[Isocops] FW: LO comparison

[DeMajistre, Bob]

------ Forwarded Message
From: Bob DeMajistre <Bob.DeMajistre at jhuapl.edu>
Date: Fri, 9 Mar 2012 17:30:03 -0500
To: Marty Quinn <martyq92 at gmail.com>, Nathan Schwadron <nschwadron at mac.com>
Subject: LO comparison

Enclosed is a quick comparison of the output of the perl codes with a stock run of the pipeline codes.

In particular, this is a comparison of map 1 (orbits 11-33). The perl output was provided by Marty. The pipeline was run with the auto-generated good times (again provided by Marty). I’d say the results are mixed.

First, note that by default the pipeline only does maps for ESAs 5-8.  I remember conversations in the past about how maps for the first few ESAs may not mean very much (for various reasons).   I’m sure we can get this to run for the lower ESAs, but that would take a bit more time.

The first page shows a comparison of the calculated energies (in the inertial frame). The top panels are the ESA 6 calculated energies for both the pipeline and perl codes. The main things I notice (aside from the fact that they are generally similar) are 1) that there are different ‘empty’ areas in the maps. The perl codes cut more out around the magnetosphere, and there are some random missing pixels and 2) the edges of the maps are done differently. It looks like the perl codes are not ‘mixing’ ram and wake orbits like the pipeline does.

The bottom plots are just a line plots of the pixel energies for both cases. They lay on top of each other with the exception of the boundary pixels where the pipeline results pop out.

The next pages are direct comparisons of the maps in rectangular form. I would say that the maps are qualitatively similar but quantitatively different. There is no easy explanation of the differences. I was hoping to see things like one being systematically high or low, or there being a ram/wake difference, but no such luck.

I’d like to look at these differences some more. It may be that the calibration factors etc. may be different for the two methods (I’m not sure if the pipeline cals have been updated for a while).

Sorry its taken so long, I had this just about ready to write up when Maher caught the annoying memory bug...

Bob

------ End of Forwarded Message
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.sr.unh.edu/mailman/private/isocops/attachments/20120312/483b74d0/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: perlpipe_comp.pdf
Type: application/octet-stream
Size: 73671 bytes
Desc: perlpipe_comp.pdf
Url : http://lists.sr.unh.edu/mailman/private/isocops/attachments/20120312/483b74d0/perlpipe_comp-0001.pdf