2MASS "Hot-Pixel" Analysis: Calibration vs. Science Scans

2MASS "Hot-Pixel" Analysis: Calibration vs. Science Scans

This is a comparison between hot pixels for
68 970424n calibration scans and 28 970424n science scans, and also for
35 970622n calibration scans and 28 970622n science scans.

Figure 1--- Comparison of Cal vs. Sci Scan FREX Distributions
DateScan Type
(# Scans)
#OR Hot Pixels Sum of FREX Images#Scans with FREX>=4 HistogramCDF
970424nCal
(68)
20
Tabular Data
OR mask in Fowler's Format
FITS Image

k.scan_cnt.tbl
Sci
(28)
40
Tabular Data
OR mask in Fowler's Format
FITS Image

k.scan_cnt.tbl
970622nCal
(35)
27
Tabular Data
OR mask in Fowler's Format
FITS Image

k.scan_cnt.tbl
Sci
(28)
52
Tabular Data
OR mask in Fowler's Format
FITS Image

k.scan_cnt.tbl

Total Hot Pixel Counts (OR mask) in Calbration vs. Science Scans

For an equal number of cal scans and sci scans, if hot pixels events are random in time, we would expect (exp_time_sci/exp_time_cal) = 6 times more hot pixels in sci scans than in cal scans. On 970424n we have (28/68)=0.4 as many sci scans as cal scans, so we expect 6*0.8 = 2.4 times as many hot pixels in the science scans; we observe (40/20)=2.0 times as many. Similar logic applied to the 970622n data means we expect 6*(28/35) = 4.8 times as many hot pixels in the science scans; we observed (52/27)= 1.9 times as many.

Conclusion: The chance a hot pixel event will occur is not random in time.

Persistent Hot Pixels (AND mask) in Calbration vs. Science Scans

On 970424n there is no hot pixel persitent across all K-array 68 calibration scans; likewise for the 28 science scans.

On 970622n, there is no hot pixel persitent across all K-array 68 calibration scans on 970424n; however among the 28 science scans there is 1 persistently hot pixel: (53,251). Note also that pixel (53,251) does trigger as "hot" in 32 out of the 35 calibration scans on this night.

Commonality Between Hot Pixels in Sci. Scans and Cal. Scans

970424n:

970622n:


Last update: 26-Aug-97, Joe Mazzarella (mazz@ipac.caltech.edu )