1 day Ago By Akanksha Mishra
New Delhi: Earthquake predictions on social media have become a headache for Indian government officials, even when they occasionally get it right. On 28 February this year, a little-known X account called Epic (Earthquake Research and Analysis) predicted that a 6.5 magnitude earthquake would hit the Mandalay region in Myanmar.
Exactly a month later, a 7.7 magnitude earthquake occurred in Myanmar’s Mandalay—the region predicted by the X post—and killed more than 3,000 people. The X account is run by a Hyderabad-based research centre called Seismo Research and Development Center, and it was their 22nd ‘correct’ earthquake forecast since 2020, out of 100 total forecasts.
It isn’t just Seismo, though. India is seeing a rise in self-proclaimed earthquake forecasters, who use everything from animal behaviour to solar radiation to anticipate the next big shake. They come armed with public X accounts, unverified methodologies, and a promise to succeed in predicting earthquakes hours or days in advance.
On 31 March, days after the Myanmar earthquake, some officials in the Department of Science and Technology (DST) and the National Centre for Seismology (NCS) received a message warning them of an incoming earthquake on the India-Nepal border. The notice was from a company called Vikash Geosensing, run by a certain Vikas Kumar, saying there were ‘signals’ indicating a 5–7 magnitude earthquake in the next 24 hours and that the vibrations would be felt as far as Delhi. Needless to say, no earthquake occurred on 31 March in the region.
But it did put the NCS on high alert against fake earthquake notices. In a public post pinned to the NCS X account on the same day, Dr O.P.
Mishra, director, NCS, wrote a strongly worded message calling Vikas Kumar “irresponsible” for spreading “baseless news” to the public. “Throwing around 200 different earthquake predictions and getting a few right is a coincidence, not a science. If you have figured out the process and can really ‘predict’ an earthquake, you should be right every time,” says Mishra.
Also Read: Nursed to health after fall in India, Eurasian Griffon soars over Afghanistan, back on migratory path A ‘correct forecast’, according to Siva Sitaram, the founder of Seismo Research and Development Center, is when an earthquake happens within six months of their prediction in a similar location. It isn’t an exact science, because the Myanmar earthquake happened one month later, 170 km away, and 1.2 magnitude higher than Seismo’s prediction.
“22 correct predictions out of 100 — that is a 22 percent success rate and not bad at all,” says Sitaram, who has been interested in earthquake prediction ever since the 2004 earthquake and tsunami in the Indian Ocean. Sitaram refuses to share his methodology until he has conducted further research, but he’s certain he is onto something. While he hasn’t published any papers yet on his work, he claims to use 20 different environmental parameters to determine the occurrence of earthquakes.
The X account, he says, is just to record the timestamps of his predictions, to then corroborate them once an earthquake actually occurs. Vikash Geosensing, on the other hand, is a two-year-old start-up incubated at IIT Delhi and run by 24-year-old Vikas Kumar. It is building a machine called the ANDSS, and Kumar claims that it can sense underground signals of incoming earthquakes through a ‘bio-organic filter’ that he has created.
Built in 2018, the ANDSS is a machine made out of red cardboard that can signal when an earthquake is coming. At first glance, it looks a bit like a docking pod on a spaceship, but inside it has over a hundred wires and test tubes, and it works on the principle of ‘animal behaviour.’ “Before an earthquake, there are low frequency electromagnetic waves that can only be detected by animals — especially wild cats,” explained Kumar.
“I have built a bio-organic filter that mimics a cat’s skin and can detect these electromagnetic waves. If you connect the machine to the earth, it can detect signals within a 300-km radius.” He says that using his machine, he had even predicted the 2023 Turkey earthquake from his home in Dwarka, even though it was beyond the 300-km radius.
“When earthquakes are so strong, you can feel the signals,” he says. In a similar vein, Sitaram too claims to have predicted the 7 January earthquake in Tibet and Nepal, even though his prediction came two months earlier in November 2024. Sitaram says that he is in the midst of perfecting his process and will very soon have it down to the tee.
Neither Kumar nor Sitaram has published a paper or study detailing exactly how their earthquake detection process works. While there are existing studies on animals — including cats — being able to detect ‘P waves’ before an earthquake, these usually happen minutes or hours before earthquakes. The United States Geological Survey , one of the foremost earthquake monitoring authorities in the world, has also stated that such behaviours are not consistent or reliable.
“This isn’t the first time. There have been many warnings to Vikas Kumar to stop sending such notices to public officials — he hasn’t been verified by the NCS, and all he does is spread panic amongst people,” OP Mishra told ThePrint. “Moreover, his notice did not even make sense scientifically.
The signals that he spoke about — there’s no meaning to it!” Kumar, however, says he was vindicated because an earthquake did end up occurring in Nepal — four days later, on 4 April . Unlike his prediction, the earthquake was just a 5 magnitude one, but Kumar still says that it is proof his machine works. Kumar is not the first person to use animal behaviour as a predictor for earthquakes.
