AGU Fall Meeting, Chicago, IL & Online Everywhere, 12 - 16 December 2022

EPIC @ AGU Fall Meeting, 12 – 16 December 2022


EPIC is happy to announce that we will be presenting at the AGU Fall Meeting, December 12 – 16,  2022.  It is the most influential event in the world dedicated to the advancement of Earth and space sciences.

The meeting will be held in Chicago and online everywhere. More than 25,000 attendees from more than 100 countries will convene to explore how Science Leads the Future. AGU will host a diverse community of scientists, students, journalists, policymakers, educators and organizations who are working toward a world where scientific discovery leads to scientific solutions, and where our global collaborations and partnerships can carry us into a sustainable future.

Find out more about the AGU Fall Meeting.

Registration rates are available on the AGU Meeting Registration page:


Mandy Parson
Mandy Parson
Technical Trainer

Mandy graduated from the University of Massachusetts, Amherst with a B.A. in Communications and Information Technology. She has also completed the Flatiron Full-Stack Software Engineering Program as of August 2021. Before coming onto the EPIC program, Mandy taught STEM to middle and high school students. Currently, she is teaching community modeling basics on the EPIC program.

Natalie Perlin
Natalie Perlin
AUS Team Member
Performance Solutions

Dr. Natalie Perlin has a background in engineering meteorology, atmospheric sciences, and modeling, and she received her Ph.D. from Tel-Aviv University. Dr. Perlin brings more than 25 years of experience in numerical modeling and data analysis in the areas of atmospheric and ocean sciences. She has worked on many projects that included model improvement and software development for users’ community, design, and implementation of coupled modeling systems, developing post-processing algorithms, and visualization methods applied to observational and modeling data in environmental sciences. Dr. Perlin started to work on NOAA EPIC program in 2022 as a Sr. Systems Engineer. 

Yi-Cheng Teng
Yi-Cheng Teng

Dr. Yi-Cheng Teng graduated from Virginia Institute of Marine Science (VIMS), College of William and Mary, with a PhD degree in Marine Science.   Yi-Cheng has more than 10 years of experience in physical oceanography and numerical modeling. Before joining and EPIC program, Yi-Cheng has been an assistant professor and a principal investigator for multiple projects supported by NSF, DOE, and NOAA.  Currently, he is a senior software engineer and working on infrastructure development of UFS Hierarchical Testing Framework.

Photo of Mike Kavulich
Stylianos Flampouris
Product Owner/Vice President
EPIC Platform Team/

Mike graduated from Worcester Polytechnic Institute with a B.S. in Physics, and from Texas A&M University with an M.S. in Atmospheric Science. He has worked as an Associate Scientist at NCAR since 2012 on various projects in development and user support for numerical weather prediction, data assimilation, and other software packages in the realm of WRF and the UFS.

Photo of Jong Kim
Jong Kim
Lead Configuration/Code Manager

Jong Kim works as a lead configuration and code manager for the NOAA EPIC program to support the UFS weather and application releases. Before joining EPIC, he led the marine data assimilation team at NOAA-NWS-NCEP-EMC. Through the UFS-R2O marine reanalysis project, the next generation global ocean data assimilation system (NG-GODAS) was successfully demonstrated in the JEDI-based sea-ice ocean coupled assimilation framework. Jong also spent about 12 years as a lead software engineer for the global modeling and assimilation office at NASA-GSFC. His early career includes a computational scientist position at the mathematics and computer science division of the DOE Argonne National Lab. He holds a Ph.D in Chemical Engineering from the University of Utah.

EPIC's Workshop at AGU

Earth Prediction Innovation Center (EPIC) Workshop: Running the Unified Forecast System (UFS) Short-Range Weather Application in the Cloud
Saturday, December 10th from 1:00pm – 5pm CT

This short course will teach students how to configure and run NOAA’s Unified Forecast System’s (UFS) Short Range Weather (SRW) Application on the Amazon Web Service (AWS) cloud computing platform. The UFS SRW App is a numerical weather prediction framework that targets predictions of atmospheric behavior on a limited spatial domain and on time scales from less than an hour out to several days. It is the foundation for improving NOAA’s predictions of hazardous convective storms made by convection-allowing ensemble forecast systems. The application does this by bringing together the essential components of the UFS regional modeling system under a single umbrella that runs these components as separate but interdependent tasks. These components include the Finite Volume Cubed (FV3) based prognostic UFS Weather Model, pre- and post-processing codes, verification, and a workflow for running the system end-to-end. The App is designed to be used by both researchers and in operations.

NOTE: This workshop can be added to an attendee’s cart when registering for the Fall Meeting.

Workshop Agenda

Activity/Method Content Description
Introduce speakers and their backgrounds. Let students know how they can ask questions and to whom they should direct questions. Introduce slack and how to use it.
The UFS, EPIC, and the Short Range Weather Application
Discuss the Short Range Weather App and it’s uses.
Log In and Run Control Case
  • Have students SSH into AWS using the PEM file.
  • Run through the control case (25k GFS_v16 physics).
  • Generate the plots.
Modify Test Case
  • Change from 25k to 3k high resolution case  with GFS_v16 physics.
  • Run new test case
  • Provide more details on how to define a custom grid.
  • Generating the plots.
Modify Test Case Again
  • Modify physics (RRFS_v1beta physics with 3km grid).
  • Run new test case
  • Generate the plots.
Compare Outputs
Have students run python scripts to compare the forecast outputs produced by their previous two cases.
  • Have students run another test case on their own.
  • Open for Q&A and ad hoc facilitation.
  • Have students fill out survey.
  • Address any remaining questions.


The following abstracts from The EPIC Team were accepted for submission:

Advances in Continuous Development and Continuous Deployment of Earth System Models

by Jong Kim

Collage of different weather conditions for AGU 2022 Abstracts

Summary: Jenkins offers a transparent and continuous integration (CI) framework for the UFS Weather Model and application codebases. As key components in the EPIC platform and infrastructure services to the UFS community, the Jenkins CI pipelines are reviewed in the presentation.

Building the Community Infrastructure for the Unified Forecast System 

by Stelios Flampouis

deployment pipeline image for AGU abstracts

Summary: EPIC’s Platform Team, led by Stelios Flampouris, will discuss the transformation and evolution of the infrastructure required for the UFS Weather Model (WM) and applications to provide the necessary support to the UFS community. The presentation will focus on the main developments and upgrades to code management, continuous integration and deployment, data management, and testing capabilities of the UFS WM, components, and applications.

EPIC Tweets