[ferret_users] Re: Your Ferret mail list questions

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

Hi,
We'll try subscribing you to the Ferret list under both @me.com and 
@mac.com.  Next time you have a question try writing to the ferret_users 
address again and we'll see what happens.

If you always set the time to 1, then you'll be over-writing the data at 
each step. The netCDF library and Ferret are saying, oh here's some 
replacement data for time step 1.

Define a 12-step time axis.  It doesn't have to be in terms of calendar 
months, just /t=1:12:1 if that's what you want.  And then when you're 
appending, add  the timestamp 0*t[gt=timeaxis,t=($imon)] to the variable 
before writing. I think that'll do it.

Ansley


On 11/30/2011 11:53 AM, mpm33@xxxxxx wrote:
> Hi Ansley,
>
> Thanks so much for the quick response.  @me.com and @mac.com are essentially the same email address.  mac.com was replaced by me.com recently.  but perhaps this is the problem, the emails are still sent from @mac.com and not @me.com?? hence bouncing as not a member
>
> Thanks for the help, Im really stuck with this issue!  Unfortunately, I did try to first do that timestamp creation via the following:
>
> 1) make new netcdfs that have time stamps of 1.  This worked as i opened the "over.nc" files and they have a timestamp of 1, using the command below:
>
> repeat/range=1:12/name=m (define symbol imon = `m` ; if `m lt 10` then define symbol imon=0`m`;define symbol fname = MIL3MAE.004_2001($imon)01_000000.G3.nc ; set data ($fname); save/clobber/file=MIL3MAE.004_2001($imon)01_000000.G3_over.nc OPTICAL_DEPTH_AVERAGE_GREEN + 0*t[T=1]; cancel data ($fname) ; cancel memory )
>
> 2) then retry this command with this statement but it still only prints the last value.  its odd but when i create the new _over.nc files they change the var name to E436 or some other E### value but do give it an L of 1.  Not sure if this is an issue but I adjust accordingly in the statement below.
>
> repeat/range=1:12/name=m (define symbol imon = `m` ; if `m lt 10` then define symbol imon=0`m`;define symbol fname = MIL3MAE.004_2001($imon)01_000000.G3_over.nc ; set data ($fname); save/file=misr_2001AOD.nc/append E436[l=1]; cancel data ($fname) ; cancel memory )
>
> Nevertheless it doesnt work.  Really frustrating...
>
> Marc
>
>
>
> On Nov 30, 2011, at 2:46 PM, Ansley Manke wrote:
>
> > Hi Marc-
> > There seems to be something wrong with your membership on the Ferret List. You asked to be subscribed earlier today under the address mpm33@xxxxxx, and that subscription seemed to go through successfully, but your question bounced back to me with some error message from our mail servers.  I have inquired about why that is.
> >
> > The message you wrote from the address mpm33@xxxxxxx also bounced, because that mail address is not subscribed to the list.  I'll let you know what I find out about all of this.
> >
> > In the meantime, I can get you started with an answer to your question. The topic you want to look up in the Ferret documentation is "timestamp": defining a time axis where the variable has no time axis. There is an index entry pointing to that topic within the documentation for SAVE. You are correct that this is the source of your problem with the data.
> >
> > Ansley