CALSTIS Version 2.21b was released on 2006-June-28 with the new OPUS 2006.4 release.
Update: This version of CALSTIS has not been installed in STSDAS yet, but will be incorporated into the expected fall 2006 STSDAS release. Users who want to have their data calibrated using this version should retrieve the data from the archive and should not recalibrate it locally.
The following OPRs are implemented in the release of CALSTIS 2.21b:
| OPR 55696: | Changes are needed to CALSTIS blaze shift correction. |
| OPR 55705: | CALSTIS uninitialized memory if CTECORR is not PERFORM. |
| OPR 55746: | Trace rotation needs to be implemented in CALSTIS. |
| OPR 55998: | CALSTIS6 scattered light correction with 31-arcsec slit. |
The following OPRs also went into OPUS build 2006.4:
| OPR 51357: | Extra line breaks in STIS trailer files. |
| OPR 55219: | STIS GO wavecals and fringeflats should be associated with science exposures. |
| OPR 55364: | Allow additional STIS grating-aperture combinations to calibrate in OTFR. |
These OPRs are discussed in more detail below.
OPR 55696: Changes are needed to CALSTIS blaze shift correction.
CALSTIS now properly takes into account the HST orbital velocity when calculating the expected shifts of the echelle blaze function throughput.
Part of calculating the correction involves finding the observed wavelength at the middle of the reference spectral order. If the on-board Doppler shift was applied (or for time-tag data), this should be taken out when finding the index into the wavelength array. (The index also needs to be determined slightly differently for high-res data.) Also, the blaze shift correction should not be applied to the wavelength array with the PCTAB table which gives the correction for finite extraction height.
Within this PR CALSTIS6 was also modified to write the average blaze shift as the value of BLZSHIFT.
OPR 55705: CALSTIS uninitialized memory if CTECORR is not PERFORM.
This change fixes a bug in the CALSTIS code which could cause CALSTIS (x1d extraction) to crash when the CTECORR correction is skipped for G750L or G750M data.
OPR 55746: Trace rotation needs to be implemented in CALSTIS.
The spectral traces of first order modes have been discovered to have rotated slowly over the lifetime of the STIS instrument. New trace reference files (*_1dt.fits) will contain two new columns. Column DEGPERYR will define the rate of rotation in degrees per year, while column MJD will specify the time at which the rotation is zero.
For each trace in the reference file (column A2DISPL) for a given
OPT_ELEM and CENWAVE, CALSTIS now performs a time-dependent rotation
as follows:
newA2DISPL = A2DISPL-(indgen(1024)-512.0)*tan((3.14159/180.0)*DEGPERYR*(TEXPSTRT-MJD)/365.25)
In particular for G140L two reference files will be needed to handle monthly MSM offsets that placed the target ~3" higher on the detector before 15 MARCH 1999 and ~3" lower on the detector on and after that date.
OPR 51357: Extra line breaks in STIS trailer files.
A fix was installed to avoid inserting unexpected line breaks in STIS trailer files.
OPR 55219: STIS GO wavecals and fringeflats should be associated
with science exposures.
This PR will allow user specified wavecals to be retroactively associated with spectroscopic observations and used by OTFR in the same way as auto-wavecals. Note that while this PR allows such new associations to be defined, they will not actually appear in OTFR until the data is re-processed and re-ingested into the MAST archive. The re-processing and re-ingest of the data is likely to start some time in the fall of 2006 and will take a few months to be completed.
OPR 55364: Allow additional STIS grating-aperture combinations to
calibrate in OTFR.
During review of the STIS archive, it was noted that there were a number of observations made with non-standard grating/aperture combinations which could be sensibly calibrated to produce 1D and/or 2D extracted spectra.
This PR requested modification of the calibration rules file to allow these combinations to be treated equivalently to the more standard fully supported combinations.
For each of the following gratings, observations with the listed aperture(s) are now fully calibrated by OTFR.
| OPT_ELEM | APERTURE(S) |
| E140M | 0.3X0.2, 0.2X0.09, 0.5X0.5 |
| E140H | 31X0.05NDA, 31X0.05NDB |
| E230M | 0.1X0.09, 0.3X0.2 |
| E230H | 1X0.06, 0.2X0.06, 0.3X0.2, 1X0.05NDA |
| G140L | 1x0.2, 0.3x0.2, 2X2, 31X0.05NDA, 31X0.05NDB, 31X0.05NDC |
| G140M | 36x0.05P45 |
| G230L | 1x0.2, 0.3x0.2, 2X2, 31X0.05NDA, 31X0.05NDB, 31X0.05NDC |
| G230LB | 0.1x0.09 |
| G230MB | 1X0.06 |
| G430L | 2X2 |
| G750L | 36x0.6P45 |
| G750M | 36x0.6P45, 2X2 |
In order to ensure that each of the above grating/aperture combinations can be calibrated, the PCTAB reference files were updated to include rows for each of the newly supported grating/aperture combinations.
Calibration rules for the following aperture/grating combinations were changed to match the rules currently used for the 52X0.2F1 aperture, namely, calibrate, but set X1DCORR to "OMIT":
| OPT_ELEM | APERTURE(S) |
| G230LB | 52X0.2F2 |
| G430L | 52X0.2F2 |
| G430M | 52X0.1F2, 52X0.2F2 |
| G750L | 52X0.2F2 |
| G750M | 52X0.1F1, 52X0.1F2, 52X0.2F2 |
OPR 55998: CALSTIS6 scattered light correction with 31-arcsec slit.
A bug in the scattered light correction, in 1-D spectral extraction, for data taken with the 31X0.05NDA slit, was fixed. In the before mentioned cases, CALSTIS failed with a segmentation violation. The problem was due to attempting to read outside allocated memory, because the 31-arcsec slit extends beyond the limits of the MAMA detectors. It was not noticed before because the 31-arcsec slits were not supported for 1-D spectral extraction.