IIT Delhi’s AI model predicts normal monsoon in 2023
A new climate model developed by a research team led by Professor Saroj K. Yeshra of IIT Delhi has predicted that 2023 will be a normal monsoon year. It said the All India Summer Monsoon Rainfall (AISMR) of about 790 mm in the upcoming monsoon season means a normal monsoon.
4, and categorical data of the Indian Ocean Dipole (IOD) for the period 1901-2001.
Researchers at the Center of Excellence in Climate Modeling at IIT Delhi, in collaboration with IIIT Delhi, MIT USA, and JAMSTEC Japan, have developed a cutting-edge machine learning model for seasonal rainfall prediction.
The AI/Machine Learning model developed and tested has proven to perform better than the current physical models used for monsoon forecasts in the country. It has shown a remarkable expected success rate of 61.9% for the examination period 2002-2022. This depends on whether the model can predict AISMR within +/- 5% of the actual values observed each year.
The AI/ML model can be predicted months in advance, depending on Niño3.4 availability and IOD projections. It can be updated according to its development. Thus, data-driven models are flexible with respect to the inputs and can better capture the nonlinear relationships between monsoon drivers while being computationally less intensive.
A few people running these models on a personal computer within a limited time can provide a more accurate prediction of seasonal rainfall than the resource-intensive process involved in traditional physical models.
“This study is of great importance for the entire country, as accurate forecasting of the monsoon early on is pivotal for making critical decisions in various social and economic sectors, including agriculture, energy, water resources, disaster management and health,” said Professor Saroj. K. Mishra, PI, DST Center of Excellence in Climate Modeling and Professor, Center for Atmospheric Sciences, IIT Delhi.
Professor Mishra added that the data-driven techniques will be expanded to provide a country-by-state forecast of monsoon rainfall, making it more useful for regional applications.
The research team also includes Professor Tanmoy Chakraborty, Department of Electrical Engineering, PhD researcher Pankaj Upadhaya from IIT Delhi, Udit Narang and Kushal Juneja from IIIT Delhi, Professor Swadhin Behera from JAMSTEC Japan, and Popat Salunke from MIT, USA.