India Meteorological Department has been at the centre of India’s efforts to minimise risk and damage from natural disasters through its early warning systems. As another powerful cyclone hit the Indian coast this week, the forecasts and warnings provided by IMD proved crucial to saving thousands of lives. IMD director general K J Ramesh speaks about Cyclone Fani and India’s success in the last few years in enhancing its capabilities to deal with natural disasters like cyclones, heat-waves, and floods. He also explains why IMD’s monsoon forecasts are the best that any weather agency offers, and why private forecasters are no competition.
AMITABH SINHA: In recent years, states have been successful in minimising damage caused by cyclones. How has this been achieved?
This is because response actions and preparatory precautions are now systematically implemented. The target of our team is to ensure there is no casualty. Across the world, there is only one example of a cyclone with zero casualty, in Hong Kong last year. During Cyclone Phailin (in 2013), less than 50 people died. These were people who refused to move to relief camps because they did not want to leave behind their assets.s
MONOJIT MAJUMDAR: Will Cyclone Fani impact this year’s monsoon?
The monsoon has not been formed yet. The earliest possibility of monsoon setting over the South Andaman Sea is around May 18… This will need at least two to three weeks. So monsoon is unlikely to be affected.
MONOJIT MAJUMDAR: Why do we see so many tropical storms in the Bay of Bengal and not in the Arabian Sea?
This is due to the warmer ocean water in the Bay of Bengal compared to the Arabian Sea. In recent times, however, we have seen almost the same temperature (in the Arabian Sea as well). This is because of global warming. Oceans are now warming up significantly… Take the example of (last year’s) Cyclone Luban. It travelled through the Bay of Bengal, crossed the peninsula and entered the Arabian Sea. It then intensified again and travelled up to the coast of Oman. The cyclone was alive for 14 days.
LIZ MATHEW: What is the reason for such frequent cyclones?
We have two principal cyclone seasons in India — in April-May over the Bay of Bengal, and October-November and partly December. This happens because of the movement of the sun from the Southern Hemisphere to the Northern Hemisphere. It leads to the formation of a low-pressure area called the Intertropical Convergence Zone. Wherever heating is maximum, based on the sun’s inclination, there will be a seasonal belt of low-pressure areas. Along these areas, (cyclone) systems develop. So when the sun moves from equator to the Tropic of Cancer — it goes up to 23.5 degrees north by June — we have the April-May season… A similar process will again take place when the sun moves back to the Southern Hemisphere in October-November after monsoon. This is the time when both the basins are warmer by 27 degrees, a prerequisite for cyclone formation or intensification. The favourable conditions in the ocean and atmosphere work in a contributory manner. That is why cyclones that develop during the season are always stronger.
In the case of Cyclone Fani, warm ocean and land surfaces because of the early heating helped sustain and intensify its life cycle.
HARISH DAMODARAN: Will there be a cooling of landmass because of Fani?
The cyclone Fani’s path was through the coastal districts of Odisha, followed by a few districts of Bengal and then onto the hills in Meghalaya through Bangladesh. Had it been moving towards Central India — East Rajasthan or West Madhya Pradesh — there could have been a possibility (of disruption of the monsoon cycle). But even then, it would not have affected (the monsoon) because heating in northern India will continue till the end of June.
Also Read | Cyclone Fani: Puri confronts the devastation
AMITAVA CHAKRABORTY: How well are states equipped to deal with the disruptive effects of cyclones. What more needs to be done in this respect?
The first pre-requisite is that a good early warning system has to be there for any (disaster management) action to be planned or strategised or implemented. That has already happened. Since the decades of 1970s, if you see, about 70,000-80,000 people used to get killed during a cyclone. Now we have come to a stage where we can save most of these lives. So a warning system is one big contribution… But, you need a good response system in states. That has happened after the establishment of the National Disaster Management Authority (NDMA).
Twenty-four major hazards have been identified by the NDMA, both natural and man-made, and guidelines for their management prepared after consultations with states. These guidelines now act as the basis for dealing with any hazard.
