[Datastream #SXB-674928]: GOES East West imagery...

Hi Gilles,

re:
> Thanks for your prompt reply,

No worries.

re:
> Very interesting, and thanks also to propose to help with creating the
> composites. I do not realize the quantity of work involved to extract the
> data and to make the composites again.

I created a McIDAS-X procedure to create the GOES-East/West composites that
are being sent in the UNIWISC feed of the IDD.  Working through a set of
archived images should be straightforward as long as GOES-East and GOES-West
image sectors are available for the same times (GOES-West scans at 0 and 30
minutes past the hour; GOES-East scans at 15 and 45 minutes past the hour).

re:
> It seems I was careful enough to have saved most of the last 3 years of
> composite imagery for interesting events making the data local and backup
> the data regularly, so we have this part done already.

Very good.

re:
> Therefore the period we would need specific days of composite imagery would
> be from 2000 to 2011.

This is a long period!  Also, at some point in the past, the GOES-East and
GOES-West sectors are only available every hour, not every half hour.  I
do not recall off of the top of my head when I moved the creation of sectors
to half hourly, but I seem to recall that it was after 2000 sometime.

re:
> Unfortunately the rain gauge data from Mexican weather service are not yet
> available hourly, we only get daily accumulation (by the end of the year
> the archive should be updated to hourly data).

OK.  Does this mean that your time frame for getting composite images
is by the end of this year?

re:
> Most precipitation here is over a short time, and we want to consider only
> days with more than 20 mm of precipitation.
> 
> From early 2000 to late 2011, there are 153 such events,
> My mistake, it is not 153 days but 142, still quite a lot...

Yes, it is still a lot.  I just want to make sure that you will need
to request the days of data you want to SSEC.  We do not control the
loading of archived data back to online disk; SSEC does.

re:
> So, that means we have to look at 153 days of full day composite imagery
> (more or less 48 composite per day). So, it is quite a bit of data actually
> (at least size-wise).

Again, at some point in the past, the GOES-East and West sectors were only
available every hour, so at most there would be 24 composites for each day
for that part of the time range of interest.  Also, while the archive is
pretty complete, there could be gaps during your periods of interest.

re:
> To give you an idea, please find attached a csv formatted file (space
> separated) which includes:
> 
> year month day precipitation

OK.

re:
> Actually that would be interesting to learn all the process for making
> these composites with McIDAS-X, and also to be able to make things myself
> without asking to much from you, but I have no idea the work involved to do
> the learning and to do the job correctly afterward.

The process would be pretty much:

- download and install McIDAS-X on a workstation there

  The machine requirements are much the same as for GEMPAK, so you should
  already have a machine that would work.  The challenge is that our
  license with UW/SSEC to distribute McIDAS-X is open for U.S. universities
  but has restrictions on non-U.S. universities.  The restrictions are
  basically that McIDAS-X must be used by students in the pursuit of
  a degree, and that McIDAS-X code can not be shared outside of the
  institution.  I can send you the full list of restrictions if you are
  interested in pursuing getting McIDAS-X running in your university.

re:
> When I saved the composites for the last 3 years, it is saved in .area
> format, no idea in which format are the data before you work on them with
> McIDAS-X,

They are in AREA format as well.  In case you didn't know, AREA is the format
that UW/SSEC developed in McIDAS as a standard format for satellite imagery.

re:
> Please let me know,

I am happy to help get McIDAS-X setup and running on a machine at your
university if your use of McIDAS-X meets the restrictions of UW/SSEC.
If this happens, then I can help you setup the processing of the GOES-East
and GOES-West sectors into composites.

By the way, you have never mentioned which wavelength channel(s) you
are interested in (e.g., VIS, WV, IR).  It sounds like you may just
be interested in the IR composites (I base this assumption on your
comment about 48 composites per day which would be the number for
one wavelength channel).  If you want composites in multiple
wavelength channels (5 are available currently in the UNIWISC IDD feed,
VIS, 3.9um, WV, IR, and 13.3um), the amount of data you will need to
download from the Unidata-Wisconsin archive will increase.

Cheers,

Tom
--
****************************************************************************
Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
----------------------------------------------------------------------------
Unidata HomePage                       http://www.unidata.ucar.edu
****************************************************************************


Ticket Details
===================
Ticket ID: SXB-674928
Department: Support Datastream
Priority: Normal
Status: Closed