[ncl-talk] query bout "ESMF_regridding.ncl"

[Dennis Shea]

method conserve will do what it says 'conserve' the overall mean. It will
*not* preserve the min/max values.

method neareststod will get the nearest value to the target grid point.
That may or may not be the the min or max value in the target grid box.



On Wed, Nov 5, 2014 at 9:20 AM, Sha Feng <sfeng at mail.usf.edu> wrote:

> I was thinking "conserve" and "neareststod" . If "conserve" it not an
> option, how about "neareststod"? I couldn't find the description of this
> option on ESMF webpage.
>
> Thanks,
> -Sha
>
> On Nov 5, 2014, at 6:45 AM, Dennis Shea wrote:
>
> re: "After the interpolation, I want to largely keep the peak points in
> the source data (2D) in place instead of being smoothed out as well as
> largely keep the total amount conserved"
>
> I doubt this can be done by any of the available methods:
>
> bilinear, finite element patch recovery or first-order conservative
> regridding
>
>
>
> On Tue, Nov 4, 2014 at 10:11 PM, Sha Feng <sfeng at mail.usf.edu> wrote:
>
>> Hi there,
>>
>> I am trying to interpolate a regional 0.01 deg scattering-distributed
>> filed (source) to a 1.3-km WRF domain (destination) using ESMF regridding
>> tool but puzzled by what to be assigned to "InterpMethod" .  After the
>> interpolation, I want to largely keep the peak points in the source data
>> (2D) in place instead of being smoothed out as well as largely keep the
>> total amount conserved. Do you have suggestion for this purpose?
>>
>> A reference would be much helpful as well if available. I couldn't find
>> clear description in the ESMF_regridding.ncl code.
>>
>> Thanks,
>> -Sha
>>
>> _______________________________________________
>> ncl-talk mailing list
>> List instructions, subscriber options, unsubscribe:
>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20141105/01ed668d/attachment.html