[ncl-talk] wrf cape oddities

[Dennis Shea]

All questions of this type (underlying WRF computations) should be
sent to wrfhelp at ucar.edu
wrfhelp is included here.

Good Luck


On Thu, Aug 27, 2015 at 2:03 PM, Jon Meyer
<jonathan.meyer at aggiemail.usu.edu> wrote:
> Good day everyone,
>
> I've been using the wrf_user_getvar to extract max CAPE values and I'm
> seeing large regions filled with missing values. Under the hood, the getvar
> function is simply performing the necessary variable I/O and conversion to
> correct units, which are then passed into the wrf_cape_2d function.
>
> The documentation of the wrf_cape_2d function states "Previous to version
> 6.1.0, this function returned a value of -0.1 for out-of-range values. This
> has been replaced with a missing value, and a "_FillValue" attribute is also
> set."
>
> I added some print statements to the getvar function to peek at the arrays
> being passed into the wrf_cape_2d function and there are no missing values
> in any of the arrays, and the min/max range is appropriate for all arrays*,
> yet the cape values returned have large regions of missing (although lots of
> reasonable CAPE values are returned). The min value of CAPE that is returned
> is 0.1, so I'm curious about whether the missing values should be set to
> zero values, or left as missing, which will impact subsequent analysis.
>
> I'm wondering what the out-of-range values represent physically? What causes
> an out-of-range value, as opposed to a small value such as the 0.1's that
> are being returned. Without being able to crack open the wrf_cape_2d
> function, I'm hoping someone with experience there can chime in.
>
> Thanks in advance,
> Jon
>
> *As a side note, the documentation for wrf_cape_2d states the PSFC field
> needs to be in hPa. The getvar function looks to be passing in Pa but with
> reasonable CAPE fields returned, I'm guessing there is some login in the
> wrf_cape_2d function that sets correct units. I tested by changing the units
> from Pa to hPa as the documentation requests, and the number of missing
> values being returned increases, so something goofy is going on that I'm not
> able to see unless I can crack open the wrf_cape_2d function
>
>
>
> _______________________________________________
> ncl-talk mailing list
> ncl-talk at ucar.edu
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>