Science & Technology

What caused heavy rains in Tamil Nadu

An independent weather forecaster tries to explain the record-breaking December rains in Tamil Nadu

 
By Akshay Deoras
Last Updated: Saturday 05 December 2015

(NOTE: Views expressed in this blog are personal and they don’t reflect the views of any organisation)
 
It wasn’t a monsoon depression or a tropical cyclone. Then what was it? The 24-hour rainfall recorded in Chennai, parts of Tamil Nadu and Puducherry on December 1-2, 2015, raises this big question for researchers. Heavy rains and flooding potential in Chennai and other areas were anticipated, but the record-breaking rainfall figures have stumped everyone. 
 
In just 24 hours (till 08:30 am IST on December 2, 2015), Chembarambakkam, a lake southwest of Chennai, received a whopping rainfall of 475 mm. In the same period, Kattankulathur (near Chennai) received 445 mm rainfall, followed by 429 mm rainfall at Kattupakkam. The astonishing fact is that as per a report, 24-hour rainfall in at least three places in Tamil Nadu is more than maximum 24-hour rainfall of 380 mm recorded at Banki, Odisha during Tropical Cyclone Phailin
 
24-hour rainfall in Tamil Nadu and Puducherry ending on 08: 30 am IST on December 2, 2015 (Courtesy: Vagaries of Weather)
 
Daily mean rainfall in Tamil Nadu and Puducherry between October 1 and December 2, 2015. Fifty-eight per cent above normal rainfall has been recorded in this period (Courtesy: IMD RMC Chennai)
Chennai airport got 345 mm rain in a day whereas city-side got 294 mm rain. Rainfall statistics too present an unusually heavy rainfall picture with the star mark representing December’s heavy rainfall event. 
 
Tropical cyclones are mostly responsible for bringing such heavy rainfall but Tamil Nadu’s heavy rain wasn’t caused by any tropical cyclone. It was caused by an ordinary low pressure system in the Bay of Bengal which performed extraordinarily. 
 
Monsoon low pressure systems 
 
India (in fact South Asia) experiences two monsoon seasons in a year. We are often familiar about the southwest monsoon season (June to September) which is a primary source of rain in most of India. For residents in parts of India (excluding southern) it’s winter period right now. But, for southern parts of India such as Tamil Nadu, it’s the real monsoon season. The state receives about 48 per cent of its annual rain in the northeast monsoon period (October to December). Coastal districts of the state get upto 60 per cent of their annual rain in this period. So, it’s completely natural for rain in these areas at this time of the year and there isn’t any "unseasonal" factor involved. 
 
Northeast monsoon (NEM) is known to be more variable (both in space and time) than the southwest monsoon (SWM). Technically, co-efficient of variation of NEM is 25 per cent as against SWM’s 10 per cent which makes its forecasting even more challenging. The most important drivers of both these monsoon seasons are low pressures (known as monsoon low pressure systems), particularly those which form in the Bay of Bengal and move towards various parts of India. Every year, India witnesses many such monsoon low pressures some of which even intensify into monsoon depressions or even a tropical cyclone. Intensity-wise, tropical cyclones are the strongest systems followed by monsoon depressions and then monsoon low pressures. 
 
This year, around 11 low pressure systems were reported during the SWM season. Such monsoon low pressure systems usually have a life span of about 3-4 days and they bring heavy rain (often more than 100 mm in a day). This often inundates the areas through which they pass. But, when one compares Tamil Nadu low pressure system with other monsoon low pressure systems, some puzzling questions arise.  
 
How a weaker system brought so much rain? 
 
Comparison of monsoon depression in central India (left) and Tamil Nadu monsoon low pressure (Image courtesy: University of Dundee)
 
