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

Re: Satellite images: why 4/hr? (fwd)


===============================================================================
Robb Kambic                                Unidata Program Center
Software Engineer III                      Univ. Corp for Atmospheric Research
address@hidden             WWW: http://www.unidata.ucar.edu/
===============================================================================

---------- Forwarded message ----------
Date: Tue, 29 Dec 98 8:20:27 CST
From: Tom Whittaker <address@hidden>
To: Gilbert Sebenste <address@hidden>
Subject: Re: Satellite images: why 4/hr?

Gilbert Sebenste wrote:
> 
> That's just it. To get a good loop of images, the 15 minute imagery is
> simply beyond comparison to a loop with 60 minute increments. But for
> those who can handle the 1KM feed, why NOT make it available to those who
> want it? I mean, with the increasing technology (heck, even at NIU now
> we're on a T3!), the bandwidth and costs will go up/down respectively and
> I suspect every UNIDATA member will eventually be on the VBNS or multiple
> T3's to handle it in the next few years. In the last 3 years, we've gone
> from a 9600 baud feed from a bird for McIDAS sat pix, to a feed as fast as
> the Internet can go.
> 
> Just my $.02. Gee, can you tell I want those 1 KM pix? :-)

But...if you want to _do_ anything with the imagery then you have to
remember that it's remapped into a conformal projection and rescaled to
8 bit data.  While that's probably just fine for some applications
(including some ops work), it doesn't lend itself to much multispectral
research and instruction.  I know we've been using 8 bit data for all
the imagery so far, but I'd like to see the extra bandwidth also be
used to increase the precision (as well as more bands).

Another option for the 1km data would be to have the community agree
that most of the time they'd only want 1km images over a single,
limited (say, 640x480) region at their institution, and then to explore
the use of "pull" technology to move the data.  Sort of like about 120
"floater" images, where each school would decide what/where/when for
themselves.  Of course, we'd need a server with access to the data and
appropriate software for subsecting and sending the images...  I don't
know what's practical within the UPC budget, but this model seems more
appropriate for us than sending 25MB of image every 15 minutes to
(potentially) everyone.

I've always thought that the beauty of the IDD is that we can decide
what data is appropriate and needed...and don't have to always rely on
the content of someone else's datastream.

tom

--
Tom Whittaker  (address@hidden)
University of Wisconsin-Madison
Space Science and Engineering Center
Phone/Voicemail: 608/262-2759  
Fax: 608/263-6738