For cyclones, states were given grants by the Central government through the National Cyclone Risk Mitigation Project. This has helped in the construction of multi-purpose cyclone shelters. The shelters built earlier were used only during cyclones and were not properly maintained throughout the year. But the multi-purpose cyclone shelter, which has now been introduced, is a community asset for panchayats… While they collect some funds, the government also provides funds for its upkeep. They have become community centres rather than cyclone shelters.
As these shelters are located in remote areas, the approach roads were a problem. After NDMA guidelines were introduced, all-weather approach roads have been laid in all 72 coastal districts. Such measures have helped us significantly increase our ability to manage impact. That has changed the manifestation of impacts.
Also, community-level awareness has increased. People have identified the most vulnerable structures, particularly those with mud walls and thatched roofs. With PM Awas Yojana and similar schemes in the last three decades, we have been able to minimise the existence of those vulnerable structures.
AMITAVA CHAKRABORTY: If Fani was hitting a city, would it have been manageable? How would you have managed that?
Managing a city is easier than rural areas. City structures are relatively stronger. And now with social media platforms, they can very quickly disseminate information. People can safeguard themselves more easily.
AMITABH SINHA: But in the cities you do have the problem of urban flooding, which has proved not easy to manage. There are still problems in dealing with several other kinds of disasters.
See, urban flooding is not because of only intensive rainfall but equally due to unregulated development, not respecting the flood plains and not allowing enough carrying capacity for natural drains. Poor development priorities is how you can describe them. Urban planners in the past are to be blamed, to be frank… Multi-storey apartment complexes have come up, and the carrying capacity of drains has not been increased. With that, solid waste management-related problems come up. Green areas are carpeted with concrete in many cities, parks are not fully developed to act as extra-water absorbers or for water recharging. In Mumbai, for example, earlier, 20 to 30 cm of rainfall in 24 hours did not cause flooding. This has changed in the last 10 years.
In terms of other extreme events, considerable success has been realised in the case of heat waves. In 2015, we had around 2,800 mortalities. Subsequently, we started hot weather season outlook, followed by weekly identification of possible hot spots. We attempted heat wave prediction. The NDMA came into the picture, calling state governments, preparing lists of do’s and don’ts, state- and city-level plans, exposure of people was reduced through advance information. In 2018, we only had 28 deaths. This is one example that if the intent is strong, response can be effective.
Last year onwards, we have started working on thunderstorm prediction… Since April 1, we are giving 24-hour advance lightning predictions. This science (for lightning prediction) in fact emerged only eight months ago, globally… so we are on a par with all the other centres in the world. Simultaneously, we are setting up lightning detection networks all over the country. Whenever lightning activity increases, SMSes go to people.
HARISH DAMODARAN: How about forecasting droughts and monsoons? IMD predictions have not been pretty accurate there.
There is a little more to it. We give one value (of expected rainfall in monsoon season) for all of India, first in April and an updated one in June and verification (of the prediction) is done with the actual rainfall at the end of the season. Take the case of last year, for example. We gave 97% (figure in April) and maintained that in the second stage also. Ultimately, the rainfall (over the entire country in the monsoon season) turned out to be 91% (of normal). This happened because of two factors. A good monsoon means a good distribution of rainfall through the four months. If rainfall is good, two things get manifested on the ground — inflows into the reservoirs, which solve the irrigation problem, and the second is increase in acreage of rain-fed crops week after week during the sowing period up to the end of July. Both these factors were higher than 10-year averages in 2018. Because of that, food production in 2018-19 was only 1.26 per cent lower than the previous year when India had the highest food production ever. So, if you look at these two factors, you cannot say monsoon was not good last year.
However, in the Northeast, rainfall reduced by 27% of its normal. That made the all-India value come down. In real sense, if distribution had not been good, we would have seen drought-like conditions.
Any such abnormal variability, particularly deficient rainfall, can be captured by models we have now for seasonal and monthly predictions. These models will take care of global factors like El Nino, or unusually high durations of snowfall, and other six global factors that control monsoon variability. If any one of these combined effects is contributing significantly to altering monsoon rainfall, we will be able to capture it. Our ability to capture the distribution of rainfall has improved and it is making our predictions better.