 
To bring out the aforementioned question, I have compared it with one monsoon depression which was observed in central India during September 16-19, 2015. As per reports, more than 150 places in Maharashtra recorded more than 100 mm rain in a day because of this monsoon depression but there isn’t any observation of rainfall as high as 475 mm of Chembarambakkam. Silod, a city in Aurangabad district of Maharashtra got 262 mm rain on September 17, while Vakadi (a place in Jalgaon district) got 220 mm rain. The low pressure system which brought rains in Tamil Nadu was very feeble (in terms of air pressure) as compared to the September system. The thumb rule is that the stronger a system is, the lower will be its air pressure and such a system will produce lot of rain under favourable conditions. From preliminary investigation, it looks like Tamil Nadu low pressure system had mean sea level pressure of around 1,008 millibar unlike around 997 millibar in the case of the September system. But naturally, maximum moisture content with the September system appears to be higher than that of Tamil Nadu system as reflected by total cloud water and total precipitable water data (which measures total atmospheric water vapour content in a vertical column of air between two levels). In the above image, compactness of clouds of monsoon depression (photographed on September 17) is another indicator of its strength over Tamil Nadu system (photographed on December 1). Thus, its definitely intriguing to see such a feeble system producing more rainfall than a powerful system. 
 
Internet is already flooded with several articles where without providing substantial evidences, people have linked this heavy rain event with factors like climate change or global warming. Below, we have a look at what could have made this low pressure system so powerful.
 
Variations in North-East Monsoon 
 
M Rajeevan, director of Indian Institute of Tropical Meteorology, Pune, and a senior monsoon researcher says that there is an increasing trend in the amount of northeast monsoon rainfall (NEMR) in the recent years and this rainfall posseses a lot of variability on inter-annual scale . But, he also agrees to the fact that it’s not normal for a feeble low pressure (which impacted Tamil Nadu) to bring so much rain. He says that the total precipitable water for this system was about 1.4 times higher than the normal levels. According to him, even the weather models couldn’t precisely capture the high rainfall associated with this system. He asserts that it will be too early to link this heavy rain event with climate change and more evidences will be needed to prove climate change’s role. 
 
21-year moving average of northeast monsoon (NEM) rainfall during 1901 to 2010. Dashed horizontal line indicates long-term average whereas recent increase in the rainfall is shown by red circle (Source: M Rajeevan et al 2012)
When asked about the influence of El Niño on Tamil Nadu rains, Rajeevan said that probability of getting a normal or above normal NEMR in an El Niño year is more than getting below normal NEMR. An analysis by Sreekala and others in 2011 states that there is 22 per cent chance of El Niño year being deficient NEMR as against 41 per cent chance of it being an excess NEMR year. So record El Niño in the Pacific is a sort of necessary but not sufficient condition to explain this year’s excess NEMR. Between 2001 and 2010, there were four El Niño events (2002, 2004, 2006 and 2009). In 2002 and 2004, NEMR was below normal but in 2006 and 2009, it was slightly above normal. Also, in this decade, the statistical relationship between El Niño Southern Oscillation and NEMR has weakened as per Rajeevan. So El Niño’s does have a role to play in this year’s NEMR but its role isn’t that straightforward. 
 
Role of vortex suspected 
 
Weather forecaster Rajesh Kapadia who writes a blog called "Vagaries of Weather" believes that Tamil Nadu heavy rains were caused by development of a vortex within the low pressure near surface levels. Vortex basically refers to rotating motion observed in a fluid (say air in this case). According to him, such vortex (or called vortices when in high number) developing within a low pressure can bring sudden heavy rain. Usually such vortex lasts till 3-4 hours after it’s formed but in rare cases (like July 26, 2005, Mumbai rain), it can last even for a day which then brings such a huge rainfall. As such vortices have high unpredictability, weather models don’t easily catch them which unnecessarily underestimates any rainfall forecast (say in the case of Chennai). 
 
A cocktail of weather conditions 
 
GFS model map of winds at 20,000 feet above ground simulating the situation on December 1, 2015 (Courtesy: earth.nullschool.net)
 
The atmospheric setup on December 1, 2015, appears to be more complex than was anticipated. It’s a cocktail of various atmospheric conditions. As can be seen from above image, a high pressure "H" (region of clockwise winds) anchored over eastern Bay of Bengal caused an inflow of moisture rich warm air from Bay of Bengal in Tamil Nadu. To represent this, I have used red dotted arrows. "C" refers to a cyclonic circulation (low pressure) which was observed east of Chennai. As monsoon has withdrawn from most parts of India, the air over central and northern parts of India is drier and colder than the warm moist air coming from southeast. Such an interaction of warm moist air from south-east and cold dry air from northeast crates a lot of instability in the atmosphere which is known to triggers strong thunderstorms. 
 
