Re: vertical interpolation

From: William Daniel Scheftic <scheftic_at_nyahnyahspammersnyahnyah>
Date: Tue Apr 22 2014 - 16:38:37 MDT

Sorry,

w =*linint1_n* (sht, ww, False, 5000., 0, 0)

On 4/22/2014 3:36 PM, William Daniel Scheftic wrote:
> The loops are causing the slowdown
>
> I think you can just do this with linint1_n
>
> w =*linint1_n* (sht, ww, False, xo, 5000., 0)
>
> w would have have dims of (/ 1, 1350, 1200 /)
>
> Take Care,
>
> Bill
>
>
> On 4/22/2014 3:22 PM, Yang Yang wrote:
>> Dear Sir/Madam:
>> I have some model-level output data from a terrain-following model. I
>> want to interpolate the model level data to a level surface, say:
>> 5000 m. I did as:
>> do i = 0, 1350-1
>> do j = 0, 1200-1
>> w(i,j) = linint1(sht(:,i,j),ww(:,i,j),False,5000.,0)
>> p(i,j) = linint1(sht(:,i,j),th(:,i,j),False,5000.,0)
>> end do
>> end do
>> Where the "sht(:,I,j)" is the height of model levels at a grid point.
>> This way takes a lot of CPU time. I am wonder if there is a better
>> way to do this. Thanks very much!
>> Ed
>>
>>
>> _______________________________________________
>> 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 Tue Apr 22 16:39:18 2014

This archive was generated by hypermail 2.1.8 : Tue Apr 29 2014 - 09:04:20 MDT