AMITABH SINHA: Sometimes it seems the IMD itself starts believing the expectation that some people seem to have — that it is somehow the IMD’s responsibility to deliver good rains. And therefore, sometimes, there seems to be a reluctance on the IMD’s part to give bad news out.
Not like that. This season, for example, from November to January, El Nino was picking up, and across the world, it was believed a strong El Nino would emerge and that would be detrimental for the monsoon of 2019. But March onwards, El Nino weakened. None of the models captured the decline in Pacific warming.
AMITABH SINHA: Probably the only time in recent years that the IMD gave bad news in the first forecast itself was in 2015. There seems to be a tendency to start closer to normal, and then, as things become more clearer, bring the forecast down, if that is what is becoming evident.
See, monsoon itself has a 30-year cycle… We are out of that lower ebb of the monsoon. Before 2015, we did not have many above normal years. We are now entering in the next phase of 30-year variability pattern. That is why we are seeing good distribution of rainfall. But now, we also make 15-day predictions, these compute only rainfall anomalies… Then we have started adding probability to those rainfall predictions since last June. These things are making an impact on agriculture and water resource usage.
RAVISH TIWARI: Is it technically possible to give monthly and regional distribution of rainfall in April itself?
Right now, we are only able to capture ocean and atmosphere interactions, for our monthly and seasonal rainfall predictions. For higher perfection, we need to account cryosphere interaction as well. Sea and ice interactions, and aerosols-cloud interactions. These are not treated very robustly as of now in the climate models currently being used across the world. That framework is called earth system modelling. There are at least five interactions that need to be incorporated into our models. That science is slowly emerging. We have this kind of modelling available for climate change scenarios at the Indian Institute of Tropical Meteorology (IITM) in Pune, but that is still at a very coarse, 100 kmx100 km resolution… Science is available, but we need to upgrade our capabilities.
RAVISH TIWARI: GPS has been monetised by Ola and Uber and several others. You have expertise that can be monetised in several sectors. Has it occurred to the IMD to monetise that expertise?
You may not be aware that our aviation safety services are paid services. To 70 airports of India, we provide 24×7 aviation safety services. And we have set up our own observing systems. Earlier, we used to get about Rs 40-50 crore per annum from airports. Today it has gone up to about Rs 130 crore per annum. Whatever we earn goes to the Consolidated Fund of India. We don’t make profits from these. But it is a measure of the commercial value of our services. And it is growing regularly. On the agriculture front, we provide district-level agriculture advisory services, covering four major crops. There are 22 major crops in different districts… This service, according to one study, potentially contributes Rs 50,000 crore per annum to our GDP from our existing services only. When this covers all the 22 crops, this can go up to Rs 3.48 lakh crore per annum. This can also give you a sense of the commercial value of our services.
ANOOP PHILIP: Do you think the IMD faces a challenge from private weather forecasters?
No challenge at all. You need a lot of research backup to improve the formulation of the models or treatment of physical processes into those models… We have research institutions like the IITM and the National Centre for Medium Range Weather Forecasting in Pune, where we have 200-250 experienced scientists working… Last year we invested about Rs 420 crore into High Performance Computing. And, as far as monsoon goes, the IMD’s prediction is the best. No global agency understands the Indian monsoon better than us. The agencies in the US, Europe or Australia do not understand the basic characteristics of the monsoon model very well.
None of the private organisations has this kind of research back-up, within India or outside. They, at the very best, take public domain products, create some new infographics, do some value additions, and try to commercialise to make revenue out of it. But at the competence level, I don’t think there is any competition from private organisations.
AMITAVA CHAKRABORTY: Do we have a good talent pool of atmospheric scientists in the country?
There are about 10 institutions, universities, IITs that have centres or departments in atmospheric sciences, or earth sciences. Five-year integrated courses are also offered at some IISERs (Indian Institutes of Science Education and Research). Three IITs are offering MTechs in courses like atmospheric science and technology, ocean sciences… So there are 15-20 institutions from where students come to us.