Re: Converting axis coordinates

Adam

Thanks. I think that solved all the problems. Apologies for missing the
PostDraw setting.

Also, thanks for the real-time support!!!

gerry

On 09/15/2011 02:15 PM, Adam Phillips wrote:
> Hi Gerry,
> (see below)
>
> On 09/15/2011 01:08 PM, Gerald Creager wrote:
>> Hello, Adam,
>>
>> I've tried that but I get an error:
>> (0) gsn_add_cyclic: Warning: The range of your longitude data is not 360.
>> (0) You may want to set gsnAddCyclic to False to avoid a warning
>> (0) message from the spline function.
>> warning:_NhlCreateSplineCoordApprox: Attempt to create spline
>> approximation for X axis failed: consider adjusting trXTensionF value
>> warning:IrTransInitialize: error creating spline approximation for
>> trXCoordPoints; defaulting to linear
>
> When you are plotting regional data on a map, you need to tell NCL that
> the data is regional so it doesn't try to wrap your data around the
> cyclic point. This is one of the more common error messages, and in this
> case it tells you what to do: set gsnAddCyclic = False.
>
>>
>> Attached is the script and a .png of the output.
>>
>>
>> ALSO, I would like to display land in gray with a black shoreline. Isn't
>> there an easy way to accomplish this?
>
> See my 1st email. The black shoreline should be there by default. If
> not, set mpOutlineDrawOrder = "PostDraw". As you are wanting to fill the
> land _after_ you fill your data, you must tell NCL to draw the map fill
> last via mpFillDrawOrder = "PostDraw"...
> Adam
>
>>
>> Thanks, again,
>> gerry
>>
>> On 09/15/2011 01:57 PM, Adam Phillips wrote:
>>> Hi Gerry,
>>> I see you are using gsn_csm_contour as opposed to gsn_csm_contour_map..
>>> (I didn't recognize this yesterday when I sent my original reply.) Are
>>> you doing that on purpose? If so, then you can simply set vpWidthF and
>>> vpHeightF to whatever values you wish between 0 and 1.
>>>
>>> If not, and you meant to use gsn_csm_contour_map, then the aspect ratio
>>> will depend on the map projection you are using. By default the
>>> projection is cylindrical equidistant. If you want to change the aspect
>>> ratio when using one of the gsn_csm_*_map* routines, beyond setting
>>> vpWidthF and vpHeightF, you also need to set mpShapeMode = "FreeAspect"
>>>
>>> Hope that helps..
>>> Adam
>>>
>>> On 09/15/2011 12:43 PM, Gerald Creager wrote:
>>>> Hit Send too fast. That did, in fact, cure the problem. Thanks for the
>>>> gentle reminder.
>>>>
>>>> Now, an aesthetics problem: The plot looks "squished" horizontally. Is
>>>> there a suggested aspect ratio I could set?
>>>>
>>>> On 09/15/2011 08:16 AM, Gerald Creager wrote:
>>>>> Argh... No, Adam, it doesn't:
>>>>>
>>>>> printVarSummary(model1Ssh)
>>>>>
>>>>> Variable: model1Ssh
>>>>> Type: float
>>>>> Total Size: 37677960 bytes
>>>>> 9419490 values
>>>>> Number of Dimensions: 3
>>>>> Dimensions and sizes: [90] x [261] x [401]
>>>>> Coordinates:
>>>>> Number Of Attributes: 1
>>>>> _FillValue : -9999
>>>>>
>>>>>
>>>>> On 09/14/2011 04:56 PM, Adam Phillips wrote:
>>>>>> Hi Gerry,
>>>>>> Does your model1Ssh array have associated latitude/longitude
>>>>>> coordinate
>>>>>> variables? You can see if it does by using printVarSummary:
>>>>>> printVarSummary(model1Ssh)
>>>>>>
>>>>>> You should see something like this:
>>>>>>
>>>>>> Variable: model1Ssh
>>>>>> <snip>
>>>>>> Coordinates:
>>>>>> time: [17067072..17618592]
>>>>>> lat: [90..-90]
>>>>>> lon: [ 0..357.5]
>>>>>> (Last 3 lines show the coordinate variables.)
>>>>>>
>>>>>> If coordinate variables aren't present, you need to assign them.
>>>>>> Looking
>>>>>> at your script though, I see you are doing this:
>>>>>> model1Ssh=100 * model1->sea_surface_height_above_sea_level
>>>>>> This will result in only the data being passed from the right-hand
>>>>>> side
>>>>>> of the equal sign to the left-hand side, and not the coordinate
>>>>>> variables associated with sea_surface_height_above_sea_level array.
>>>>>>
>>>>>> I'd recommend changing this:
>>>>>> model1Ssh=100 * model1->sea_surface_height_above_sea_level
>>>>>> to this:
>>>>>> model1Ssh=model1->sea_surface_height_above_sea_level
>>>>>> model1Ssh = model1Ssh*100
>>>>>>
>>>>>> I would venture to guess that that will fix your coordinate
>>>>>> variables,
>>>>>> and therefore the lats/lons shown on your plot...
>>>>>>
>>>>>> As far as the land mask issue: As you are plotting SSH there is no
>>>>>> data
>>>>>> over land, so those areas are getting color filled white. You need to
>>>>>> tell NCL to color fill the map after the contour field gets color
>>>>>> filled. Try setting res@mpFillDrawOrder = "PostDraw", and set
>>>>>> res@mpLandFillColor = "gray70", as you are adding gray70 to your
>>>>>> colormap, and it's best to exactly specify the color your want NCL to
>>>>>> use.
>>>>>>
>>>>>> Let the group know if that doesn't solve everything..
>>>>>> Adam
>>>>>>
>>>>>> On 09/14/2011 03:34 PM, Gerald Creager wrote:
>>>>>> -------- Original Message --------
>>>>>> Subject: Converting axis coordinates
>>>>>> Date: Wed, 14 Sep 2011 16:34:13 -0500
>>>>>> From: Gerald Creager<gerry.creager@tamu.edu>
>>>>>> Reply-To: gerry.creager@tamu.edu
>>>>>> Organization: AATLT, Texas A&M University
>>>>>> To: ncl-talk@ucar.edu<ncl-talk@ucar.edu>
>>>>>>
>>>>>> I'm creating an image of a model output (attached), and the X and Y
>>>>>> coordinates are currently in model indices. I'd like to have them
>>>>>> represented as Lon/Lat.
>>>>>>
>>>>>> Also, I tried to create a gray landmask but it didn't seem to
>>>>>> work. If
>>>>>> someone's got a suggestion for that, I'd appreciate it. The ncl
>>>>>> script
>>>>>> is also attached.
>>>>>>
>>>>>> Thanks, gerry
>>>>>
>>>>
>>>
>>
>

-- 
Gerry Creager -- gerry.creager@tamu.edu
Texas Mesonet -- AATLT, Texas A&M University
Cell: 979.229.5301 Office: 979.458.4020 FAX: 979.862.3983
Office: 1700 Research Parkway Ste 160, TAMU, College Station, TX 77843
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk