Data in rotated grid coordinates

[Gudrun Nina Petersen]

Dear Diandong,

thank you very much for your reply.

The information I have of the grid are the following:

Number of columns: 326
Number of rows:    420
Column spacing:    0.11 degrees
Row spacing:       0.11 degrees
First latitude:    -14.00
First longitude:   342.00
North Pole lat:     36.00
North Pole lon:    140.00

The rotated grid is a rectangular lon/lat grid, the actual grid is much
wider at the top than at the bottom.

I can calculate the location of each point in the actual geographical
grid by using the attached program.
What is does is that is calculates
the geographical (lon/lat)_A (A=actual)
from the (lon/lat)_I of the north pole
and the (lon/lat) in the rotated grid.
This results in two 2-D arrays lon_A(i,j) and lat_A(i,j).

But how to make GrADS understand this is yet beyond my understanding.

I would really appreciate help here.


Cheers,

Nina


Ren Diandong wrote:
> First, you need to know the GrADS (COARSIAN) convention and the
> projection used to create the data. For polar steorographic projection,
> for example, you need to know the reference longitude, the reference
> latitude, the grid spacing (on the flat map) in x- (W-->E) and y-(N-->S)
> directions.
> A key parameter is the pole's grid counting in the grid space, which can
> be easily figured out if you know the projection rules.
>
> I am not sure of the projection your data is produced with. If you can
> provide the projection code, I can write a correspondi
> ng code to find the parameters for use pdef (use GrADS to display
> preprojected data).
>
> Best wishes,
>
> Diandong
>
>
> On Wed, 2006-11-08 at 11:59 +0100, Gudrun Nina Petersen wrote:
>> I have data from the UK Met Office Unified Model (UM) that is in rotated
>> grid coordinates. My domain is roughly centred on 60N.  North post is brought to
>> equator to make equal area grid for the calculations.
>> If I plot the data directly,
>> the longitude is 18W to 17E and latitude 14S to 32N
>>
>> However, in reality the data is within the following boundaries
>>
>>    (75N,120W)-----------------(86N,40W)--------------------(75N,40E)
>>        |                          |                             |
>>        |                          |                             |
>>        |                          |                             |
>>        |                          |                             |
>>        |                          |                             |
>>        |                          |                             |
>>        |                          |                             |
>>        |                          |                             |
>>    (37N,62W)------------------(40N,40W)---------------------(37N,18W)
>>
>>
>> I've read the documentation on PDEF but I still don't understand how to plot
>> the data in the correct geographical location.
>>
>> I can write a pdef-file that contains x(i,j),y(i,j),wrot(i,j), information
>> about the correct location of each point: e.g. x(1,1)=-62 and y(1,1,)=37.
>>
>> However, I can't understand how to use my pdef-file in the ctl-file as the
>> data is not on a linear grid anymore when moved to the north.
>>
>>
>> Any help would be greatly appreciated.
>>
>> Best regards,
>>
>> G. Nina Petersen
>> School of Environmental Sciences
>> University of East Anglia
>> Norwich
>> UK
>>

--

Guðrún Nína Petersen
School of Environmental Sciences
University of East Anglia
Norwich
UK
-------------- next part --------------
A non-text attachment was scrubbed...
Name: um_xy_nc.f
Type: text/x-fortran
Size: 3393 bytes
Desc: not available
Url : http://gradsusr.org/pipermail/gradsusr/attachments/20061109/9334273d/attachment.bin