Clean ACIS Background in VFAINT Mode

Related acis_process_events threads

There are other threads that should be considered, since they may affect how acis_process_events is run. The Create a New Level=2 Event File thread shows how to combine all of these options into a single run of acis_process_events.

• Apply the Time-Dependent ACIS Gain Correction
• Apply an ACIS Gain Map
• Remove Pixel Randomization
• Apply/Remove PHA Randomization
• Apply the ACIS CTI Correction

Determine the eventdef parameter

If you are unsure of the event mode of your observation, the information can be found in the READMODE and DATAMODE values stored in the file header:

unix% dmkeypar acisf00884_000N001_evt1.fits READMODE echo+
TIMED

unix% dmkeypar acisf00884_000N001_evt1.fits DATAMODE echo+
VFAINT

This is indeed a very faint mode observation; all VFAINT observations will also have READMODE equal to TIMED. The proper eventdef parameter, used in the next section, is "stdlev1". The full parameter syntax of the eventdef string may be found in plist acis_process_events.

Run acis_process_events

Running acis_process_events with the SDP level=1 event file as the input and with check_vf_pha=yes will produce a new level=1 event file with possible background events flagged in status bit 23.

At the same time, the latest CALDB will be applied, meaning that the newest gain map will be picked up. The ACIS CTI correction is also applied while cleaning the background. This may not be appropriate for your source; please make sure to read the Combining with ACIS CTI Correction caveat.

unix% punlearn acis_process_events
unix% pset acis_process_events infile=acisf00884_000N001_evt1.fits
unix% pset acis_process_events outfile=acis_884_new_evt1.fits
unix% pset acis_process_events badpixfile=acis_884_new_bpix1.fits
unix% pset acis_process_events eventdef=")stdlev1"
unix% pset acis_process_events stop=none
unix% pset acis_process_events check_vf_pha=yes
unix% acis_process_events
Input event file or stack (acisf00884_000N001_evt1.fits): 
Output event file name (acis_884_new_evt1.fits): 
aspect offset file ( NONE | none | <filename>) (NONE): 

It is important to note the unusual syntax of the eventdef parameter; the tool will not access the predefined string if the leading ")" is missing (see example 6 of ahelp parameter).

The content of the parameter file may be checked using plist acis_process_events.

You may see a warning about the number of event islands that contain one or more bad pixels:

# acis_process_events (CIAO 3.4): The following error occurred 26941
  times: dsAFEBADPCNTERR -- WARNING: Event island contains 1 or more bad pixels.

It is explained in this FAQ and may be ignored.

The trail parameter is also used when check_vf_pha=yes. The idea is that for the pixels above the center of the event island, the amount of charge in these pixels is due in part to a fraction of the charge that is trailed from the other pixels as a result of charge transfer inefficiency (CTI). In this case, the charge due to the CTI trail should be ignored when checking to see if the pixel is above the split threshold. By default the algorithm ignores 2.7% of the charge in the central pixels. This value comes from empirical data taken at the focal plane temperature of -110C. Those experiments have shown very little impact of trail parameter on the particle rejection efficiency. This is likely due to the fact that the CTI coefficient only affects testing of 1 or 2 of the 16 border pixels. Even if the value is far off, the rejection efficiency is degraded by 10% at most - that is, you reject 18% (instead of 20%) of the particles.

Apply grade/status filters

Filter for bad grades (using ASCA grades) and for a "clean" status column (i.e. all bits set to 0). Since the background particles are flagged as one ("bad"), they will be filtered out of the event file:

unix% punlearn dmcopy
unix% dmcopy "acis_884_new_evt1.fits[EVENTS][grade=0,2,3,4,6,status=0]" \
      acis_884_flt_evt1.fits

Apply GTI filters

The Good Time Intervals (GTIs) supplied by the pipeline now need to be applied. Simultaneously, an unnecessary column is eliminated from the output:

unix% punlearn dmcopy
unix% dmcopy \
      "acis_884_flt_evt1.fits[EVENTS][@acisf00884_000N001_flt1.fits][cols -phas]" \
      acis_884_evt2.fits 

Be sure to include the @ symbol in the filter expression; the command will not be executed properly if it is omitted.

Elimination of real events

To see if this is a problem in your data, create a file of the photons rejected by the check_vf_pha screening. Here the filter is defined with status bit 23 marked as "bad" (1) and all other bits may be good or bad (x); note that the bits are counted from the right, starting at zero. This filter retains those events that have been flagged on account of the 5x5 screening:

unix% dmcopy "acis_884_new_evt1.fits[EVENTS][status=xxxxxxxx1xxxxxxxxxxxxxxxxxxxxxxx]" \
      acis_884_23bad.fits     

Figure 1  shows acis_884_new_evt1.fits (left) and acis_884_23bad.fits (right) displayed side-by-side. As you can see, there are some source photons identified as background particles by acis_process_events; this is evident because the photons are clumped around the source location. In this case, you may decide to keep the photons (i.e. not use the newly screened file).

If the rejected photons are uniform across the field, then pile-up was not a problem and it is perfectly safe to filter out those events. There are more examples of this comparison on the Reducing ACIS Quiescent Background Using Very Faint Mode page.

Combining with destreak

When the destreak tool is run on data that has been processed with check_vf_pha=yes, it is not as efficient at detecting (and removing) streak events because of the large number of nonzero status bits set by the background-cleaning algorithm. While the destreak mask parameter may now be used to include the nonzero status events in the destreaking, the following is still the recommended method for combining these two processes:

1. run destreak on the level=1 event file, as shown in the Destreak the ACIS-S4 Chip thread
2. perform the VFAINT background cleaning with acis_process_events (i.e. run this thread)

Combining with ACIS CTI Correction

It has been determined that for faint sources the number of events flagged by the VFAINT algorithm is insensitive to the use of the CTI adjustment. As a result, the VFAINT background algorithm and the CTI adjustment may be used in the same run of acis_process_events, as shown in this thread.

If the source is not faint, it is suspected that simultaneous use of the VFAINT background algorithm and the CTI adjustment may flag more source events as background. In this case, one may wish to apply the VF mode filter first, then rerun acis_process_events with the CTI adjustment. To do so:

1. Run acis_process_events as written in this thread, but turn off the CTI correction and the time-dependent gain correction:

   unix% pset acis_process_events apply_cti=no
   unix% pset acis_process_events doevtgrade=no
   unix% pset acis_process_events apply_tgain=no
   

   It is not necessary to recalculate the event grading (doevtgrade=no). Since the tgain correction requires CTI to be applied at the same time, it is turned off to avoid error messages.

2. Re-run acis_process_events to apply the CTI, as shown in the Apply the ACIS CTI Correction thread.