July 2018 STAN
This STAN provides various updates to STIS calibration, including the Exposure Time Calculator (ETC), the new E140M blaze shift reference file, the current levels and impact of jitter, spatial scanning, and the reprocessing status of pre-SM4 data.
Updates to ETC Version 26.1
Version 26.1 of the Exposure Time Calculator (ETC) was delivered in advance of the Cycle 26 call for proposals. It includes multiple data updates to better reflect the predicted state of the instrument at mid-cycle 26, or May 01, 2019.
For all STIS ETC pages, the dark rates for the CCD and FUV-MAMA, and the read-out noise for the CCD have been updated to reflect the latest performance of the instrument. Throughputs for the NUV modes have also been updated to reflect the expected levels based on recent fits to the TDS monitoring data of white dwarf standards. Details may be found in the Detector Background section of the ETC User's Guide.
More Accurate E140M Blaze Function Shifts in FUV PHOTTAB Delivery
Accurate flux calibration of STIS echelle observations depends on the correct definition and alignment of the spectral blaze function, or characteristic efficiency of each spectral order (as described in August 2017 STAN and STIS ISR 2018-01). The blaze function is impacted by shifts on the detectors and shape variations over time that are due to changes in the alignment of the STIS echelle optics. These changes can result in flux discrepancies of 10% or more in overlapping spectral regions of adjacent orders (see Figure 1). Flux mismatches of this magnitude can become problematic when combining all spectral orders into a single, continuous flux-calibrated spectrum over the full echelle spectral range.
The blaze function shift is calculated in the CALSTIS pipeline as a function of spatial location of the spectrum on the detector and as a function of observation date. After SM4 the time coefficients of the blaze function shift for all echelle modes, including the E140M grating, were set to zero in the PHOTTAB reference files when new sensitivity curves were derived. The temporal component of the blaze function shift has evolved as time has elapsed since SM4, and requires characterization for proper flux calibration. Time coefficients for the E140H mode settings were previously added to the FUV-MAMA PHOTTAB in May, 2017. The new reference file, 26p1601ko_pht.fits, now includes time coefficients for the E140M blaze function shifts.
When the new coefficients were applied to recent E140M observations in the CALSTIS pipeline, some new artifacts appeared in the flux calibrated spectral orders. These features include spikes in the flux values at the bluest ~1-2 Angstroms of a subset of spectral orders (see Figure 2), as well as a noticeable curvature in some of the longest wavelength orders (see Figure 3). Further analysis has revealed that the blaze shapes of the E140M orders have changed with time, in addition to the progression of the blaze shifts. Despite the introduction of the new artifacts, the inclusion of the time-dependent coefficients for the blaze function shifts does improve the flux agreement in overlapping spectral regions for the majority of E140M data sets. The average relative flux agreement for overlapping regions with the updated reference file is within our relative photometric flux accuracy requirement of 5% (STIS Instrument Handbook, Table 16.2), for 85% of post-SM4 data sets.
For users who are severely impacted by the spikes at the shortest 1-2 Angstroms of the order edges when merging the spectral orders together, we recommend trimming the affected spectral regions before combining the spectral orders, for now. (Note that the spikes at the blue edges of the orders are only an artifact present in the flux calibrated spectra and not in the observed net and gross count rates.) To address the changes in shapes of the blaze functions, the STIS team has executed a special calibration program in Cycle 25 (PID 15381) to obtain new observations of a spectrophotometric standard star with the E140M grating, for the purpose of deriving an updated set of sensitivity curves. Analysis of these observations is now underway in Summer 2018.
The STIS team also created a Python module, stisblazefix, to optimize the alignment of the blaze function shifts for individual STIS echelle spectra, with a possible improvement over the pipeline data products. The tool is summarized in STIS ISR 2018-01, and also includes installation instructions.
Users who retrieved post-SM4 E140M observations from the MAST archive prior to June 25, 2018, may wish to retrieve their data again for the most up-to-date calibrated version using the new PHOTTAB reference file. Alternatively, you may download a copy of the new reference file to manually reprocess your observations from the following page: https://hst-crds.stsci.edu/browse/26p1601ko_pht.fits.
New time coefficients for the NUV echelle gratings, E230M and E230H, are being tested. PHOTTAB updates for each grating will be released separately, later this year.
