Re: [ferret_users] a time axis for a single point?

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

Hi Helene,
You can do this following the part of the Users Guide about "NetCDF, 
append  time steps", but instead of reading in a set of ASCII files and 
appending them to the netcdf file, you will read the first month's data, 
compute the mean, and then save that to a file. You can define the mean 
by regridding to a time axis with one time cell, defined to span the 
month's data:

DEFINE 
AXIS/EDGES/T="01-NOV-2001:00:43":"01-DEC-2001:00:18"/EDGES/NPOINTS=1/units=  
/T0=   tmonth

(with your choice of units and time origin)

LET bt_mean = bt[gt=tmonth]    ! or another regridding transformation if 
you wish
SAVE/file=outputfile.nc bt_mean

Then read the second month's data, redefine the time axis tmonth to 
correspond to the time range of that data, regrid to form the mean, and 
use  SAVE/APPEND to add it to the file. 
Ansley

Helene Brogniez wrote:

> Hi,
>
> I am working on the strorage of satellite data. I have 2 files of 
> observations for the months of November and December 2001, with a 
> quite irregular time step: for each month I have around 432 values, 
> corresponding to 1h30 of observations.
> I would like to concatenate my 2 files with the data regridded to 
> monthly means (weel, I do know how to make the time average ;-) ), 
> such as the first mean will be associated to t="15-nov-2001" and the 
> second mean to t="15-dec-2001".
>
> One of the file looks like:
>
> yes? use AMSUB_2.5d_11.2001.nc
> yes? sh da
>     currently SET data sets:
>    1> ./AMSUB_2.5d_11.2001.nc  (default)
> name     title                             I         J         
> K         L
> BT       BT (K)                           1:145     1:25      
> ...       1:432
>
> yes? stats BT
>
>             BT (K)
>             LONGITUDE: 178.8E(-181.2) to 178.8W
>             LATITUDE: 31.2S to 31.2N
>             Z:  N/A
>             TIME: 01-NOV-2001 00:43 to 01-DEC-2001 00:18
>             DATA SET: ./AMSUB_2.5d_11.2001.nc
>
> Total # of data points: 1566000 (145*25*1*432)
> # flagged as bad  data: 1383895
> Minimum value: 188.69
> Maximum value: 278.88
> Mean    value: 252.02 (unweighted average)
> Standard deviation: 7.9381
>
>
>
>
> thanks in advance,
> Helene