AI Data-driven Models and Very Large Ensembles for the Prediction of Extreme Events

Abstract

We propose to develop a novel regional artificial intelligence (AI) data-driven model to improve the predictions of atmospheric rivers and the associated extreme weather events. We will leverage the recently developed global AI data-driven models and the Center for Western Weather and Water Extremes high-resolution reanalysis (6 and 2 km domains over the Central and Eastern Pacific and Western US). These models can generate a forecast in just a few seconds to minutes, which will allow us to develop very-large ensembles (+7000 members) both regionally and globally, covering regions of interest to the Navy worldwide. We will also investigate the abilityof both the new regional and the existing global AI data-driven models to represent physical relationships across different variables in space and time. We will collaborate with colleagues from the Naval Research Laboratory, the European Centre for Medium-Range Weather Forecasts, and the Norwegian Meteorological Institute. Approved for Public Release

Document Details

Document Type
DoD Grant Award
Publication Date
Nov 09, 2024
Source ID
N000142412731

Entities

People

  • Luca Delle Monache

Organizations

  • Office of Naval Research
  • United States Navy
  • University of California, San Diego

Tags

Fields of Study

  • Environmental science

Readers

  • Artificial Intelligence
  • Ocean-Atmosphere Mesoscale Modeling, Data Assimilation, and Flux Boundary Layers
  • Seismology

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • Space