[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20020319: UDUNITS date conversion: non-Gregorian calendar systems

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: 20020319: UDUNITS date conversion: non-Gregorian calendar systems
  • Date: Tue, 19 Mar 2002 09:14:52 -0700
Emanuele,

> To: address@hidden
> cc: address@hidden
> From: emanuele lombardi <address@hidden>
> Subject: udunits & years of 360 days
> Organization: ENEA  CLIM  Casaccia
> Keywords: 200203191316.g2JDGpa05055 UDUNITS

The above message contained the following:

> I have troubles using udunits to convert numbers representing days into 
> "dates" when the calendar is known to be "climatological" (360 days).
> I know some software (like ferret) handle properly such situations taking 
> in account the value of the attribute "calendar" of the input netcdf files.
> This attribute can be given the value "360" .
> 
> But my own Fortran codes cannot act this way since udunits and its related 
> routines (e.g. utcaltime) have no way to be told which calendar is to be used.
> 
> So accessing the same netcdf file from Ferret I get the correct dates, 
> while accessing it from Fortran I get wrong dates.
> 
> It seems to me that a solution could be to add an input parameter (360/365 
> integer) to the routine "utcaltime"
> or  to add the info about the calendar length in the UNIT string given to 
> routine "utdec" .
> 
> Could you, please, tell me what is the best solution and suggest me a way I 
> could "hack"
> udunits so that it can understand 360 days calendar?
> 
> Thank you very much from Italy

The UDUNITS package wasn't designed to handle calendar systems other
than the Gregorian.  As a consequence, FERRET has to handle time
parameters separately.  The same would have to be done in Fortran.
You would have to port the FERRET time-handling code to Fortran or
create your own.

I suggest that you contact the FERRET people or try posting an inquiry
to the netCDFgroup mailing-list.

Regards,
Steve Emmerson   <http://www.unidata.ucar.edu>