Re: calculating laplacian in NCL

Hi Dennis,
  Lap.f90 is your code, I only corrected a typo. Please see attached code.

NCL code:

input variable:
Variable: ssta
Type: float
Total Size: 51840 bytes
           12960 values
Number of Dimensions: 2
Dimensions and sizes: [lat | 90] x [lon | 144]
Coordinates:
           lat: [ -89.. 89]
           lon: [1.25..358.75]
Number Of Attributes: 5
 time : 197001
 missing_value : 1e+20
 units : W/m^2
 longname : monthly surface upward longwave radiation
 _FillValue : 1e+20

  Thanks
Mingxuan
On Feb 29, 2012, at 8:54 PM, Dennis Shea wrote:

> This is not clear. What is lap.f90?
>
> What does printVarSummary(...) of the input variable show?
>
> Include a *clean* version of your compute code/
>
> On 2/28/12 5:08 PM, Mingxuan Chen wrote:
>> Hi Dennis,
>> Please see test results. Both figures are from same file (
>> 144X90,without missing values). Figure1 is using lap.f90, second one is
>> using ncl function lapsF to calculate laplacian.
>> How do you evaluate the lap.f90?
>> Thanks
>> Mingxuan
>>
>> Fig1
>>
>>
>> Fig2
>>
>>
>>
>>>
>>>
>>> On Feb 24, 2012, at 3:43 PM, Dennis Shea wrote:
>>>
>>>> This is offline.
>>>>
>>>> Maybe someone has done it. Don't know.
>>>>
>>>> Attached is an old fortran subroutine that is untested.
>>>>
>>>>
>>>> On 02/24/2012 10:26 AM, Mingxuan Chen wrote:
>>>>> Hello,
>>>>> There are functions lapsF and lapsG to calculate laplacian in NCL,
>>>>> but they don't handle missing values. How do calculate laplacian
>>>>> with missing values in NCL? Did anyone already write a
>>>>> function/subroutine to handle missing values?
>>>>> Thank you for your help.
>>>>> Mingxuan
>>>>> _______________________________________________
>>>>> ncl-talk mailing list
>>>>> List instructions, subscriber options, unsubscribe:
>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>
>>>> <lap.f>
>>>
>>

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk