3.3 Structure of CALACS
The CALACS package consists of four individual tasks listed in Table 3.2. These tasks are called automatically by CALACS, but each may be run separately should some variation in the normal processing be desired by the user.
Table 3.2: Tasks in the CALACS pipeline
|
CCD specific calibrations |
|
Cosmic-ray rejection task |
|
Basic MAMA and CCD calibrations |
|
|
The flow of data through the ACS calibration pipeline and the decisions made while working with associated data are diagrammed in Figure 3.1. These calibration steps are also outlined below:
- Calculate a noise model for each pixel and record in the error array
- Flag known bad pixels and saturated pixels in the data quality array
- Subtract bias-level determined from overscan regions (CCD only)
- Subtract bias image (CCD only)
- Subtract post-flash image, if required (CCD only)
- Perform cosmic-ray rejection and combining of
CR-SPLIT data (CCD only)
- Perform global linearity corrections (MAMA only)
- Scale and subtract dark image and calculate mean dark value
- Divide by flat field and multiply by gain
- Calculate photometry header keywords for flux conversion
- Calculate image statistics
Additional calibration steps are performed by MultiDrizzle, including:
- Compute offsets between dithered images (using PyDrizzle)
- Correct the geometric distortion of non-dithered images
- Perform cosmic ray rejection using the FLT files as input
- Convert the data units from electrons to electrons per second
- Combine calibrated dithered observations into a single product
As indicated in Figure 3.1, the calibration tasks have been split to handle CCD-specific calibrations (ACSCCD) separately from those steps which can be applied to any ACS data (ACS2D). For example, the MAMA data obtained with the SBC do not have the overscan regions found in CCD data.
The initial processing performed on CCD data alone is shown in Figure 3.2. The reference files appropriate for each processing step and the calibration switches controlling them are also given beside the name of the task they control. The output (overscan-trimmed image) from ACSCCD is then sent through ACS2D as shown in Figure 3.3. Raw MAMA data are processed from the start with ACS2D, which initializes the error and DQ arrays (already performed for CCD data) and which performs linearity corrections.
Figure 3.1: Flow Diagram for ACS data with CALACS task names in bold
.
Figure 3.2: Flow diagram for CCD data using ACSCCD in CALACS. 
Figure 3.3: Flow diagram for MAMA and overscan-trimmed CCD data using ACS2D in CALACS. 