After a preliminary investigation, to me, Tamil Nadu’s heavy rain event seems to be a result of a Mesoscale Convective System (MCS) which was pretty intense given the above unstable conditions. Mesoscale in meteorology refers to a scale or horizontal extent of about 1-100 km. We know about thunderstorms and some of which we come across are single cell thunderstorms (meaning their influence is restricted to a certain area only). Such thunderstorms mostly pop up during the afternoon hours, produce bad weather and die. But, when such individual thunderstorms cluster together, they become organised on a larger scale and can last for several hours. Such a large cluster of thunderstorms is known as MCS which is observed in many parts of world like United States of America. In satellite or doppler radar images, such MCS can be linear or round in shape (appearing sort of round in Tamil Nadu’s case) and it often produces severe weather events such as heavy rains which lead to flooding. 
 
In Tamil Nadu’s case, role of MCS can’t be ruled out. In addition to this, development of vortices can’t be ruled out as well which will be clear after investigation of doppler radar images. In the case of the September monsoon depression, such a large instability in the atmosphere or a cocktail of atmospheric conditions weren’t observed. That’s how Tamil Nadu low pressure system is different from other low pressure systems and perhaps this caused it to produce so much rain.

Subscribe to Weekly Newsletter :

Comments are moderated and will be published only after the site moderator’s approval. Please use a genuine email ID and provide your name. Selected comments may also be used in the ‘Letters’ section of the Down To Earth print edition.

  • Very good analysis.

    However, it is a fact that in the past also occurred such rains in Nellore region in Andhra Pradesh and Chennai
    region in Tamil Nadu during the northeast monsoon season. Nellore received 444 mm of rain in 24 hours in October 1950, 357 mm in November 1936. Chennai received 262 mm in 1901. November height rainfalls are: 982 mm in 1915 for Nellore and 1088 mm for Chennai in 1918. By December, the system of northeast monsoon is more active along Tamil Nadu Coast over Andhra Pradesh Coast.

    Rainfall data is available since 1871 -- IITM published for sub-division-wise. Coastal Andhra sub-division presented very large variation in rainfall in the starting of the period and now the same is repeated. The annual rainfall presents a 132 year cycle. The below the average 66 year period started around 2001. The southwest mnsoon and northeast monsoon rainfall presents a 56 year cycles but in opposite direction. In the northeast monsoon rainfall presented steep change in the beginning of the 132 year cycle and the same pattern was repeated at the end of the cycle in the southwest monsoon rainfall. The present steep change in the northeast monsoon is repeated now. If Rajeevan would have extended backward to 1871, he would have seen this, probably.

    Dr. S. Jeevananda Reddy

    Posted by: Dr. S. Jeevananda Reddy | 3 years ago | Reply
  • Dear Dr. S. Jeevananda Reddy,
    Thank you for going through this analysis and for your comment.

    It's interesting to note 400mm+ rainfall figures from Andhra Pradesh also. Were they (Nellore's rain) related to any tropical cyclone? I have written this particular piece in the context of rainfall in Tamil Nadu (specifically the 400mm+ events in 24hr period) hence didn't investigate the rainfall in Andhra Pradesh. As far as the variability is concerned, I would like to have a look at the mentioned 132 years cycle. Is it possible for you to share any link (say a graph) showing this trend?

    Regards,
    Akshay Deoras

    Posted by: Akshay Deoras | 3 years ago | Reply
    • please send the e-mail ID and I can send it to you.

      Dr. S. Jeevananda Reddy
      E-mail ID: jeevanandareddy@yahoo.com

      Posted by: Dr. S. Jeevananda Reddy | 3 years ago | Reply
  • Good Post. But if you had observed carefully the year 2015 alone had 3 extreme rainfall events, 1 in Cuddalore district on 9th Nov when Neyveli recorded 45 cms in 6 hours then on 15th Nov when Chennai & Surrounding areas recorded extreme rains with many stations crossing 25 cms. The final one is 1st Dec. 2015. If one goes through the NEM rainfall pattern we have seen such instances happen in the past also like year 2005. The Nov. 24 hours all time record for Chennai is 45 cms on 25th Nov. 1976

    Posted by: Srikanth | 12 months ago | Reply