One type of wind rose plot is a sort of 2-D histogram. Both azimuth and magnitude are binned, and the resulting azimuth sectors are plotted with magnitude on a color scale and frequency represented by the length of the sector (see example figure from Wikipedia below).
I thought that making a figure like this one in GMT would be simple, but it turns out to non-trivial. psrose does not have this functionality, as far as I can tell.
Has anyone ever tried to make a wind rose plot like this one using GMT? I’m sure it must be possible with some clever preprocessing. I’d love to have an example.
I’m trying to use psrose to plot windrose diagrams, but am also having problems figuring out how to do some things, e.g. figuring out if or how radial and azimuth coordinates can be better labelled. Plots like your diagram can be done using the windRose and pollutionRose functions in the R ‘openair’ package.
Could you tell me more about the per-sector “binning”? Since this example has some odd m/s boundaries between colors, I wonder if they got computed by the rose plot tool or if a user specified a CPT with those boundaries? Would specifying a CPT file with the limits be the way to set this up?
