Re: [ferret_users] Ferret v.6 and save data in a netCDF file

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

Hi Emile,
This appears to be a bug that occurs when you write data that's simple 
time series that came in from a netcdf or other file to a new netcdf 
file. If you look at the individual input files and the output file with 
ncdump you'll notice that Ferret has failed to write the time axis units 
attribute to the output file!  This does not happen if we are writing a 
variable that has other dimensions, such as a subset of an X-Y-T variable.

> ncdump -h input.1987.01-03.nc  
  (will have the time axis with attribute TIMEAXIS:units = "DAYS since 
1900-01-01-01"  or something)

> ncdump -h output.1987.nc
  ( has the units attribute missing!)

I see where this can be fixed in the code and will get a new bugfix 
version out shortly.  I have some ideas for a workaround which I'll talk 
with Emilie with and if they seem useful will let others know.

Ansley

Emilie Vanvyve wrote:
> Hi Ferret users!
>
> I have recently changed to the last version of Ferret and now 
> encounter a problem with a script that just worked perfectly with the 
> previous version.
> All I do is to open a few netCDF files (each one = 3-month long) and 
> write some selected variables into a new one (1-year long). So mainly 
> it's just a time concatenation. My time axis is lost somewhere and I 
> just don't understand why since it didn't happen before.
>
> Here is the script:
>
> use input.1987.01-03.nc
> use input.1987.04-06.nc
> use input.1987.07-09.nc
> use input.1987.10-12.nc
> let nb = 4
>
> define symbol file = output.1987.nc
>
> save/file=($file)/clobber lon,lat,sh,isol  !a few time independent 
> variables
>
> repeat/range=1:`nb`:1/name=f ( \
>   save/file=($file)/append uairdy[d=`f`,k=40] ;\
>   save/file=($file)/append eta_tv[d=`f`] ;\
>   save/file=($file)/append evapot[d=`f`] ;\
>   save/file=($file)/append vairdy[d=`f`,k=40] ;\
>   save/file=($file)/append tairdy[d=`f`,k=40] ;\
> )
>
> Why does it not work anymore?
>
> Many thanks for any clue on this!
>
> Emilie V.
--
~>   ~>   ~>   ~>   ~>   ~>   ~>   ~>   ~>   ~>   ~> 
Ansley Manke, NOAA/PMEL   ansley.b.manke@noaa.gov
7600 Sand Point Way NE, Seattle WA
Phone 206-526-6246,  FAX 206-526-6744