Re: ESMF_regrid does not give results

[1] There is no regrid limitation with the surface variables.

[1] I think there is something inconsistent about the units
     for 'apcp' => "kg/m^2" and the
     long_name: Monthly average total precipitation at Surface

The 1st time step shows:

(0) ===> Robust Dispersion Statistics: Monthly average total
precipitation at Surface <===
(0) [0] Mean=1.1555
(0) [1] StdDev=87.4397
(0) [2] Min=0
(0) [3] LowDec=0.0335617
(0) [4] LowOct=0.0411911
(0) [5] LowSex=0.0518723
(0) [6] LowQuartile=0.077816
(0) [7] LowTri=0.123596
(0) [8] Median=0.244152 <====
(0) [9] HighTri=0.355553
(0) [10] HighQuartile=0.412014
(0) [11] HighSex=0.476109
(0) [12] HighOct=0.531044
(0) [13] HighDec=0.57835
(0) [14] Max=100.003
(0) [15] Range=100.003
[snip]

There is no way that the "Monthly average total precipitation"
is on average 0.24 kg/m2 over the NARR.

Attached is a script. I'm sure it can be improved.

See
http://www.cgd.ucar.edu/~shea/esmf_regrid.bilinear.apcp.png

On 12/17/12 9:59 PM, Kevin Kuo wrote:
> I have a perplexing problem with ESMF_regrid.
>
> I have already used ESMF_regrid to regrid some (time, level, lat, lon)
> NARR data downloaded from
> http://www.esrl.noaa.gov/psd/data/gridded/data.narr.pressure.html
>
> with success. However, I'm getting non-nonsensical results when I try
> the same script with (time, lat, lon) data from the same source, in
> particular the total accumulated precipitation and latent heat flux.
>
> The code is very simple and it has worked regridding some NARR
> datasets. I am wondering if there is a problem or limitation with
> ESMF_regrid with surface level data? I have already tried it with and
> without short2flt (for scale and offset) and thought that it might be
> a problem with the data visualization software (panopy) but have ruled
> out those sources of error.
>
> I have attached my script. If someone could test it out I would be
> very grateful. Can rcm2rgrid do the same thing? My first attempt with
> it was very poor. As a side note, when InterpMethod is set to
> "conserve" or "patch" ESMF_regrid gives an error about weights not
> generated correctly.
>
> Thank you in advance.
> --
> Kevin C. Kuo
>
>
>
> _______________________________________________
> 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