‘Will be able to predict cloudbursts sooner’

KOLKATA:India will be able to forecast extreme weather events such as cloudbursts at least two days in advance from 2022, the ministry of earth sciences (MoES) has informed a parliamentary committee.

“The ministry (MoES) stated that unlike cyclones, forecasting a cloudburst is very difficult due to the dynamics of the rapidly developing clouds over a very small area. However, they could make a probabilistic forecast on the possibility of cloudburst events over a specific area 48 hours in advance with an enhanced computing facility,” the committee stated in its report submitted to the Rajya Sabha earlier this month.

Cloudbursts are extreme weather events in which an area registers over 100 millimetres of rain in one hour. For instance, the 2013 floods in Uttarakhand were triggered by cloudbursts and had killed over 5,500 pilgrims and local residents, while many were reported missing.

India’s weather forecasting system depends on high-performance computing (HPC) system with a capacity of 10 PetaFlops (PFlops). Plans are afoot to increase the current HPC from 10 PFlops to 40 PFlops by 2022 and to 100 PFlops by 2024.

“We will be able to forecast extreme weather events such as cloudbursts at least two days in advance by 2022 after the computing facility is upgraded. Now, our forecasts have a precision up to 12 kilometres. The advanced computing system will help us give a more elaborate forecast up to 5 km,” said Madhavan Nair Rajeevan, secretary of MoES.

Experts said the advanced technology will allow authorities a period of at least two days to initiate pre-emptive actions which will help save more lives. Cloudbursts are being predicted — using Doppler Weather Radars and Satellite Data — two to three hours in advance.

“Science is not the limiting factor, but it is the computing facility. To forecast extreme events like cloudbursts and very heavy rainfall, which are highly localised, we need to increase our horizontal resolution to less than one kilometre and also increase our vertical resolution,” said KJ Ramesh, former director-general of the India Meteorological Department.

“An upgraded HPC system of 100 PFlops will help meteorologists to analyse the fast-changing microphysical characteristics of clouds and also improve the forecast,” he said.

In the changing climate scenario, central and northern India and the western Himalayas have become more prone to extreme rainfall events, whereas north, northwest and central India are prone to heatwaves. At least eight cyclonic storms formed over the Arabian Sea and Bay of Bengal last year. Usually, five cyclones develop in the Indian seas in a year.

We will be able to forecast extreme weather events at least two days in advance by 2022 after the computing facility is upgraded. MADHAVAN NAIR RAJEEVAN, secretary of MoES