkhanpet, a neighbourhood in southern Chennai, started to inundate because of torrential rains. By five in the morning, almost 80 percent of the city was submerged under four metres of water. The situation continued for the next 72 hours, killing more than 500 and destroyed infrastructure worth Rs 500 crore. It was most intense of the three spells that had battered the city in just one month. Earlier, the city had witnessed incessant rains from November 11 to 13, and then from November 15 to 17.
Chennai received 1,200 mm of rainfall in November 2015, which was the highest rainfall the city received in November in the past 100 years. The city on an average receives 407.4 mm rainfall in November. On December 1, Chennai received 300 mm rainfall, making it the wettest December day ever recorded in the city. The normal rainfall for Chennai in December is 191 mm.
A closer look suggests that the intensity and the resultant losses due to the recent floods could have been greatly reduced. Tamil Nadu faltered on several accounts. Firstly, it failed to act despite a clear warning from the India Meteorological Department (IMD) of heavy rains. Secondly, the state administration has over the years done little to prepare for disasters despite being flood-prone. Thirdly, Chennai and its neighbouring areas have witnessed unplanned urbanisation in recent years that has destroyed the city’s natural flood sinks such as marshlands and river channels.
IMD in mid-October issued a forecast that predicted 11-12 percent above normal rains in the southern states with a probability of about 90 percent. It had said that the northeast (winter) monsoon, caused by retreating monsoon winds that attain moisture from the Bay of Bengal on their way back to the south from the northeastern direction, would be stronger. These winds are responsible for the rains in the southern states of Andhra Pradesh, Tamil Nadu, Kerala and parts of Karnataka between October and December. The IMD forecast, though, did little to prepare the states for the situation.
On December 1, an ill-prepared Tamil Nadu administration decided to open the gates of the Chembarambakkam reservoir, and the released water inundated the city. A public interest petition filed in the Madras High Court against the Tamil Nadu government suggests 1,104 cubic metres per second of water were released into the Adyar river, which meets the water requirements of the city. Highlighting how the decision was delayed, the petition, filed by Chennai-based businessperson Rajiv Rai, says that the water was released after a warning was issued at midnight. “If one studies the levels of water in the various catchment tanks on a daily basis, one can see that the reservoirs had much greater inflow than outflow right through November 2015,” says the petition. It alleges that state chief secretary K Gnanadesikan waited for three days after the Public Works Department (PWD) wrote to him on November 29 to release water from the reservoir. “The chief secretary, even though he was well aware of the reports that there was going to be a heavy downpour for a few more days, didn’t direct the release of water when he received the warning from PWD,” the petition says.
Clueless, yet confident
The administration was caught unawares despite the fact that floods are not new to the state. According to a 2013 Comptroller Auditor General of India (CAG) report, there have been 50 cyclonic storms in the region between 1900 and 2009. Even the hazard profiles of coastal districts prepared in the aftermath of the 2004 tsunami that hit the Tamil Nadu coast shows that most of the coastal districts experience flooding during the retreating monsoon, which normally accounts for 48 percent of the rainfall in the state. “Heavy rains during the months of October, November and December inundate low-lying areas, coastal areas and the areas nearby major irrigation sources,” states the disaster profile prepared by the district administration.
The level of unpreparedness of the city administration can be gauged from the fact that the Chennai Metropolitan Development Authority (CMDA), the key urban planning agency for the city, still relies on an outdated hazard profile of Tamil Nadu, which says none of the areas in the state is flood-prone. According to CMDA’s hazard profile document, “From the flood hazard map of India (mapped by IMD, New Delhi), it is seen that no area in Tamil Nadu falls in the risk zone.” The document says that few areas in Chennai might get flooded due to heavy storms and for this, flood affected areas have to be mapped. According to N Madhavan, who has worked with the Municipal Corporation of Chennai, ward-level vulnerability mapping was done after the 2004 tsunami. “These efforts went waste as boundaries of wards and zones changed following the expansion of the city,” he said.
The callousness of the state administration does not end here. The State Disaster Management Authority, which was set up in 2008 under the Disaster Management Act, 2005, has not met even once, according to the 2013 CAG report. Almost seven years later, the state is yet to come up with disaster management rules. It was only in May 2015, following the Nepal earthquakes, that the Municipal Corporation of Chennai had started to prepare a comprehensive disaster management plan which pre-defines the roles of officials during a disaster.
Only to get worse
The northeast monsoon is generally known to bring a few heavy showers amidst scanty rains to the southern states. Between October and the first week of December, 27 of the 34 districts in Tamil Nadu and Puducherry had received 10 percent excess rainfall with several districts registering more than double the normal rainfall. While normally it rains heavily for three to four days and is followed by long dry spells, this year it rained continuously with practically no respite for more than a month. Chennai received 1608.6 mm rainfall between October 1 and December 16, 2015, which was more than twice the amount of rainfall the city received during the last three months of 2014 (719.7 mm).
“There are several factors responsible for the performance of the northeast monsoon in southern states. The El Niño Southern Oscillation (ENSO) along with the Indian Ocean Dipole and the Madden-Julian Oscillation (MJO), which are heating patterns in the Indian and Pacific oceans, has caused a lot of atmospheric activity. These have culminated in the rain in the southern states,” says B P Yadav, IMD director.
Several studies since the early 2000s have pointed to a positive correlation between ENSO and the northeastern monsoon. “ENSO has been known to suppress south-west monsoons while enhancing the northeast monsoons. This year we have seen this to a great extent,” says Gibies George, a senior research fellow at the Indian Institute of Tropical Meteorology (IITM), Pune.
The other global pattern that has influenced the northeast monsoon in India this year is the MJO—a traversing global oceanic and atmospheric phenomenon that has been found to have global climatic ramifications. MJO has a 30-60 day cycle that is characterised by eastward moving clouds in the tropics over the Indian and Pacific Oceans. It has been seen that when the MJO cycle is close to 30 days, it aids rainfall in the country while a cycle longer than 40 days causes suppression. “A strong MJO remained practically stationary over the Indian Ocean all through November and this has been a major factor in the heavy rainfall. The differential due to MJO has forced rain-bearing winds into the southern region of the sub-continent. The influence of MJO has been in line with previous observations,” says D S Pai, director, Long Range Forecast division, IMD.
These connections are not the only newly emerging patterns regarding the northeastern monsoon. A close look at the seasonal rainfall levels in recent decades across districts in coastal Andhra Pradesh, Karnataka and Tamil Nadu reveals that the northeast monsoon has become progressively more bountiful. A 2012 study in Theoretical and Applied Climatology makes use of homogenous rain gauge data maintained by IITM to show that winter rains in peninsular India have exhibited a positive rainfall trend of 0.4 mm per day per decade between 1979 and 2010. At the same time, an increase in the number of extreme rainfall days and a decrease of normal rain days have also been observed in several parts of south India. According to George, this trend is similar to what has been observed during the summer monsoon. “We have observed that with the warming of the atmosphere, the moisture carrying capacity of the air also increases. Hence, with warming, we can expect to see an increase in the magnitude of rain events and the frequency of extreme rain events. Normal rain days will become fewer and less frequent,” says George. Experts warn that rainfall during the northeast monsoon is likely to increase in the future, another reason the state administration should get its act together.