ERDC, Microsoft agreement aims to analyze risk of extreme weather in the cloud

U.S. ARMY ENGINEER RESEARCH AND DEVELOPMENT CENTER
Published July 14, 2021
Updated: July 14, 2021
A new agreement between the U.S. Army Engineer Research and Development Center and Microsoft will improve climate modeling and natural disaster resilience through the use of predictive analytics-powered, cloud-based tools and Artificial Intelligence services.

A new agreement between the U.S. Army Engineer Research and Development Center and Microsoft will improve climate modeling and natural disaster resilience through the use of predictive analytics-powered, cloud-based tools and Artificial Intelligence services.

A new agreement between the U.S. Army Engineer Research and Development Center and Microsoft will improve climate modeling and natural disaster resilience through the use of predictive analytics-powered, cloud-based tools and Artificial Intelligence services.

A new agreement between the U.S. Army Engineer Research and Development Center and Microsoft will improve climate modeling and natural disaster resilience through the use of predictive analytics-powered, cloud-based tools and Artificial Intelligence services.

VICKSBURG, Miss.—  Modeling the risk of extreme weather and natural disasters along the nation’s coastline is critical to the U.S. Army Engineer Research and Development Center (ERDC) mission of delivering innovative solutions for safer, better world.

Increasing this modeling capacity and better dissemination of data to climate research is the goal of a new agreement between the ERDC and Microsoft Corporation. This government/industry collaboration is aimed at improving climate modeling and natural disaster resilience planning through the use of predictive analytics-powered, cloud-based tools and Artificial Intelligence (AI) services. 

The agreement seeks to demonstrate the scalability of the code of ERDC’s premier coastal storm modeling system, CSTORM-MS, inside Microsoft’s Azure Government, a cloud computing service for building, testing, deploying and managing applications and services through Microsoft-managed data centers specifically for the U.S. Government. CSTORM-MS is a comprehensive integrated system of highly skilled and highly resolved models used to simulate coastal storms. The models provide for a robust, standardized approach to establishing the risk of coastal communities to future occurrences of storm events and for evaluating flood risk reduction measures. With its physics-based modeling capabilities, CSTORM-MS integrates a suite of high-fidelity storm modeling tools to support a wide range of coastal engineering needs for simulating tropical and extra-tropical storms, as well as wind, wave and water levels. 

Currently, CSTORM-MS models are run at ERDC’s Department of Defense Supercomputing Resource Center, one of the DoD High Performance Modernization Program’s (HPCMP) supercomputing centers. In 2020, ERDC and the HPCMP performed a commercial cloud for high-performance computing workload assessment. This initial testing included a feasibility study of the CSTORM-MS models, and was successfully conducted using Microsoft’s Azure cloud.

Through the Cooperative Research and Development Agreement (CRADA) between ERDC and Microsoft, two goals have been set for this second phase of the project: 

  • To demonstrate scalability of the CSTORM-MS code on Azure through running the entire North Atlantic Coast storm suite with a sea level rise value not previously simulated; and
  • To create an opportunity for researchers to use the model results and replicate the workflow on their affected coastlines. 

Microsoft’s participation in this effort stems from their Microsoft AI for Earth, a working group within Microsoft established in June 2017 that provides cloud-based tools and AI services to organizations working to protect the planet across five key areas: agriculture, biodiversity, conservation, climate change and water. AI for Earth awards grants to support projects that use AI to change the way people and organizations monitor, model and manage Earth's natural systems. 

The CRADA between ERDC and Microsoft is made possible through the Federal Technology Transfer Act of 1986. The act provides that federal laboratories’ developments, such as those of ERDC, should be made accessible to private industry and state and local governments for the purpose of improving the economic, environmental and social well-being of the United States by stimulating the use of federally funded technology developments or capabilities.