This page (revision-83) was last changed on 02-Feb-2017 13:19 by David R Williams

This page was created on 27-Jan-2009 03:17 by David R Williams

Only authorized users are allowed to rename pages.

Only authorized users are allowed to delete pages.

Page revision history

Version Date Modified Size Author Changes ... Change note
83 02-Feb-2017 13:19 43 KB David R Williams to previous
82 19-Mar-2009 13:50 41 KB David R Williams to previous | to last
81 18-Mar-2009 21:12 40 KB David R Williams to previous | to last

Page References

Incoming links Outgoing links
TeamSciMeetingMar09
TeamSciMeetingMar09

Version management

Difference between version and

At line 42 changed one line
* Warm pixels have mean + (5 sigma) to mean + (50 sigma) pedestal values. The rate of increase has had multiple level-offs. After the levelling off in 2008, there was an increase again in July, then levelled off in December. Seems to be a temperature dependence throughout Earth's orbit about the Sun (January is hottest).
* Warm pixels have mean + (5 sigma) to mean + (50 sigma) pedestal values. The rate of increase has had multiple level-offs. After the levelling off in 2008, there was an increase again in July, then levelled off in December. Seems to be a temperature dependence throughout Earth's orbit about the Sun (January is hottest). By 2010, 16% of pixels may be affected; 2012, 26% might be affected. Call on PRY to present happier news about this, though...
At line 44 added 22 lines
* Peter Young
Sounds like a huge fraction of the CCD.
What's the best way to treat warm pixels for scientific analysis. My thought from the start of the mission was to ignore them, and neglect them in the analysis. But HPW and PRY talked and found that interpolating actually gave good results.
Wrote a document to be posted on this Wiki after the meeting presenting some of the results.
Used the standard EIS_PREP processing.
Then artificially inserted 30% bad pixels. So how badly were the fits degraded by ignoring these fake "bad pixels".
HEM: Interpolatin after fitting?
PRY: This is done by EIS_PREP. And is done in the solar_Y direction
KPD: How do you know they're better fits?
PRY: I have the original data, and identify the places where there never were bad pixels. Then I process these data by inserting fake bad pixels, interpolate, and then compare the line fit parameters.
KPD: I think you're inventing data...
PRY: That was my worry, too, but it seems to work!
HEM: What about clusters of missing pixels?
PRY: There are different methods of tackling these.
You could reproduce the original data 97%, but the bad interpolations tend to happen in the high-intensity areas. My recommendation is to use interpolated data rather than just ignoring the missing pixels. I was amazed how well you could do even with 30% warm pixels, but it seems we can continue longer than expected by a couple, even if we don't remove them.
It may be related to EIS's oversampling w.r.t. its resolution, so there is correlation of data.
This page (revision-83) was last changed on 02-Feb-2017 13:19 by David R WilliamsTop