Deadly thunderstorms test India’s weather warning system

With yet another round of thunderstorms killing over 60 people by mid-May in northern India following a similar event involving 125 deaths earlier in the month, scientists are calling for better forecasts, dissemination and actions based on a clearer understanding of their inherent uncertainty.

In the early May events, Uttar Pradesh, Rajasthan and Punjab accounted for 100 deaths, while Andhra Pradesh in the south lost 14 people in 41,000 lightning strikes reported in the state. In the later event, most of the deaths were in Uttar Pradesh, India’s most populous province.

Most casualties were reportedly caused by collapse of buildings in intense downward wind called downbursts, sandstorms, and thunderstorms. While storm warning and building safety issues are very much in the agenda of government disaster management agencies, lightning forecasts, dissemination, and risk reduction measures require better attention, scientists pointed out.

The India Meteorological Department (IMD) had forecast thunderstorms before both events. Madhavan Nair Rajeevan, secretary in the Ministry of Earth Sciences, and an accomplished scientist specialising in extreme weather and monsoon rains, has noted that the forecasts were timely and accurate.

Dust storms, thunderstorms and lightning activity are common in North India during April and May. Some storms gain severe intensity with devastating lightning, Rajeevan said. Lightning rapidly discharges electrical energy in the atmosphere within a cloud, between clouds, or between a cloud and the ground. It is the ground hits that kill or injure people.

Preceding the monsoon that starts in June, the prevailing weather pattern leads to dry and hot weather conditions in north India. However, these patterns change when eastward moving weather systems called Western Disturbances strike this region.

“These systems are areas of lower atmospheric pressure (and thus moisture-laden) and thus cause precipitation (rain, snow, hail etc.),” Akshay Deoras, an independent weather forecaster, explained to indiaclimatedialogue.net. “Recently, we have been seeing reduced frequency of Western Disturbances during the winter season, but increased frequency and intensity during India’s summer period.”

Thunderstorm activity

The thunderstorm activity on May 2 was triggered by one such western disturbance. The atmospheric instability was increased as a result of high surface temperatures over the plains of North India. In fact, the instability and energy required for thunderstorms were significantly high on that day. This is also due to presence of dry air in the mid-tropospheric levels. That made the thunderstorms more intense.

The damaging storms got triggered close to the arid regions located at the Rajasthan-Haryana border by afternoon. “Initially, they were small in size and of single cell structure, but in a few hours, rapidly evolved to form a system called Mesoscale Convective System (MCS). Such MCS are known to produce damaging winds, typically due to the presence of gust fronts,” Deoras said.

The troposphere is the lowest region of the atmosphere, up from the Earth’s surface to about 10 km. MCS is a collection of thunderstorms that act as a system. An MCS can spread across an entire state and last more than 12 hours. A thunderstorm is a rain shower with thunder and lightning.

A gust front is the leading edge of rain-cooled air that clashes with warmer thunderstorm inflow, characterised by a wind shift, temperature drop, and gusty winds out ahead of a thunderstorm. Gusts denote rapid fluctuations in the wind speed with a variation of 10 knots (18.52 kmph) or more between peaks and lulls. The speed of the gust is measured as the maximum instantaneous wind speed.

Short lead

Thunderstorms could be forecast two days in advance, while they are still forming, though their intensity and the exact location could be predicted only closer to the event, leaving very short lead time of say, two hours, scientists said. Sridhar Balasubramanian, a meteorology expert and associate professor at the Indian Institute of Technology-Bombay, told indiaclimatedialogue.net that radar observation could track and predict the direction of thunderstorm and issue local forecasts even within that lead-time and save lives.

Now the challenges are three-fold — technical measures that are needed for such localised forecasts, the expertise in interpreting and communicating the event, and mechanism to disseminate forecasts.

For the forecast, IMD is well placed with a line of Indian Space Research Organisation weather satellites and a nationwide system of weather stations with trained scientists and forecasters. They detect thunderstorms through satellites and radars.

Satellites take snapshots of the Earth from space, locating clouds, and detecting temperature differences. They watch for fast growing clouds, a tell tale mark of a potential thunderstorm. Clouds with cold tops, for instance, are usually very high up, and could be a thunderstorm. The satellites track these clouds to see where the storm hits next.

On the ground, radars play and important role. IMD has a radar network, offering a range of forecast services and products. IMD’s modernisation includes expanding its Doppler radar network. Doppler radars work by sending out electromagnetic waves that bounceback from moving objects in the air such as rain clouds.

