Old version - superceded Nov 2010. Now that we have many years worth of CTI and background data, we can better determine the anti-correlation between ACIS CTI and background rate directly from the data. Only front-illuminated CCDs show this strong correlation; back-illuminated CCDs show very little background related variance. By sorting the CTI data into groups with equal background rate (± 1 cts/frame), we can remove any background variance. Any remaining change in the CTI must then be due to real changes.
The figure below shows the background rate (as measured by the S3 high energy reject counter) and the measured CTI as a function of time. The colors and symbols identify the equal-background groups that were used to determine the real CTI increase. The CTI increase is measured to be (3.27 ± 0.02) x 10-6 / year.
The figure below uses the same color/symbol combinations but shows the background rate as a function of CTI.
If we assume that the background-CTI relation is linear, we can find the background detrending factor that best produces the known CTI increase. This detrending factor is 3.23 ± 0.10 x 10-7 CTI / Bkg (cts/frame) such that:
Corrected CTI = Measured CTI + 3.23 x 10-7 x Bkg rate (cts/fr)
The corrected CTI is higher than the measured CTI and represents the CTI in the absence of any sacrificial charge. For example, if the measured CTI is 12.0 x 10-5 and the background rate is 60 cts/frame, the corrected CTI is 13.9 x 10-5.
This detrending factor is three times larger than that derived from alternating exposure mode data. I do not currently have a complete explanation, but it may be due to events in the framestore. Changing the frametime can simulate changes in background rate for events in the imaging array, but the dwell time in the framestore is always the same in the alternating exposure mode tests, so this additional varying background component is not simulated.
The figure below shows the measured CTI and the CTI corrected for the background variations. Most of the small deviations in the measured CTI were due to background fluctuations and do not appear in the corrected CTI. Most notable of these is the large jump due to a CME in Oct 2003 that is almost completely removed.