Re: RotatedMercator map projection

From: Seth McGinnis <mcginnis_at_nyahnyahspammersnyahnyah>
Date: Tue, 16 Sep 2008 12:48:06 -0600

I got it figured out earlier today, so I'm all good. You got me pointed in the right direction, though, so thanks!

Suppose your data is in Lambert Conformal projection, with appropriate @lat2d & @lon2d arrays. To plot the data in its native projection, so that you have it properly placed relative to geographic features on the globe but the grid still ends up nice and square, you pass these resource values to gsn_csm_contour_map() (where "proj" is a variable in the NetCDF file that contains information about the map projection):

      res_at_mpProjection = "LambertConformal"
      res_at_mpLambertMeridianF = proj_at_longitude_of_central_meridian
      res_at_mpLambertParallel1F = proj_at_standard_parallel(0)
      res_at_mpLambertParallel2F = proj_at_standard_parallel(1)
      res_at_tfDoNDCOverlay = True

The equivalent for data in a rotated pole or rotated lat-lon projection (which I now know is *not* the same as rotated Mercator) is this:

      res_at_mpProjection = "Mercator"
      res_at_mpCenterLonF = proj_at_grid_north_pole_longitude - 180
      res_at_mpCenterLatF = 90 - proj_at_grid_north_pole_latitude

Cheers,

--Seth

>Hi Seth,
>
>I am not sure I can visualize what you are trying to do. Can you send the script and the data
>or at least an image of what you have now?
> -dave
>
>On Sep 16, 2008, at 12:56 AM, Seth McGinnis wrote:
>
>> Ah! I get it now. I didn't realize they were different. Thank you!
>>
>> So, it turns out my data is actually rotated pole. I have lat2d &
>> lon2d arrays, so the data is showing up in the right place on the
>> map, but I want to use a map projection that will make it square.
>> Any thoughts? In Ezmap terms, I believe it has non-zero PLAT &
>> PLON. Is that what mpCenterLatF & mpCenterLonF specify for the
>> normal Mercator projection?
>>
>> --Seth
>>
>>
>>> Hi Seth,
>>> The RotatedMercator projection is not actually the same as a
>>> rotated pole projection. It ignores the 'mpCenterLatF' value,
>>> and the 'mpCenterRotF' value behaves in a different manner from
>>> the normal Mercator projection.
>>> You will see the difference if you give 'mpCenterRotF' a non-zero
>>> value.
>>>
>>> For an explanation of the difference between these projections
>>> from the perspective of the underlying software, see:
>>>
>>> http://www.ncarg.ucar.edu/supplements/ezmap/#PROJECTIONS
>>>
>>> Scroll down to 'Cylindrical Projections'
>>>
>>> If the grid is plotted in its native projection then you must
>>> either have the tfDoNDCOverlay flag set True, or you are not
>>> communicating
>>> the 2d coordinates correctly to the plotting routine.
>>>
>>> -dave
>>>
>>>
>>>
>>> On Sep 15, 2008, at 4:17 PM, Seth McGinnis wrote:
>>>
>>>> Does anyone have any examples of how to plot data that's in a
>>>> rotated Mercator (aka rotated pole or rotated lat-lon) coordinate
>>>> system?
>>>>
>>>> I've been making plots using the following map projection resources:
>>>>
>>>> res_at_mpProjection = "RotatedMercator"
>>>> res_at_mpCenterLonF = 83
>>>> res_at_mpCenterLatF = 42.5
>>>>
>>>> and as far as I can tell, it doesn't look any different than
>>>> leaving the projection unspecified. (The data has lat2d and lon2d
>>>> arrays. Using the same attributes with the corresponding values
>>>> for data in a polar stereographic or transverse mercator projection
>>>> results in the native grid being plotted square on the page.)
>>>> Maybe I have the wrong values there, but I tried it with
>>>> mpCenterLatF = 0, 45, and 90, and it made no difference to the
>>>> resulting plot. Should I be specifying the projection parameters
>>>> in a different way?
>>>>
>>>> The only examples I can find of the RotatedMercator projection on
>>>> the NCL website are in the "Map Only" application examples 12 & 13,
>>>> and it looks to me like the RotatedMercator example is identical to
>>>> the Mercator example.
>>>>
>>>> Thanks,
>>>>
>>>> --Seth
>>>>
>>>> ----
>>>> Seth McGinnis
>>>> Associate Scientist
>>>> ISSE / NCAR
>>>> ----
>>>> _______________________________________________
>>>> ncl-talk mailing list
>>>> ncl-talk_at_ucar.edu
>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>
>>
>

_______________________________________________
ncl-talk mailing list
ncl-talk_at_ucar.edu
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Tue Sep 16 2008 - 12:48:06 MDT

This archive was generated by hypermail 2.2.0 : Wed Sep 17 2008 - 10:32:36 MDT