Wadia Institute to develop AI-based warning for glacial lake outbursts
DEHRADUN: City-based Wadia Institute of Himalayan Geology will develop an early warning system using artificial intelligence (AI) for glacial lake outbursts, snow avalanches, rockfall and flash floods, its director said on Tuesday.
The announcement comes after a deadly flood -- apparently triggered by a glacial burst -smashed through two dam projects in Tapovan, Chamoli district on Sunday.
Wadia Institute director Kalachand Sain said they were working on developing the early warning system for major glacier regions in vulnerable areas in Himalayas to save lives if Chamoli like incidents reoccur.
There are around 1000 to 1400 glaciers in Uttarakhand, said Sain.
“Under this initiative, we will set up ground observatories near accessible areas in glacier regions and monitor various parameters like temperature, rain, snow, wind speed, wind direction, local weather phenomenon and other related things.
“This data coupled with regular satellite imagery of the areas will help in developing a warning system for such Chamoli and Kedarnath like tragedies in the coming time. We will use the Internet of Things and artificial intelligence to make it more accurate and generate alerts in real-time,” he said.
The Internet of Things means a network of physical things that are embedded with sensors, software and other technologies for connecting and exchanging data with other devices and systems over the internet.
“Whenever there will be excessive rainfall or snowfall or related event, coupled with our
ground report and satellite imagery, we can make a model whether there are chances of any glacial lake outbursts or snow avalanches or rockfall which can lead to Chamoli like disasters. Based on our ground data and local weather data and satellite imagery, we will be able to come up with warnings for particular areas and alert authorities accordingly,” he said.
Sain said based on our ground assessment and availability of funding, the number of observatories to be set up will be decided. “The more observatories we have the better will be our warning system. And with the accumulation of data over the years, the warning system will get better,” he said.
Pankaj Chauhan, scientist and expert on use of artificial intelligence, who will be the main resource person in the project, said: “What we are actually developing is taking data from ground observatories, satellite imagery, water level sensors downstream, past data trends in local weather and climate near major glacier valleys like Kedarnath or Nanda Devi Biosphere
area, and feed the data to Internet of Things that will be calibrated to generate alerts to authorities concerned in realtime, based on thresholds we set for different glacier regions.”
“Suppose there is heavy rainfall or snow in some area. If it exceeds a certain threshold, we set for glacier region, an alert or warning will be generated about Chamoli like incidents and people downstream could be informed to move away from the river area and its banks. This system will help in saving lives,” he said.
On how artificial intelligence will be used in this project, Chauhan said: “Artificial intelligence will analyse long term data of various parameters like rain, snow and other things in particular areas and identify patterns or cycles over the years.
“These patterns will also help un in predicting whether a certain glacier area will receive excessive precipitation this year or not. And coupled with live data from ground and satellite imagery, the early warning system will keep improving over the years,” he said.