![](https://secure.gravatar.com/avatar/b951736e99a6f8d9129537b5fea565fa.jpg?s=120&d=mm&r=g)
I'm sorry, pleading ignorance here... But is my approach correct? Could you possibly provide a link to an example where data is projected first before the interpolation? Or, alternatively, a brief example (if only in sudo code). I just am not sure what I am doing here is correct: 149 # In this approach we work with projected data 150 x,y = m(x,y) 151 xres,yres = res 152 newx= np.arange(m.xmin,m.xmax,(m.xmax-m.xmin)/xres) 153 newy= np.arange(m.ymin,m.ymax,(m.ymax-m.ymin)/yres) 154 Znew = mlab.griddata(x,y,z,newx,newy) It seems to work, yes, but what do you mean there is nothing to do? Should my 'xres' simply be the lengths of the original x,y which are lon,lat ? Thanks again! Robert Kern-2 wrote:
On Wed, Aug 12, 2009 at 18:06, John [H2O]<washakie@gmail.com> wrote:
Robert Kern-2 wrote:
Are you using lat/lon as X/Y for griddata? Or are you projecting it first? You should project. Not seeing your code or data, I'm not sure I can diagnose what is going wrong with the interpolation.
-- Robert Kern
I have been trying to follow your suggestion of projecting first.. see example later, but I'm not sure if I am doing it correctly. One question is how then would I return the interpolated array back to a lat/lon grid... it seems transform_scalar is set up to go from lat/lon TO projection. I guess it could be used to 'unproject' the data as well?
The output Z values will be in the same order as the inputs. There is nothing to do.
-- Robert Kern
"I have come to believe that the whole world is an enigma, a harmless enigma that is made terrible by our own mad attempt to interpret it as though it had an underlying truth." -- Umberto Eco _______________________________________________ SciPy-User mailing list SciPy-User@scipy.org http://mail.scipy.org/mailman/listinfo/scipy-user
-- View this message in context: http://www.nabble.com/2d-interpolation%2C-non-regular-lat-lon-grid-tp2490968... Sent from the Scipy-User mailing list archive at Nabble.com.