Predicted Impacts of Increased Observatory Jitter on STIS Spectroscopic Modes
Gyro-1 onboard HST failed on the morning of April 21, 2018. Routine procedures subsequently took place to power on Gyro 6 and HST resumed nominal operations with a new combination using Gyros 2-4-6 within days of the failure. Since the combination switch occurred, increased spacecraft jitter (from routine 5-7 mas RMS to ~12 mas) has been observed and is being actively monitored. It is important to note that STIS’ slits are oriented ~45° from V2, thus 12mas of jitter in V2 corresponds to 8.5mas of jitter along the dispersion direction.
For STIS, observations using the narrowest slits in the FUV will be the most affected as an increase in spacecraft jitter could potentially translate into a measurable degradation of the spectral resolution and of the overall spectroscopic data quality (S/N and flux calibration). The STIS Team has identified those spectroscopic modes that would potentially suffer the worst degradation in spectral resolution using both calibration data taken since mid-April and simulations of the impact that increased jitter might have compared to a "zero-jitter situation."
In order to determine upper bounds on the degradation of the STIS spectral resolution, we convolved modeled STIS Line Spread Functions (LSFs) with various amounts of jitter to compare the impact to the zero-jitter case. Note that the lower levels of jitter prior to the gyro switch and the effect of STIS' defocus (see STIS ISR 2017-01) are not included in the simulations below. Thus, the observed marginal degradation is less than the predicted numbers given.
|Detector||Optical Element||Aperture||Wavelength||% Resolution Degradation|
|Jitter 6.0 mas||Jitter 10.0 mas||Jitter 12.5 mas|
Above: Table of simulated resolution loss due to jitter, as compared to the zero-jitter case. The bluest FUV modes and smallest slits are the most affected.
Note that these predictions are extremely conservative. Initial analysis of recent data suggests that the impact of the increased jitter on the spectral resolution is minimal.
The spacecraft's pointing is monitored during observations in order to characterize any impact spacecraft jitter may have on science. These pointing telemetry data are provided in jit.fits files a few days after observations are available in MAST. Users are encouraged to examine their jitter files to assess the impact upon their observations.
Due to the increased jitter, HST has experienced decreased performance in guide star acquisitions and reacquisitions. This may cause the Take Data Flag (TDF) to go down and prevent STIS from recording data, or shortening exposures from their planned duration. As such, users should check their data for success promptly after execution and within the usual 90-day period attached to any HST observations and file a Hubble Observation Problem Report (HOPR) for any issues discovered. Observers are reminded that exposures deemed failed will become publicly available immediately. In the past few weeks, the mission has been implementing changes to the guiding system to help mitigate these losses.
Testing Spatial Scans with the STIS CCD
Spatial scanning with the STIS CCD is a recently enabled, available-but-unsupported mode for obtaining high S/N ratio spectra of relatively bright targets by trailing the target in the spatial direction within one of the long STIS apertures. Possible scientific applications include the reliable detection of weak stellar and interstellar absorption features (particularly in the red and near-IR where ground-based observations can be severely compromised by strong telluric absorption) and the accurate monitoring of stellar fluxes (both broad-band and in narrower spectral intervals) for characterizing transiting exoplanets. As discussed in the 2017 March STAN and in a new section in the STIS Instrument Handbook for cycle 26, spatial scanning offers the potential advantages of improved flat fielding, improved fringe removal (beyond about 7000 Å ), and better correction for cosmic rays, compared to the previously used method of deliberately saturating the CCD detector at a fixed pointing. In this article, we briefly summarize the results achieved in one completed GO program which utilized spatial scans and provide some preliminary results from the ongoing analysis of data obtained for a special calibration program designed to test some aspects of this observing mode.
GO program 14705 (PI M. Cordiner) obtained scanned spectra of the heavily reddened B0 I star BD+63 1964 with the STIS G750M/9336 setting, in an attempt to detect weak diffuse interstellar bands that have been attributed to C60+ (Cordiner et al. 2017, ApJ, 843L, 2). Use of the 52x0.1 aperture for both the trailed stellar spectra and the tungsten lamp exposures enabled very effective removal of the fringing pattern; see Fig. 12.8 in the STIS Instrument Handbook. The S/N ratios in the final reduced spectra range from 600-800 per resolution element, consistent with expectations from the total counts in the combined spectrum; see Fig. 1 in Cordiner et al. (2017). Based on that successful application of spatial scanning, two follow-on programs are applying the technique to obtain similar spectra of additional targets.
