5A.1 The Unified Forecast System: Improving Research to Operations

Tuesday, 14 January 2020: 10:30 AM
252A (Boston Convention and Exhibition Center)
Richard B. Rood, Univ. of Michigan, Ann Arbor, MI; and H. L. Tolman

The Unified Forecast System: Improving Research to Operations

Richard B. Rood and Hendrik Tolman for the Unified Forecast System - Steering Committee

The Unified Forecast System (UFS) is a community-based, coupled, comprehensive Earth modeling system. It can be configured into multiple applications, which span local to global domains and predictive time scales from sub-hourly analyses to seasonal predictions. UFS is designed to support both the United States Weather Enterprise, including National Weather Service forecasts, and to be an exploratory platform that can include research elements that are not part of operational applications.

This talk introduces the research to operations (R2O) transition process to support NOAA’s efforts to develop a Unified Forecast System (UFS). The description relies on policy documents as well as analysis of research to operations transitions that have occurred in the past. To define, better, and to improve research to operations transitions, we advocate using upcoming transitions as use cases to develop a detailed understanding of the processes and, hence, support improvement.

For the purpose of organization, we identify three types of research to operations transitions. These three types are of different complexity, time spans, and cost. Examples are provided in the text. 1) From the UFS systems perspective, the Systems Level transition requires the selection of system-level components from the community, configuration for a particular application, evaluation of candidate systems, and then transition to operations. This is strategic level transition of order five years in effort with decadal length organizational outcomes. 2) From a UFS application perspective, the Application Level transition considers a set of defined changes to, primarily, a single component in an existing application. This is order of months to two years in efforts. 3) Incremental Level transitions are more frequent and target narrow changes to existing algorithms and functions in application system of high Readiness Level.

We identify the functions that need to exist in a robust, repeatable research to operations transition capacity. These functions are Management and Decision Making, Workflow, Code Management, Systems Integration, Developer and User Support, and Testing, Verification & Validation. None of the functions are at full capacity or, even, fully defined for the current application suite. It is essential to identify the functions and gaps in the end-to-end research-to-operations process prior to allocating resources to improve the process. Systematic improvement of the operations-to-research interface follows from this systems-based improvement.

The full document can be accessed here: https://drive.google.com/open?id=14IDAKWA_-FVZaJrhqV625fwMePSJTBnc

Unified Forecast System Web Portal: https://ufscommunity.org/

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner