X-Sender: lynnes@localhost
Date: Tue, 15 Apr 2003 14:41:50 -0400
To: direct_broadcast_users@listserv.gsfc.nasa.gov
From: Christopher Lynnes <Christopher.S.Lynnes@nasa.gov>
Subject: New Lookup Tables for Direct Broadcast Level 1B code
Sender: owner-direct_broadcast_users@listserv.gsfc.nasa.gov
Reply-To: Christopher Lynnes <Christopher.S.Lynnes@nasa.gov>
Status:   

New lookup tables are now available for the Level 1B 
(calibrated radiance) code, at:
ftp://g0dug03u.ecs.nasa.gov/pub/.dbs/lookup_tables/

The example L1B output at:
ftp://g0dug03u.ecs.nasa.gov/pub/.dbs/examples_2_2/aqua/2003_04_06_171048
has also been updated, as my previous output had been processed with (slightly) 
obsolete lookup tables.

Note that in order to use the updated lookup tables, you must change your PCF 
files (or run_l1_aqua.pl/run_l1_terra.pl) to reflect BOTH the new lookup table 
filenames (Logical Unit Numbers 700050, 700060 and 700070), AND the 
MCSTLUTVersion (LUN 800610).

Also, the MODIS Characterization Support Team has posted some quite thorough 
documentation about the lookup tables at 
http://www.mcst.ssai.biz/mcstweb/L1B/product.html.

So far, we have been getting a new LUT every week, with Aqua and Terra 
alternating.  We will try to get the new LUTs out a day or two after they go 
into the DAAC's operational mode (i.e., after testing).  However, we may fall a 
few days behind from time to time...
======================================================
Christopher Lynnes
Code 902, NASA/GSFC
301-614-5185


******************************************************************************


X-Sender: lynnes@localhost
Date: Tue, 15 Apr 2003 16:15:49 -0400
To: direct_broadcast_users@listserv.gsfc.nasa.gov
From: Christopher Lynnes <Christopher.S.Lynnes@nasa.gov>
Subject: Re: New Lookup Tables for Direct Broadcast Level 1B code
Sender: owner-direct_broadcast_users@listserv.gsfc.nasa.gov
Reply-To: Christopher Lynnes <Christopher.S.Lynnes@nasa.gov>
Status:   

At 02:41 PM 4/15/2003, Christopher Lynnes wrote:
>New lookup tables are now available for the Level 1B (calibrated radiance) 
code, at:
>ftp://g0dug03u.ecs.nasa.gov/pub/.dbs/lookup_tables/
>
>The example L1B output at:
>ftp://g0dug03u.ecs.nasa.gov/pub/.dbs/examples_2_2/aqua/2003_04_06_171048
>has also been updated, as my previous output had been processed with (slightly) 
obsolete lookup tables.

With the weekly updates, it is fair to ask:  "Should I take the trouble to 
download the new LUTs every week and integrate them into my system?"

This is unfortunately a difficult question to answer.  In comparing an Aqua 
oceanic scene computed with LUTs 4.1.3.3 and 4.1.3.4, I found 40% of the 
EV_1KM_RefSB to be different, but none of the Emissive values.  The average 
difference of EV_1KM_RefSB was 1.6 counts, with a maximum difference of 24.

So, no doubt about it, there are some differences in the values; they are 
relatively small, but larger than, say, roundoff error.  However, it would take 
a systematic study to see whether these differences are typical either for other 
data granules, or for other lookup table changes.  (Perhaps a good undergraduate 
project...)

For those of you running higher level algorithms (cloudmask, fire, etc.), it may 
be worthwhile to run L1b data from two different lookup table versions through 
your higher levels to see if it changes the results appreciably.
======================================================
Christopher Lynnes
Code 902, NASA/GSFC
301-614-5185