Code Reproducibility Practices in the UFS Weather Model and Land DA Application Development

The Earth Prediction Innovation Center (EPIC), launched by NOAA/WPO to accelerate the community's development and integration of innovations to the Unified Forecast System (UFS) Weather Model and Land Data Assimilation (DA) System. A significant aspect of the center’s work is the continuous improvement and development of the UFS-WM and Land DA Applications in regard to code reproducibility and operational readiness of the code in support of the community development.

This poster presentation summarizes a series of operational readiness and performance testing for these UFS applications on NOAA RDHPC systems. Of importance when examining this topic is the Operational Requirement Tests currently in use by UFS developers when determining the suitability of a high-powered computing platform or how proposed code changes will affect the current Weather Model development repository. In addition to this, the newly developed Land DA “ctest” suite (unit tests), using the ERA5-Land dataset, will be examined to showcase the new test’s use cases and benefits to the UFS community.

The code reproduction and operational testing methods discussed within this presentation are meant to educate and provide a demonstration to users of the best practices and intended uses of the UFS-Weather Model Operational Requirement and Land DA ctests, so community members have documented practices to assist them in preparing to test new code development and ensuring platform readiness.