Re: [ferret_users] short and missing data

["Ansley C. Manke" <ansley.b.manke@xxxxxxxx>]

Title: Signature

Hi Bruno,
This is a bug in the Ferret/PyFerret handling of the _FillValue and missing flag.  Because the _FillValue or a missing_value attribute os taken into account when the data is packed, the logic is to first check the _FillValue and then scale the data.  The file is correct.  When Ferret is checking whether there is a the_FillValue or a missing_value attribute, it is doing the wrong thing in the case of  this file which has a _FillValue but no missing_value attribute.

We will get this corrected.  In the meantime, a workaround would be to add this attribute to your dataset:

   VMDR_SW1:missing_value = -32767s ;


>  ncatted -O -a missing_value,VMDR_SW1,a,s,-32767 file.nc

On 5/24/2019 9:04 AM, Bruno Levier wrote:

Hi

Could anyone explain why one of the point of the field I try to display appears as missing (and how to fix it)? See attached figure (point at 19E and 59.2N), and also the file with the data.

The file is written is short. And the value of the point which appears as missing is zero. So applying the offset and scale factor, the value of the point should be 180. I tried to modified the "FillValue" attribute but it didn't change anything. It seems like the value "zero" of a short file is considered as a missing value.

Thanks for your help,
Bruno Levier

ps: I use Ferret 7.43

--

--
Ansley Manke
NOAA/PMEL Science Data Integration Group
7600 Sand Point Way NE
206-526-6246