Analysing the way they bounce back, forecasters determine how heavy the rain is, where it is going, and whether there is hail. They can even show how the wind is blowing near and inside a storm system. As a US National Severe Storm Laboratory (NSSL) primer explains: “This is helpful in understanding what kinds of hazards the thunderstorm might have (tornado, microburst, gust fronts etc.) associated with it. It also helps us understand how the thunderstorm is feeding itself.”

Nowcasting

The Doppler radars can aid forecasts two or three hours prior to severe weather events. The process is called nowcasting, a technique for very short-range forecasting that maps the current weather, and uses an estimate of its speed and direction of movement to forecast the weather a short period ahead.

Though IMD officials are confident about their forecasting capabilities, independent experts said the nowcasting could be vastly improved. “The forecast of thunderstorms can be improved if we have a higher resolution weather model that is updated more frequently than the existing models, as capturing localised convection initiation is crucial,” Deoras pointed out to indiaclimatedialogue.net.

Convection denotes upward atmospheric motion because of surface heating, lifting moisture in the air along with it. Rain and thunderstorms are the result of convection. Still the physics of thunderstorms is not properly understood and this acts as an issue in improving the models, Deoras said. Inside these systems, complex processes and interactions keep on happening on very short time scales.

Even the best forecasts are useless if people do not get them in time. Earlier this year, disaster experts discussing Cyclone Ockhi at a conference in Thiruvananthapuram pointed out that India’s storm forecast dissemination system needs drastic changes. Sometimes radars are not maintained properly, and satellite images and weather information are not readily available to the public in an accessible manner.

IMD officials blamed the disaster messaging system of the government that depends on several levels of bureaucracy that are not necessarily very efficient in mass communication even with all the technologies such as instant access to wireless and mobile phone networks.

The same debate continues in the aftermath of the thunderstorm events as well. “As you have seen, most weather warnings — when they are issued at all — are sent to the heads of various departments, mostly bureaucrats, who simply have no idea what to do with them,” Sajan Venniyoor, a community media expert and former Prasar Bharati official based in New Delhi, told indiaclimatedialogue.net. “Thunderstorms and dust storms killed over a hundred people across north India in the first week of May this year, and the centre blamed the IMD, pointing out that its forecasting mechanisms had not prevented the deaths. The weather agency replied, sure, it had issued a warning. It had informed the chief secretaries of Rajasthan and Uttar Pradesh about the impending storms. Now, this is a great system for saving the lives of chief secretaries, but it really doesn’t help the public.”

Experts at a meeting in Thiruvanathapuram, the capital of Kerala in southern India, stressed on the need for robust community communication systems, including FM radio with options for listeners to provide feedback. Citizen’s feedback is a major input for nowcasting even at neighbourhood levels. UK Met Office weather observations website (WOW), for instance, encourages citizens to share current weather observations, including severe weather information, from all around the globe.

Rural context

In the Indian rural context, FM radio and mobile phones could be the start of a robust mechanism to share forecasts. In the US the NOAA Weather Radio (NWR) serves as the “Voice of the National Weather Service,” a network of radio stations broadcasting continuous weather information direct from a nearby National Weather Service office.

Venniyoor suggested that there is a possibility of linking existing local FM stations and new community radio stations into a network to give reliable highly localised forecasts and prepare people for impending disasters. “So we need local radio (yes, community based radio) that will not only monitor weather conditions, but will also be in constant touch with the IMD and other agencies, and pass on life-saving information to the community.” The idea is to make listening to weather forecasts a part of daily life — with clear messages and advisories slipped into radio programmes along with local music, neighbourhood news and talk shows.

The next step is ensuring that people take forecasts seriously, as that is the key to disaster risk reduction, as Dominic Kniveton, professor of climate science and society at the University of Sussex, puts it. Even with a good forecast dissemination system, public responses are mediated by perceptions of forecast accuracy.

“Forecasts are inherently uncertain and inevitably influenced by many factors,” he told indiaclimatedialogue.net in an interview. “It is like throwing pinball, you hit it right, but not always.” Kniveton, an atmospheric physicist by training, said that people need to understand this element of uncertainty. That means people need to work with possible scenarios offered by the best forecast and tune their responses accordingly, and know what to do under each scenario. India’s National Disaster Management Authority (NDMA), for instance, has recently issued a set of dos and don’t to follow before, during and after thunderstorms.

Experts say that we can mitigate the damage from thunderstorms significantly by following three activities — rapidly updating radar and satellite information, deploying an efficient real-time weather warning system with dissemination mechanisms, and generating widespread awareness of safety measures.