For 2D images, you might consider using the NCL , released by the NCAR (National Center for Atmospheric Research), which is specifically designed to manipulate and build atmospheric (meteorological) data. An impressive gallery of results is available on this site . IMO's NCL syntax can be very ugly at times, and it looks like python PyNGL shell ( gallery ) now exists
Some of my colleagues used a number of very impressive tools to visualize 3D atmospheric data, including vis5d and steam .
For maximum flexibility, mayavi2 , which is basically a MATLAB-style ipython interface for Kitware VTK library, is used in a huge range of scientific visualizations. I do not recommend this approach if you do not want to "collapse your own", but it is very nice if you need to get three-dimensional results that look right.
source share