Re: how to convert CESM lat/lon grid to WRF curvilinear grid?

This is not at all clear.

"T" is from a rectilinear grid.

"T2" is the interpolated variable on the WRF grid.
It uses the map projection information on the
WRF file you provided. The 2nd plot used
the 'tfDoNDCOverlay' resource. The 3rd plot
turned off that resource an actually did a transformation.

Both plots 2 and 3 are identical.

The problem is that you need to provide the appropriate
new map projection parameters so the plots fill the screen.

You may want to write to wrfhelp@ucar.edu
It is their job to answer WRF related questions.

Good luck

On 3/21/13 3:01 AM, Chi-Yun Wang wrote:
> Hi,
>
> ...I found you used "T" for the first example.
> Sorry, I have not clarified my problem.
> My problem is why after coverted CESM grid to WRF grid, I could not use native-grid-plot to plot the domain-filled graphic which just like plotting WRF grid variables directly from WRF output.
> Thanks.
>
> Agilen
>
> ________________________________________
> 寄件者: Dennis Shea [shea@ucar.edu]
> 寄件日期: 2013年3月20日 下午 12:39
> 收件者: Chi-Yun Wang
> Cc: ncl-talk@ucar.edu
> 主旨: Re: [ncl-talk] how to convert CESM lat/lon grid to WRF curvilinear grid?
>
> Run the attached code. The cylindrical equidistant
> 'fills' the screen. Using a Lambert Conformal (LC) projection
> with the parameters on the WRF file yields the
> plot you got. Note: the data were plotted on a LC projection
> via 2 different approaches. They yield the same answer.
>
>
> On 3/18/13 11:13 PM, Chi-Yun Wang wrote:
>> Hi, Dennis,
>>
>> Thank you for the common. I had removed something unnecessary.
>> The attachments are the code, nc file I used, and the output.
>>
>> Agilen
>>
>> ________________________________________
>> 寄件者: Dennis Shea [shea@ucar.edu]
>> 寄件日期: 2013年3月19日 下午 12:09
>> 收件者: Chi-Yun Wang
>> Cc: ncl-talk@ucar.edu
>> 主旨: Re: [ncl-talk] how to convert CESM lat/lon grid to WRF curvilinear grid?
>>
>> Sending a script with no other information makes debugging difficult.
>>
>> What about
>>
>> printVarSummary(xlat)
>> printMinMax(xlat, 0)
>> printVarSummary(xlong)
>> printMinMax(xlong, 0)
>> printVarSummary(T)
>> printVarSummary(T2)
>>
>> ---
>> CESM data span 0->360
>> WRF is normally -180 to 180
>> ---
>>
>> When you have something like the following in your code
>>
>> res@mpRightCornerLonF = xlon(0,nlon-1
>> ; not sure when I set xlon(nlat-1,nlon-1) would come u
>> ; negative value ...What does this indicate?
>> print(xlat(0,0)+" "+xlon(0,0)+" "+xlat(nlat-1,nlon-1)+" "+xlon(0,nlon-1))
>>
>>
>>
>> On 3/18/13 8:36 PM, 王啟芸 wrote:
>>> Hi, NCL users,
>>>
>>>
>>>
>>> I am trying to convert CESM global model grid which is lat/lon grid to WRF curvilinear grid. I used rgrid2rcm, but the output graphic was shift and did not fill the whole domain when I used WRF native grid method to plot. The attachment is the output that I converted CESM grid to WRF grid via rgrid2rcm first and then used WRF native grid method to plot. Can anyone know the better (correct) way to convert CESM grid to WRF grid? Thanks!
>>>
>>> p.s. lat/lon and lat2d/lon2d have the same domain
>>>
>>>
>>>
>>> Agilen
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> 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 Thu Mar 21 09:16:21 2013