Core Model Evaluation for Physical Oceanography Applications

The National Ocean Service (NOS), on behalf of the Department of Commerce’s National Oceanic and Atmospheric Administration (NOAA), is conducting market research to assess capabilities related to open-source, community-supported oceanographic models and their current and future potential. The effort is intended to support an evaluation of physical core model codes and approaches that can best meet NOAA’s mission needs while enabling modernization over the coming decade.

NOS currently supports multiple physical oceanography applications built on different community-supported models, including ROMS, FVCOM, ADCIRC, and SCHISM. While this model diversity helps address a broad mission set, it also creates resource challenges across staffing, funding, high-performance computing usage, and software maintenance.

The scope of the effort focuses on evaluating physical oceanography models against both quantitative and qualitative criteria. The selected approaches should be capable of meeting NOS mission and modeling requirements such as safe navigation and coastal resilience, while also supporting advances in computing infrastructure including HPC, cloud environments, and GPU architectures.

The evaluation will also consider the ability of models to leverage AI and machine learning, integrate with NOAA’s Unified Forecast System, and incorporate emerging approaches such as hybrid physics-AI systems. In addition to the core models currently used by NOS, NOAA is seeking information on emerging models and methodologies supported by academia and private industry that may satisfy future operational requirements.