Research

Atmosphere & Environment Sciences

Title :

A data-driven machine learning approach to dynamical balance in the tropical and extra-tropical climate and extreme weather events

Area of research :

Atmosphere & Environment Sciences, Earth

Focus area :

Climate Change

Principal Investigator :

Prof. Jai Suhas Sukhatme, Indian Institute Of Science, Bangalore, Karnataka (560012)

Timeline Start Year :

2024

Timeline End Year :

2026

Contact info :

Details

Executive Summary :

This proposal aims to investigate data-driven dynamical balances in large-scale tropical modes, extratropical Rossby Wave Packets, and their interactions with the background jet stream. Understanding the dynamical balance underlying large-scale tropical systems is crucial for predicting tropical weather and understanding the tropical climate. The moisture mode theory has been useful for describing large-scale, slowly evolving intraseasonal modes, but there is still a lack of conceptual universality. The life cycle of Rossby wave packets (RWPs) plays a significant role in modulating weather, leading to persistent anomalous weather and extreme events. Recent advances in machine learning (ML) have led to newer techniques with greater efficacy in representing and predicting the evolution of nonlinear dynamical systems. The project will use a hierarchy of models to study the data-driven dynamical balance in large-scale tropical modes and extratropical extreme events associated with nonlinear RWP evolution. The goal is to improve the fundamental understanding of these atmospheric phenomena and explore dynamically oriented dimension reduction techniques to provide more insightful descriptions of large-scale tropical oscillations.

Co-PI:

Prof. Joy Merwin Monteiro Indian Institute Of Science Education And Research (IISER) Pune, Maharashtra (411008)

Total Budget (INR):

30,60,760

Organizations involved