Re: : cnFillMode problem

On Mar 28, 2012, at 3:02 PM, Mary Haley wrote:

>
> I see your other question:
>
> "One quick question on your plots, did you use satellite projection on each granule?"
>
> I'm not sure what you mean, but Dave Brown may be able to answer better than I.

I'm not sure what you mean by that either. Can you explain further?
 -dave

>
> --Mary
>
> On Mar 28, 2012, at 2:53 PM, ncl-talk-owner@ucar.edu wrote:
>
>> As list administrator, your authorization is requested for the
>> following mailing list posting:
>>
>> List: ncl-talk@ucar.edu
>> From: xxi@gatech.edu
>> Subject: Re:: cnFillMode problem
>> Reason: Message body is too big: 1904055 bytes with a limit of 1500 KB
>>
>> At your convenience, visit:
>>
>> http://mailman.ucar.edu/mailman/admindb/ncl-talk
>>
>> to approve or deny the request.
>>
>> From: Xin Xi <xxi@gatech.edu>
>> Date: March 28, 2012 2:53:00 PM MDT
>> To: Mary Haley <haley@ucar.edu>
>> Cc: "ncl-talk@ucar.edu USERS" <ncl-talk@ucar.edu>
>> Subject: Re:: cnFillMode problem
>>
>>
>> Hi Mary,
>> Are you saying RasterFill doesnot work in this case?
>> I followed your advice on setting the cnMinlevel and cnMaxlevel. The plot with cell fill comes out funny (see attached) - I donot know why. By not looking good, I meant, the plot looks wiggling and has the white edges of each cell.
>> I intensionally set the lat/lon limits to show the domain I am interested at.
>> Thanks!
>> Xin
>>
>> On Wed, Mar 28, 2012 at 4:29 PM, Mary Haley <haley@ucar.edu> wrote:
>>>
>>
>> Dear Xin,
>>
>> What's happening with the raster fill is that you have some areas with no data, which get filled in with white, because you can't have transparency in a raster plot. You will need to use CellFill in order to get the transparency and still be able to see all the overlaid contours. I will update the example on the HDF page to reflect this issue.
>>
>> Can you explain what you mean by the CellFill not looking good?
>>
>> One word of caution: whenever you are comparing several contour plots and using one labelbar to represent them all, you need to make sure that you set the contour levels to the same range for every plot. If you don't do this, NCL is going to pick "nice" contour levels for each plot you create, and the levels may be different from one plot to the next, if your data range changes.
>>
>> To fix this, I added the following resources to your script:
>>
>> res@cnLevelSelectionMode = "ManualLevels"
>> res@cnMinLevelValF = 0.0
>> res@cnMaxLevelValF = 2.0
>> res@cnLevelSpacingF = 0.125
>>
>> Also, I don't know if this was intentional, but the min/max range you chose for the map (via mpMin/MaxLat/LonF) is not big enough to cover the full range of your actual lat2d/lon2d arrays.
>>
>> I've attached a version of your plot that was draw with the default AreaFill versus CellFill so you can compare.
>> <aod.127.cellfill.png><aod.127.areafill.png>
>> --Mary
>>
>>> Hello,
>>> I am making a plot following example of http://www.ncl.ucar.edu/Applications/Images/hdf4sds_5_lg.png. In the example script, the cnFillMode is mistakenly (since the comment reads "Raster Mode" and other rasterfill options are used) set to CellFill. In my script, when I set cnFillMode to RasterFill, but it gives me a blank map; interestingly, while I set it to CellFill, it gives me the aerosol optical depth, but the way CellFill does it doesnot look good. Please help out on this RasterFill vs. CellFill problem. Thanks!
>>> I have attached my script and data.
>>>
>>> Xin
>>> <script.ncl><data.zip>
>>>
>>>
>>>
>>>
>>>
>>
>>
>>
>>
>> --
>> Xin Xi
>> School of Earth and Atmospheric Sciences
>> Georgia Institute of Technology
>> 311 Ferst Drive, Atlanta, GA 30332-0340
>>
>> <MYD04_L2.aod.127.areafill.png><MYD04_L2.aod.127.cellfill.png>
>>
>>
>>
>>
>>
>
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Wed Mar 28 15:36:01 2012