SHI Collaboration Profiles

Profile pages for Sustainable Horizons Institute SRP 2025-2026 Project Leaders


Agniv Sengupta

Agniv Sengupta

University of California San Diego

Scripps Institution of Oceanography

Biography

Dr. Agniv Sengupta is a Sr. Computational Research Scientist and the Machine Learning Team Lead of the Center for Western Weather and Water Extremes (CW3E), Scripps Institution of Oceanography at UC San Diego. His research interests involve the prediction of high-impact weather events using artificial intelligence and machine learning. His current projects focus on improving the prediction skill of weather (0-10 days), subseasonal (1-6 weeks), and seasonal (1 to 6 months) forecasts in the Western United States. This involves exploring innovative algorithms and approaches, advancing models for predictions across multiple timescales, and developing decision-support tools and forecast products in coordination with stakeholders. Dr. Sengupta earned his Ph.D. (2020) and M.S. (2016) in Atmospheric and Oceanic Science from the University of Maryland College Park, and was subsequently a postdoctoral scholar (2020-21) at the NASA Jet Propulsion Laboratory (JPL) prior to joining CW3E.

SRP Project Title

AI Models for the Prediction of Extreme Weather Events

NAIRR Project

Development of AI Data-driven Models and Very Large Ensembles for the Prediction of Atmospheric Rivers and Extreme Weather Events

Topical Areas

Artificial Intelligence and Intelligent Systems; Atmospheric Sciences

Abstract

Accurate weather forecasting has traditionally relied on numerical weather prediction models, which require significant computational resources. In this context, artificial intelligence (AI) models have revolutionized the domain of weather prediction in the past 2-3 years, emerging as computationally efficient alternatives. As part of our NAIRR Pilot project, we are developing such AI weather modeling and prediction capabilities over the North Pacific and the western United States, specifically for extreme weather phenomena such as atmospheric rivers (ARs). In this work, we utilize state-of-the-art AI advancements, including graph neural networks and Diffusion-based Generative AI methods. Additionally, we employ these models to generate a large number of realizations of future weather, enhancing the tracking of AR storms from their genesis over the Pacific to their impact over the U.S. West Coast.

Desired Skills

Interest in AI/ML, Meteorology, or Climate Science Experience with, or a desire to learn: * Python * Statistics * Data Analysis * Data Visualization * Machine Learning tools (e.g., PyTorch, TensorFlow, Keras, Scikit-learn) Coursework or experience in Statistics or Data Science is preferred. Familiarity with Machine Learning is a plus.

Additional Comments

Opportunity to be co-hosted alongside a cohort of interns selected through the Scripps Institution of Oceanography (SIO)/CW3E Summer Internship Program at the beautiful SIO campus overlooking the Pacific Ocean.

Lightning Talk Title

AI Data-driven Models for the Prediction of Extreme Weather Events

Keywords

Weather; Meteorology; Artificial Intelligence; Machine Learning