[ncl-talk] Contour over map - projection issue
Michele Petrini
mpetrini139 at yahoo.it
Thu Jun 25 01:38:06 MDT 2015
Dear helping desk,
I am a beginner with NCL and I have the following issue. I have a netcdf
file with some data obtained from an ice sheet model run. The model
transforms latitude-longitude coordinates into native Cartesian
distances coordinates via the Lambert equal Area projection, using as a
center of projection the North Pole (0,90). Hence my data in the netcdf
file are geographically dependent on native Cartesian distances
coordinates. I am trying now to plot a contour over a map of the
Northern Emisphere of one of the netcdf file variables (let's say the
ice thickness). First of all, I imported succesfully the variables I am
interested to, depending on time, x and y, along with the latitude and
longitude variables, on x and y (where x and y are defined in native
Cartesian distances from (0,90). In order to go back from the native
Cartesian distances to latitude and longitude, I define a new array
/without /attributes (if the original variable is H, I just insert
Hlatlon= (\ H \)). Then I name the three dimensions of the new variable
Hlatlon (time, lat, lon), and I assign to the dimension number 1 and 2
(using the NCL enumeration from 0) the coordinate variables lat and lon.
Finally, I plot the contour of ice thickness over the map. What I get
when I run the script is wrong (I'm sure as, for instance, Greenland is
ice free). Where I did wrong? Is there a better (faster or more correct)
way to plot my contour with such variables?
Thank you for your help,
Michele
--
***
Michele Petrini
Ph.D. student in Earth Science and Fluid Mechanics
Università degli studi di Trieste,
Dipartimento di Matematica e Geoscienze
Palazzina C - via Weiss 1, 34128 Trieste, Italy
Email: mpetrini139 at yahoo.it
Skype: michele.petrins
Mobile: +39 3398367372
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20150625/623634bf/attachment.html
More information about the ncl-talk
mailing list