In the world’s first so-called successful earthquake prediction in China in 1975, too, the authorities relied on animal observations, among other phenomena. The ‘science’ of earthquake prediction has been around since 1975, when the Chinese authorities said they had successfully predicted the Haicheng earthquake of 7.7 magnitude and evacuated the city in advance.
The authorities had indeed noticed a lot of seismic activity in the days leading up to the earthquake, and observed changes in groundwater levels , as well as abnormal activity in snakes that came out of hibernation right before the earthquake. “When the Haicheng earthquake happened, the world really thought we had figured out how to predict earthquakes — it was a major moment,” said Kusala Rajendran, seismologist and former professor at the Indian Institute of Science Bengaluru. “That lasted for one year.
” The same Chinese authorities failed to predict the Tangshan earthquake in 1976, which killed over 240,000 people. After much academic debate, most seismologists agreed that the 1975 ‘prediction’ was just a lucky fluke. “The world has been trying to predict earthquakes for a long time, but all we have come up with is coincidences and false alarms,” says Rajendran.
“Animal behaviour, radon gas, electric activity in the ionosphere are all precursors to earthquakes for sure, but they are not definitive indicators and should not be treated as such.” Rajendran also agreed with Mishra that if one has a system to predict an earthquake, they should be able to do it in every case and not just randomly. That is what will make scientists take notice and be considered valid, she says.
Kumar’s ANDSS, which currently sits in a small office in the IIT Delhi research park, is also connected to an intricate map of India retrofitted with tiny light bulbs. As he explains it, whenever the ‘electromagnetic waves’ are captured by the machine, the map lights up, displaying the exact location from which the signals are coming. “I have predicted over 2,000 earthquakes since 2015, and only 3–4 of them have been incorrect,” says Kumar.
“Yet no one is willing to believe me.” Since Kumar did not record his predictions like Sitaram, there is no way to independently verify his claim of predicting over 2,000 earthquakes. But he says that he is willing to prove and get his technology verified only if a government organisation supports him.
“I don’t have a university degree which is why people don’t take me seriously. But I ask them — I ask NCS to first test my machine, see whether it is valid, and then refute it,” said Kumar. “How can they refute it without even trying?” NCS’s Mishra, however, has a simple answer for all private earthquake forecasters.
“If you say an earthquake will occur in a seismically active zone like Nepal which is in the Himalayas, or Myanmar’s Sagaing fault, then you are not predicting but anticipating an earthquake,” says Mishra. “We already know the seismic zones and faults in the world, we have mapped them thoroughly. If you actually have anything new to contribute, you will be able to pinpoint the exact date and time.
” Kusala Rajendran, however, admits why earthquake forecasting is still a ‘fad’. The quest has challenged the scientific community for a long time and continues to do so, she said. “There’s something evasive about earthquakes and their occurrences that is not like other natural events — even with all the knowledge of a fault line, past seismic activity and sensors on the Earth, it is never guaranteed that you will predict an earthquake,” adds CP Rajendran, seismologist and adjunct professor at the National Institute of Advanced Studies, Bengaluru.
Quoting the example of the Parkfield Experiment , Kusala Rajendran explained how geologists in the US had tracked seismic activity in Parkfield, California along the San Andreas fault in 1985. The region had seen six previous earthquakes, at an interval of 22 years each, and scientists were confident that the next earthquake would occur in 1985. They issued an official prediction, stood waiting with their recording instruments, but the earthquake did not occur until 2004.
“Even in a fault as well-studied as San Andreas, even with scientists globally predicting an earthquake, it did not end up happening — what does that tell us about this phenomenon?” said Rajendran. “There’s still too much humankind needs to learn about earthquakes.” While the ANDSS or Sitaram’s Seismo might not be Rajendran’s answer to earthquake prediction, she does believe that it is possible somewhere in the future.
“If we go by data collection and data crunching and are able to establish a pattern that can foresee earthquakes, we might have a real shot,” says Rajendran. “This much data is a lot for humans to handle so AI and machine learning could be the answer right now.” “But the answer lies in data.
It has to.” (Edited by Radifah Kabir) Also Read: Zero waste served Indian-style—why the future of food sustainability lies in the past var ytflag = 0;var myListener = function() {document.removeEventListener('mousemove', myListener, false);lazyloadmyframes();};document.
addEventListener('mousemove', myListener, false);window.addEventListener('scroll', function() {if (ytflag == 0) {lazyloadmyframes();ytflag = 1;}});function lazyloadmyframes() {var ytv = document.getElementsByClassName("klazyiframe");for (var i = 0; i < ytv.
length; i++) {ytv[i].src = ytv[i].getAttribute('data-src');}} Save my name, email, and website in this browser for the next time I comment.
Δ document.getElementById( "ak_js_1" ).setAttribute( "value", ( new Date() ).
getTime() );.
