Serving Real-Time Point Observation Data in netCDF using Climate and Forecasting Discrete Sampling Geometry Conventions
Abstract
NetCDF is a set of software libraries and self-describing, machine-independent data formats that support the creation, access, and sharing of array-oriented scientific data. The Climate and Forecasting (CF) metadata conventions for netCDF foster the ability to work with netCDF files in general and useful ways. These conventions include metadata attributes for physical units, standard names, and spatial coordinate systems. While these conventions have been successful in easing the use of working with netCDF-formatted output from climate and forecast models, their use for point-based observation data has been less so. Unidata has prototyped using the discrete sampling geometry (DSG) CF conventions to serve, using the THREDDS Data Server, the real-time point observation data flowing across the Internet Data Distribution (IDD). These data originate in text format reports for individual stations (e.g. METAR surface data or TEMP upper air data) and are converted and stored in netCDF files in real-time. This work discusses the experiences and challenges of using the current CF DSG conventions for storing such real-time data. We also test how parts of netCDF's extended data model can address these challenges, in order to inform decisions for a future version of CF (CF 2.0) that would take advantage of features of the netCDF enhanced data model.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2016
- Bibcode:
- 2016AGUFMIN23A1763W
- Keywords:
-
- 1904 Community standards;
- INFORMATICSDE: 1936 Interoperability;
- INFORMATICSDE: 1946 Metadata;
- INFORMATICSDE: 1982 Standards;
- INFORMATICS