The first visit of calibration program 15383 (PI C. Proffitt) compared long trailed images of the white dwarf GRW+70 5824 (taken through the 52x2 aperture) with images of the tungsten lamp (taken through the 52x0.1 aperture), to see how well the two are aligned for the default scan angle of 90 degrees. While a small difference (~0.065 degrees, corresponding to about 1 pixel over the roughly 1000-pixel extent of the scan) was found between the two images, it is not yet known whether that is characteristic or not. The data from visit 1 also indicated that while the forward scans executed consistently, an apparent timing issue caused the start of the corresponding reverse scans (in round-trip mode) to begin at different locations. We therefore recommend using only forward scans, positioning the target at the same initial location for each scan.
The second visit of program 15383 obtained scans of the bright G8 V star 55 Cnc (a known exoplanet host with numerous prior STIS observations) using several different scanning configurations. During the first orbit of the visit, long (48 arcsec) scans of 55 Cnc were taken through both the 52x0.1 and 52x2 apertures, in order to compare the alignment of the trailed spectra through the two apertures and with respect to deep contemporaneous flat fields obtained for fringe correction (through the default 52x0.1 aperture). During each of the second and third orbits, a series of ten shorter (12 arcsec) scans was taken through the 52x2 aperture, in order to assess the repeatability of the fluxes -- both within each orbit and from orbit to orbit. Average counts of order 1-2 x 10^4 per pixel were obtained for most of the individual columns in each scan. (Note that this visit was scheduled to execute when 55 Cnc e was not in transit -- so that no intrinsic variations in flux would be expected.) While the analysis of these (somewhat limited) data is still in progress, some preliminary conclusions may be drawn (from images corrected for bias, dark counts, and both large-scale and pixel-to-pixel flat fields, but not for cosmic rays or fringing):
- The positioning of the trailed spectra appears to be quite reproducible, and the fringes in those spectra are well aligned (to within 1 pixel) with those in the fringe flat spectra -- though there are slight (several tenths of a pixel) drifts, both during each orbit and as a function of y position on the detector.
- The average counts (per 2-D pixel, averaged over about 3500 Å, or ~70% of the wavelength coverage) are typically slightly (~0.04%) higher for orbit 2 than for orbit 3, but the standard deviations of those average counts are of order 0.01% for the 10 scans within each orbit and are less than 0.005% for the last 7-8 scans in each orbit (see Fig. 1). (These may be compared to the nominal expected absolute and relative flux accuracies of 5% and 2%, respectively, for pointed observations taken through the 52x2 aperture; see Sec. 16.1 in the STIS Instrument Handbook.)
- The agreement is somewhat poorer for narrower wavelength bins, but the standard deviations are still of order a few hundredths of a percent for 50-500 Å-wide bins.
While the exact figures will change, once more thorough reductions are performed (including removal of cosmic rays, fringing, and geometric distortion, and with careful assessment of possible effects due to jitter, scan rate variations, variations in dark rate, etc.), these initial indications are promising.
More detailed information regarding observing strategy -- e.g., determining the scan parameters for observing a particular target (exposure time, scan rate, initial target positioning) and obtaining custom flat-field exposures -- may be found in Sec. 12.12 of the STIS Instrument Handbook; specific questions may be addressed to the STScI HelpDesk. The spectra of 55 Cnc obtained for calibration program 15383 are publicly available via the MAST archive. We would encourage those who may be interested in using STIS spatial scans to download and examine those data, in order to gauge the potential utility of this technique for their own research projects.
Reprocessing Status of STIS Data Obtained Before Servicing Mission 4
The STIS team is pleased to announce that all STIS data obtained before Servicing Mission 4 (pre-SM4 data) have now been reprocessed using the latest reference files and pipeline processing software.
There are several situations in which archival STIS data is reprocessed, including the generation of new reference files, the update of existing reference files, or the update of calSTIS software. As of 2007, STIS data obtained before SM4 have been statically archived and not updated. This has resulted in the pre-SM4 data being outdated in terms of reference files, keyword headers, and other improvements. The Data Processing and Archiving Services team at STScI has recently reprocessed all pre-SM4 data as of July 2018. We recommend users currently working with STIS data obtained before 2009 to re-download data via MAST. The pre-SM4 data will be included in all future pipeline and reference file updates.
Please contact the HST helpdesk if you have any issue or question about the newly processed STIS pre-SM4 data.