Error in time unit translation

Hi,

I'm converting some grib files from my regional climate model. There seem
to be a bug in NCL's translation of the time array as it give the incorrect
reference date. Here's a
CDO sinfo output from the grib file:

 File format: GRIB
    -1 : Institut Source Ttype Levels Num Gridsize Num Dtype :
Parameter ID
     1 : unknown unknown instant 40 1 42160 1 P12 : 71.1

     2 : unknown unknown instant 40 1 42160 1 P12 : 72.1

     3 : unknown unknown instant 40 1 42160 1 P12 : 76.1

     4 : unknown unknown instant 40 1 42160 1 P12 : 11.1

   Grid coordinates :
     1 : lonlat > size : dim = 42160 nx = 310 ny = 136
                        rlon : first = 142.24 last = 278.2 inc =
0.44 degrees
                        rlat : first = -29.7 last = 29.7 inc = 0.44
 degrees
                        northpole : lon = -180 lat = 90
   Vertical coordinates :
     1 : hybrid level : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
17 18 19
                                    20 21 22 23 24 25 26 27 28 29 30 31 32
33 34 35
                                    36 37 38 39 40
   Time coordinate : unlimited steps
     RefTime = 2007-01-01 00:00:00 Units = hours Calendar =
proleptic_gregorian
  YYYY-MM-DD hh:mm:ss YYYY-MM-DD hh:mm:ss YYYY-MM-DD hh:mm:ss YYYY-MM-DD
hh:mm:ss
  2007-01-01 00:00:00 2007-01-01 03:00:00 2007-01-01 06:00:00 2007-01-01
09:00:00

snip....

But NCL interprets the file with the following output:

double initial_time0_hours ( initial_time0_hours )
long_name : initial time
units : hours since 1800-01-01 00:00

The year is off by 207 years. This is probably not a bug in NCL or maybe it
is.

I used the script "Efficient approach":
http://www.ncl.ucar.edu/Applications/gribeff.shtml

/Marston

-- 
Only the fruitful thing is true!

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Thu Apr 25 06:30:42 2013