In a groundbreaking move, the Ministry of Earth Sciences (MoES) is spearheading the integration of Artificial Intelligence (AI) and Machine Learning (ML) techniques to elevate the accuracy of weather, climate, and ocean forecasts. Various institutes operating under the ministry are at the forefront of this transformative initiative, ushering in a new era of meteorological research.
MoES has taken a significant step by establishing a dedicated AI and ML virtual center, entrusted with the crucial task of developing, testing, and refining various AI and ML techniques. This center also focuses on capacity building through workshops and conferences, fostering an environment of innovation and collaboration. To support the training and deployment of AI models, a robust computing environment and virtual workspace have been set up on a Graphical processor-based server at the India Meteorological Department (IMD).
Achievements and Outcomes:
- Enhanced Short-Range Precipitation Forecast:
- Improved forecast accuracy in 1-day, 2-day, and 3-day lead times with a notable reduction in bias.
- High-Resolution Urban Gridded Meteorological Data Sets:
- Developed high-resolution (300m) urban gridded datasets for temperature and precipitation.
- Time-Varying Normalized Difference Urbanization Index:
- Introduced a time-varying index with a spatial resolution of 30 meters, spanning the years 1992-2023.
- Very High-Resolution Precipitation Data Sets:
- Created very high-resolution precipitation datasets, crucial for verification purposes.
- Deep Learning for Precipitation Nowcasting:
- Exploring a Deep Learning approach for precipitation nowcasting using data from Doppler Weather Radars (DWRs).
MoES envisions a future where weather and climate forecasts will be driven by a hybrid technology that combines AI/ML models with traditional numerical weather prediction models. The encouragement for institutes under MoES to leverage AI and ML technological advancements is an ongoing commitment. In line with this, MoES is dedicated to enhancing High-Performance Computing (HPC) infrastructure. The implementation of AI and ML-based data-driven modeling is deemed essential for generating species-specific Potential Fishing Zone (PFZ) advisories for fishermen across coastal states.
The Union Minister of Earth Sciences, Shri Kiren Rijiju, highlighted these developments in a written reply presented in the Rajya Sabha today. The strides made in the realm of meteorological research through the infusion of AI and ML stand testament to the commitment towards precision forecasting and technological advancement in Earth Sciences.
Posted On: 21 DEC 2023 4:11PM by PIB Delhi
