Transcript Persistence is

Persistence in the WFC3 IR detector
Knox S. Long, Sylvia Baggett,
Susana Deustua, and Adam Riess
STScI 2010 Calibration Workshop
1
Summary
 Persistence is a residual image observed in most types of IR arrays
 The WFC3 IR detectors exhibit persistence from sources that approach or
exceed full well

Persistence is most easily detectable when bright targets are observed in previous
visits (of others)

But self-persistence also occurs when exposures containing bright targets are dithered
on the detector within a visit

Typically persistence results in signals of 0.2 electrons s-1 , 1000 s after a saturated
exposure.
• Persistence exhibits a power law decay with time
 Persistence can be both scientifically and cosmetically deleterious
 The effects of persistence are reduced by dithering
 Persistence can be estimated from the time history of illumination in earlier
exposures
 Post-processing can remove about 90% of the persistence signal with
algorithms that track the history of the exposure
STScI 2010 Calibration Workshop
2
Basic reason for persistence is understood
Trapped Trapped
electrons holes
Mobile
electrons
Depleted
Mobile
holes
- - - --+
- - - -- - -
- - - -- - -
++ +
+ +
+ ++
++ +
+ +
+ ++
++ +
+ +
high flux
signal
reset
dark idle
(large reverse bias)
All traps have released
their charge in depletion
region
R.Smith, SPIE 7021-22, Marseille 2008-06-24
(low bias)
As signal
accumulates the
depletion width is
reduced. Traps
newly exposed to
charge can
capture some
mobile carriers.
(large reverse
bias)
At “reset” the
wider depletion
region is restored,
but trapped charge
stays behind.
STScI 2010 Calibration Workshop
- - - -++ +
+ +
N
P
next dark exp.
(small bias
reduction)
The released charge
reduces the bias
voltage. persistence
3
Persistence is a function of total exposure
 Stimulus is electrons, not
a rate like electrons/s

Short exposures of very bright
objects

Long exposures of fairly
bright objects
 The response is Fermilike

negligible at low levels

saturates above about 105
elec.

Slow rise at very high welldepth levels
STScI 2010 Calibration Workshop
4
Persistence is ~ a power law function of time
 Persistence is ~ a power
law function of time
 Persistence is similar
over the entire array

Slight gradient and some
features
 Example:
• Used tungsten lamp
expose the array at levels
from 50,000 to 1.5 106 e
STScI 2010 Calibration Workshop
5
Finding persistence in your images
Ex. 1
 Inspect histogram
equalized images

Look for obvious patterns

Look for objects that
appear mushy
 Use multidrizzle to find
residuals

Subtract the last single
science image from the
first
Ex. 2
STScI 2010 Calibration Workshop
6
What was the problem?
Bright Stars in HII
region
47 Tuc
MAST search will find offenders:
http://archive.stsci.edu/hst/history_search.html
STScI 2010 Calibration Workshop
7
Mitigating persistence in individual images
 Assume persistence image for
each prior exposure

Pi(x,y,t)=F(x,y) (t/to)-g
 Assume only the highest exposure
level matters

Pf=Max(Pi)
 Simply subtract the persistence
image from the affected exposure
 Currently testing this algorithm on
individual cases using a python
routine that modifies the _flt.fits
files to remove the persistence
 Issue: Need access to proprietary
data to generate the persistence
image.
STScI 2010 Calibration Workshop
8
Summary – Persistence is not a virtue, but …
 Today
 Tomorrow

IR observations prohibited
after some “bad actors” in
Cycle 17

Dithering helps

A working model exists

In house tool exists to
mitigate persistence on caseby-case basis

Contact [email protected] if you
notice persistence and want
help your images

Continue checks for “bad
actors” in Cycle 18, but too
many constraints reduce
efficiency

Characterize persistence on a
pixel by pixel basis

Considering:
• Migrating tool into pyraf
• Standard production of
persistence images
More info: http://www.stsci.edu/wiki/WFC3/Persistence
STScI 2010 Calibration Workshop
9