India launches indigenously built weather forecasting system for accurate predictions

The Centre on Monday launched indigenously built 'Bharat Forecasting System' (BFS) that will enable the India Meteorological Department (IMD) to provide more accurate and localised predictions of rains.

Advertisement

BFS has become the one-of-its-kind numerical weather prediction system in the world. During the launch of the BFS, Union Minister Dr Jitendra Singh said the development of the new weather model was spearheaded by four women scientists.

Developed by Pune-based Indian Institute of Tropical Meteorology (IITM), the BFS enhances India’s weather prediction resolution from 12 km to 6 km, enabling every village in India to access more precise and location-specific forecasts. The weather model has been in development since 2022.

M Ravichandran, Secretary, Ministry of Earth Sciences said, “This weather model is unique. The tropical region has a lot of kinetic weather events. We need a high resolution system to capture spatial changes. Globally, this is the first time we have implemented this model. Earlier, we used to issue one forecast for four villages. The BFS will enable us to issue separate forecasts for each of the four villages.”

With this, the extreme rainfall prediction events will be improved by 30 per cent. “The new BFS has a resolution of 6 km at the tropics and around 7-8 km resolution at the poles. The current range of our system is only around 12 km. This will remarkably improve our forecasting accuracy,” IITM Director Suryachandra Rao said.

He further said earlier weather forecasting models used to take around 12-14 hours to make accurate weather predictions over a particular area, but with the advancement in high-performance computing systems (HPC) which was launched last year under ‘Mission Mausam’, weather can be accurately predicted in 4-6 hours.

Singh said the system would help reduce crop loss and anomalies by improving extreme weather predictions — with a 30% improvement in extreme rainfall forecasts and a 64% enhancement in core zones, using real-time modeling based on the ‘Triangular Cubic Octahedral Grid model.’

The Tco (triangular-cubic-octahedral) grid is a specialised grid system used in numerical weather prediction and climate modelling, particularly in global models. It's designed to provide a higher resolution, especially over the tropics, compared to traditional Gaussian grids.