Re: pynio/grib question

From: David Ian Brown <dbrown_at_nyahnyahspammersnyahnyah>
Date: Mon, 18 Sep 2006 11:56:20 -0600

Hi Jeff,
You are correct that PyNIO does not currently give any indication of
the precision of the data in
any attribute similar to the CDC least_significant_digit attribute. I
can look into this as a possible enhancement.
However, a problem that I can foresee based on a quick perusal of some
files with wgrib is that because the
packing for each GRIB record is handled individually, based, I believe,
on the range of the data within the
record, the least significant digit may not be constant for all
horizontal slices of a PyNIO GRIB variable. I would have to
investigate further to know whether this is really the case. If it is,
I suppose one could provide a maximum and
minimum least significant digit for the variable as a whole.

On Sep 18, 2006, at 10:55 AM, Mary Haley wrote:

> Hi Jeff,
> I'm hoping Dave Brown can answer this question. I'm CC-ing him in case
> he didn't see your message.
> --Mary
> On Fri, 15 Sep 2006, Jeff Whitaker wrote:
>> Hi all:
>> When opening a grib1 file with PyNIO, is there any way to find out
>> what the precision of the grib data is? For example, geopotential
>> height data is usually packed in grib so that it has a precision of 1
>> meter (there are no significant digits to the right of the decimal
>> point). I presume this information is in the binary data section
>> (BDS) of the grib file, but I don't think it is converted to a
>> variable attribute in PyNIO. What I'm looking for is something like
>> the least_significant_digit attribute in the CDC netCDF conventions
>> (
>> -Jeff
>> --
>> Jeffrey S. Whitaker Phone : (303)497-6313
>> Meteorologist FAX : (303)497-6449
>> NOAA/OAR/PSD R/PSD1 Email :
>> 325 Broadway Office : Skaggs Research Cntr 1D-124
>> Boulder, CO, USA 80303-3328 Web :
>> _______________________________________________
>> pyngl-talk mailing list

pyngl-talk mailing list
Received on Mon Sep 18 2006 - 11:56:20 MDT

This archive was generated by hypermail 2.2.0 : Wed Sep 27 2006 - 09:59:34 MDT