1. CF 2.0 Coming Soon? (Climate and Forecast Conventions for netCDF) Ethan Davis ESO Developing Standards - ESIP Summer Mtg 14 July 2015

2. CF 2.0 Coming Soon? (Climate and Forecast Conventions for netCDF) Ethan Davis ESO Developing Standards - ESIP Summer Mtg 14 July 2015

3. CF 2.0 Coming Soon? (Climate and Forecast Conventions for netCDF) Ethan Davis ESO Developing Standards - ESIP Summer Mtg 14 July 2015

4. What is netCDF? Binary data format(s) Software libraries / APIs – C (Fortran, C++, Python, …) and Java – Upon which many data management, analysis, and visualization tools have been built Data model – Conceptual model of the data – Independent of data format details – Independent of programming language

5. What is netCDF? Classic netCDF data model – Multidimensional arrays of data values Which share dimensions – Attributes

6. What is netCDF? Classic netCDF data model – Multidimensional arrays of data values Which share dimensions – Attributes lat lon

7. What is netCDF? Classic netCDF data model – Multidimensional arrays of data values Which share dimensions – Attributes netcdf mydataset { dimensions: lat = 12 ; lon = 19 ; time = 4 ; variables: float lat(lat) ; float lon(lon) ; float temp(time, lat, lon) ; float rh(time, lat, lon) ; attributes: :Conventions = "CF-1.6"; } lat lon

8. What is netCDF? Classic netCDF data model – Multidimensional arrays of data values Which share dimensions – Attributes

9. What is netCDF? Classic netCDF data model – Multidimensional arrays of data values Which share dimensions – Attributes Enhanced netCDF data model – Adds hierarchical groups Organize and group dimensions and variables – Adds structures – Only available in netCDF-4 (based on HDF5)

10. What is netCDF? Classic netCDF data model – Multidimensional arrays of data values Which share dimensions – Attributes Enhanced netCDF data model – Adds hierarchical groups Organize and group dimensions and variables – Adds structures – Only available in netCDF-4 (based on HDF5)

11. What is netCDF? Enhanced netCDF data model – Adds hierarchical groups Organize and group dimensions and variables – Adds structures – Only available in netCDF-4 (based on HDF5)

12. What is netCDF? Enhanced netCDF data model – Adds hierarchical groups Organize and group dimensions and variables – Adds structures – Only available in netCDF-4 (based on HDF5)

13. What Does It Mean? netcdf mydataset { dimensions: lat = 12 ; lon = 19 ; time = 4 ; variables: float lat(lat) ; float lon(lon) ; float temp(time, lat, lon) ; float rh(time, lat, lon) ; attributes: :Conventions = "CF-1.6"; }

14. CF conventions Describe coordinate systems – Spatial, temporal, wavelength, etc Describe the structure of the data – Gridded, station, soundings, etc Indicate the meaning of the data – Geophysical parameter – Standard Name - controlled vocabulary Extent over which data value is valid – Point or Cell – Characteristics: min, max, mean, etc.

15. CF conventions Historically dealt with gridded data

16. CF conventions Historically dealt with gridded data Starting to deal with observational data

17. CF conventions Historically dealt with gridded data Starting to deal with observational data CF 1.6 limited to netCDF classic data model

18. CF 2.0 Use features of the Enhanced Data Model – Enable things cannot be done with Classic Data Model – Simplify complexities from Classic Data model Organize data with hierarchical group structure – Satellite data – Members of Ensemble Models Expand types of spatial extents allowed for describing the valid extent of data values – CF 1.x: Point or Cell – Extend to polylines, polygons, more complex coverages

19. CF 2.0 Use features of the Enhanced Data Model – Enable things cannot be done with Classic Data Model – Simplify complexities from Classic Data model Organize data with hierarchical group structure – Satellite data – Members of Ensemble Models Expand types of spatial extents allowed for describing the valid extent of data values – CF 1.x: Point or Cell – Extend to polylines, polygons, more complex coverages

20. CF 2.0 Use features of the Enhanced Data Model – Enable things cannot be done with Classic Data Model – Simplify complexities from Classic Data model Use hierarchical group structure to organize data – Satellite data – Members of Ensemble Models Add new types of spatial extents used to describe the valid extent of data values – CF 1.x: Point or Cell – Extend to polylines, polygons, more complex coverages

21. CF 2.0 – Join the Discussion https://github.com/cf-convention/CF-2 http://cfconventions.org/ Questions? – Ethan Davis, edavis@ucar.edu Unidata is one of the University Corporation for Atmospheric Research (UCAR)'s Community Programs (UCP), and is funded primarily by the National Science Foundation (Grant NSF-1344155).