High-Performance Computing Cluster for Modeling Cloudy Atmosphere--Ocean Dynamics and Sea Ice

Abstract

A high-performance computing (HPC) cluster is requested for modeling cloudy atmospheric dynamics and also its coupling with the ocean and sea ice. The HPC cluster would support interdisciplinary research and research-related education on new applied math techniques for modeling and predicting the coupled atmosphereoceansea-ice system.Forecasts of sea ice are challenging due to many factors. For instance, sea ice can take many different forms, including smaller-scale discrete ice floes, or larger-scale sheets that can sometimes be treated as a continuum of ice but also are subject to relatively thin cracks. Furthermore, sea ice is influenced by its coupling with the atmosphere and ocean, and atmospheric and oceanic predictions are themselves challenging due to additional factors, such as the complexities of atmospheric water vapor and clouds. Mathematical models and computational frameworks are difficult to design for the evolution of such a complex system.A research plan is described for designing new computational frameworks, based on multiscale modeling, stochastic modeling, and data science. Such mathematical techniques will allow high-fidelity simulation of ice floe dynamics, and interaction with the ocean and cloudy atmosphere, while also remaining computationally tractable.The proposed HPC cluster consists of 800 processor cores and 6 gigabytes of memory per core. This configuration would allow both parallel computations and memory-intensive computations and data analysis, and it can support the needs of many users concurrently, including several interdisciplinary graduate students and postdoctoral researchers.

Document Details

Document Type
DoD Grant Award
Publication Date
Jun 17, 2020
Source ID
N000142012495

Entities

People

  • Samuel N. Stechmann

Organizations

  • Office of Naval Research
  • United States Navy
  • University of Wisconsin System

Tags

Fields of Study

  • Environmental science

Readers

  • Parallel and Distributed Computing.
  • Polar and Arctic Studies
  • Research Science/Academic Research

Technology Areas

  • AI & ML