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

[IDV #PMI-123940]: IDV - MSG channels 13 and 14

HP-


> >> Sorry to be again prolific in support requests. Here an other one.
> >>
> >
> > Sorry to be so unresponsive on our end.  We were in Oklahoma last
> > week teaching a workshop and are getting sidetracked left and
> > right.
> >
> As long as you get sidetracked symmetrically it's okay with me. I don't
> like too much right-wing ;-). BTW one of the Swiss German folklore
> things is Fahnenschwingen (Swiss flag brandishing and hurling) and there
> all the movements made with the flag (minimum side length 1.18m) shall
> equally distributed among left and right turns!

Well, I usually lean left, but I'll try to be more balanced. ;-)

> >> As you might (not) be aware Marianne's MSG server is providing 2 additonal
> >> channels beyond 1-11 and 12. They are channels 13 and 14. These 2
> >> channels contain the solar and thermal contribution respectively, of
> >> channel 4 (3.9micron).
> >>
> >> Would it be possible to make them available thorugh localhost? Under McIDAS
> >> you can access them under these band numbers, but only for archived data.
> >> Up to now they are only implemented on my local server and not on the ADDE!
> >>
> >> My plan is to migrate Marianne's Fortran code (which she actually emulated 
> >> from
> >> my IDL code ;-)) to Jython this autumn-winter.
> >>
> >
> > I'm not clear on what you are asking.  Here's what I understand you
> > to say:
> >
> > 1) you get channel's 13 and 14 when you do an IMGLIST in McIDAS on
> > archived data.
> >
> No, IMGLIST is only showing the real channels. Channels 123 and 14 get
> calculated on the fly when you access them explicitely through the BAND
> keyword, e.g IMGDISP 1, N.1 BAND=13

Since the image data server (msgaget?) knows how to deal with these
extra bands, it would be nice if the image directory server (msgadir)
would list them.  The data server must know how to create the appropriate
AREA header needed for the directory server, so the code is already there.
That would be the best solution. 

> > 2) when you look at the same datatsets through the IDV, they are not
> > listed.
> >
> No. You would porbably have to emulate the call that is issued via the
> BAND keyword in McIDAS, or similar?
> > 3) for realtime data (EUM_AD), these are not listed for the M9
> > dataset in either McIDAS or IDV
> >
> The EUM_AD server has these virtual bands 12/14 not implemented (yet,
> due to priority conflicts).

Okay, so the basic problem is that the directory server doesn't list
these and that's what the IDV GUI uses to populate the list of
available bands.  Otherwise, we have no way of knowing what other
bands are available.

> > If your local McIDAS system is set up so that the ADDE server and
> > the McIDAS session are both pointing to the same enviroment (same
> > executables), you should see the same bands in the IDV that you
> > see in McIDAS.  Are they sharing the same environment?
> They are sharing the same environment including Marianne's server with
> the virutal bands extraction. But as already said the bands are only
> extracted via the BAND keyword.
> > Is it
> > possible that the remote ADDE server is not picking up
> > Marianne's server, but the default MSG server?
> >
> Not implemented there (yet). See above.
> > As for the difference between realtime and archived data, is it
> > possible that the bands are added during the archiving?
> No, they are created on the fly when called. See above.
> > Or
> > are they running a different version of the server that does not
> > have this calculation in it?
> Yes, See above.
> >
> >
> > If I'm misunderstanding the situation, please let me know.
> >
> Half-half. I hope my indications have mage it a little bit clearer.

Through the GUI, adding in the extra bands is not possible at this
point, sorry.   I don't want to hardcode something like this.

You could write a Jython procedure to do the read for you.  Depending
on whether you are accessing data by time or relative position,  it
can be easy or harder. ;-)

For a relative position, you could have a procedure like:

def loadImages(num,band):
  num = int(num)
  a=[]
  for i in range(num):
     
a.append("adde://adde.ucar.edu/imagedata?group=RTIMAGES&descr=MOLL-IR&band="+band+"&pos=-"+str(num-i))
  makeDataSource(a, "ADDE.IMAGE")

where you would just evaluate the formula to load in the data.  The 
corresponding
formula would pass in the number of images and the band number.  If you needed
to subset the image, you could add in other keywords.  Docs are at:

http://www.ssec.wisc.edu/~dglo/visad/edu/wisc/ssec/mcidas/adde/AddeURLConnection.html


For time specific images, you would have  to use the &day=<value>&time=<val>
options in the ADDE URL to specify the day and time.

But, in the long run it would be easier for Marianne to change her
directory server. ;-)

Let me know if you have questions about the syntax of the Jython or formula.

Don
   




Ticket Details
===================
Ticket ID: PMI-123940
Department: Support IDV
Priority: Normal
Status: Open