The Model Diagnostics Task Force is constituted to develop, coordinate, and implement process-based model evaluation metrics and a metrics framework in National modeling center metrics packages, leveraging ongoing efforts at the modeling centers toward advancing model evaluation and development capabilities.
This Task Force is constituted of researchers whose projects were successfully evaluated as part of the FY15 competition held by NOAA’s Modeling, Analysis, Predictions, and Projections program on the topic of Process Oriented Metrics. This competition was motivated by a number of factors, including community interest in moving beyond performanceoriented metrics toward processoriented metrics, ongoing efforts to develop next-generation climate and Earth system models, evolving plans for CMIP, and a need to link model development and evaluation efforts across modeling centers. Nine projects were selected as a result of the competitive process, including eight focused on developing metrics and one that will develop metrics as well as define a direction and framework for the overall effort.
Task Force Function and Implementation
This Task Force will initiate its activities in October 2015 and will have a duration of three years. It is expected that researchers who were selected through the NOAA competitive process will participate actively in this Task Force helping to build an integrative processoriented metrics framework serving NOAA and other modeling centers. Task Force membership may also include affiliates, elected by Task Force leadership, who are not supported by MAPP funding but whose expertise and work are relevant to achieving the goals of this task force. The Task Force will coordinate with other relevant ongoing diagnostic efforts including emerging community processoriented metrics efforts such as the WCRP/WGCM Metrics Panel, PCMDI’s UVCDAT effort, and the EMBRACE ESMValTool project.
The Task Force will connect all of the funded investigators and other invitees. Beyond enhancing communication between investigators, expectations of this group include contribution by the individual projects to the development and implementation of the collective effort with a focus on implementing a cohesive metrics framework, open documentation of the efforts, and a focus on maximizing the community utility of metrics and the metrics framework. It is expected the main group will have monthly teleconferences and consider inperson meetings, as appropriate.
As of Spring 2017, the Task Force has developed a functional Application Program Interface (API) for the metrics package, which is documented here. The linked document also describes the overall status of the effort as of April 2017. The API is python based and designed to be flexible such that metrics coded in different non-proprietary software packages can be integrated into the API.
MAPP Task Force Concept and Terms of Reference
Lead: Eric Maloney, Colorado State University
Co-Lead: Andrew Gettelman, National Center for Atmospheric Research
Co-Lead: Yi Ming, NOAA Geophysical Fluid Dynamics Laboratory
Co-Lead: David Neelin, University of California, Los Angeles
Relevant MAPP Program PIs and selected additional invitees.
To view the full Participants list, please visit the Participants page.
To view the full list of Projects, please click here.