access_iris - ACCESS-NRI/accessdev-Trac-archive GitHub Wiki
Iris
Iris is an open source Python library for analysing and visualising meteorological and oceanographic data sets, initially developed at the Met Office.
https://github.com/SciTools/iris Source code repository
http://scitools.org.uk/iris/docs/latest/ Documentation
https://groups.google.com/forum/#!forum/scitools-iris Users forum
https://groups.google.com/forum/#!forum/scitools-iris-dev Developers forum
Iris uses matplotlib for graphics and cartopy (also from the Met Office) for mapping (replacing basemap).
Features
- Reads/writes netCDF, UM PP and GRIB files
- Reads UM fieldsfiles
Use on raijin
Iris 1.6 is installed on raijin under ~access/apps. To use it
module use ~access/modules
module load iris
Examples
Galley of example plots with links to scripts is here.
Issues
- cartopy dependencies make installation awkward
- Incomplete standard names and other metadata with fieldsfiles
- Inconsistent merging of fields into multidimensional variables
Comparison with cdat
Note this refers to cdat 5 rather than the new UV-CDAT which I haven't really tried yet.