INSTRUMENT SCIENCE REPORTS (ISRs)

We report on the COS instrument’s spatial resolution for observations taken using the FUV G160M grating at Lifetime Position 6 (LP6). We compare measurements made on spectra taken with all G160M central wavelength settings (1533, 1577, 1589, 1600, 1611, and 1623) against those predicted by an optical model of COS. Using these observations, we validate the model and present predictions of the COS FUV spatial resolution at LP6. We perform this model validation by comparing the cross-dispersion spread functions (CDSFs) measured in COS spectra of the white dwarf WD0308-565 to the CDSFs predicted by the optical model. After confirming that the observations agree with the model, we find that the spatial resolution generally improves as a function of cenwave, from a high full width at half maximum of ∼14 pixels = 1.26 arcseconds for cenwave1533near1500 A to a low of ∼3pixels=0.27 arcseconds for cenwave 1623 near1700A.

We present changes in the background locations when extracting spectra from NUV data taken by the Cosmic Origins Spectrograph (COS) on board the Hubble Space Telescope. This change was motivated by discrepancies between the NUV and FUV fluxes of extragalactic extended sources. We found that the background extraction boxes for the NUV XTRACTAB are too close to the Primary Science Aperture (PSA) region along the cross-dispersion direction, such that the PSA profile often overlaps with the background extraction boxes, even in the case of non-extended sources. As a result, the NUV fluxes were reported to be lower by ∼20% compared to the FUV fluxes for some extended sources. To avoid over-subtraction, we created a new NUV XTRACTAB where the background extraction locations were moved closer to the edges of the detector, away from the PSA extraction boxes while keeping the extraction box heights unchanged. In this Instrument Science Report, we show the effect of changing the background location in the NUV fluxes of both extended and non-extended targets. We observe NUV fluxes to change by 0—22 % between the old and the new NUV XTRACTAB, with the largest change observed in the case of the extended sources.

We report on the COS instrument’s FUV channel spectral resolution for observations using the G160M grating at Lifetime Position 6 (LP6). We compare measurements made on COS/FUV spectra taken with two central wavelength settings (“cenwaves” G160M/1533 and G160M/1623) to those predicted by an optical model of COS. We perform our validation by comparing ISM absorption lines found in the COS LP6 observations against the same lines found in high resolution STIS spectra after convolving them with the COS modeled line spread function (LSF). The models show that the COS G160M resolution at LP6 is generally within ±20% of its value at LP4. The resolution at most wavelengths and cenwaves falls between 10,000 and 16,000. The highest resolution attainable at LP6 is R ∼18,000, which is reached on the FUVA segment of G160M/1533.

This ISR documents the Cosmic Origins Spectrograph (COS) Target Acquisition (TA) annual monitoring programs for HST Cycles 20–24, with additional information about COS TA performance determined from data of the C17–C23 FGS-to-SI alignment programs. During this period, NUV exposures were obtained at the nominal (LP1) position, and FUV exposures were executed at Lifetime Positions LP2 and LP3. These programs were designed to monitor numerous aspects of both imaging and spectroscopic TAs, including checking the TA subarrays, the NUV SIAF entries, the telescope slew distances, and evaluating the accuracy of numerous COS flight software (FSW) patchable constants required for TA. This project verified that all three COS TA modes (FUV spectroscopic, NUV spectroscopic, and NUV imaging) were generally behaving nominally in Cycle 20–24, and determined that no SIAF or FSW parameter updates were required during this time, with the exception of changes to MIRRORB ACQ/IMAGE in 2014. These changes included a changing of the lamp current from LOW to MEDIUM, an adjustment of the LTACAL exposure time, and a modification of both the MIRRORB WCA and PSA/BOA ACQ/IMAGE TA subarrays.

We describe the flux calibrations of COS FUV cenwaves G130M/1055 and G130M/1096, known as the blue modes. These modes are the shortest wavelength modes available with HST at medium resolutions, covering 900A to 1,240A at spectral resolutions up to 13,000. A new flat field image used by both cenwaves was derived along with new sensitivity curves for each cenwave based on observations of the flux standard WD-0308-565. Comparison with the CALSPEC model spectrum of WD-0308-565 shows that the residuals of the flux calibration improved from as much as 30% to less than 1%, well within the 5% requirement. We further validate the new calibration by comparing the blue modes spectrum of WD-0308-565 to the overlapping portions of spectra taken at cenwaves G140L/800, G140L/1280, and G130M/1222. The new calibration was delivered to CRDS on 2021 February 22.

A new COS/FUV lamp template reference file (LAMPTAB) for Lifetime Position 3 (LP3) was delivered for use in the COS calibration pipeline in April 2018. The LAMPTAB file contains reference spectra used during the wavelength calibration of COS/FUV spectroscopic data. The LP3 lamp spectra data were obtained in Program 14856 in October 2016 as a part of the improved wavelength calibration effort. In this ISR, we document the data analysis techniques used for the creation of the LP3 LAMPTAB file, as well as our testing procedures.

Between March and June of 2021, exploratory focus sweeps in preparation for Lifetime Position 6 (LP6) were performed across three previously unvisited positions at +7, +9, and +11 arcsec on the COS/FUV detector with the G130M and G160M gratings through HST Program 16491. The focus sweeps were performed on the subdwarf B star Feige 48 at a range of focus settings. An auto-correlation technique was used to find the minimal line widths of absorption lines and therefore find the optimal focus for each setting. The final, absolute focus values for G130M/1222 are −967, −908, and −897 steps, and for G160M/1600 are +78, +219, and +267 steps at the +7, +9, and +11 arcsec positions, respectively.

We report on the monitoring of the zero points of the COS NUV dispersion solutions during Cycle 28 in program 16330. Select cenwaves were monitored for all NUV gratings except for G285M, which is no longer monitored due to its Available-but-Unsupported status. Comparisons to COS monitoring data obtained in previous cycles indicate internal stability within the allowed ranges of zero points. Comparisons to STIS data indicate small but persistent COS offsets of 1 – 3 pixels toward shorter wavelengths. The cause of this is not yet well understood. The March 2022 delivery of a new NUV DISPTAB with a revised zero point for stripe B of the G230L/2950 cenwave reduced this discrepancy for that setting.

We report on the monitoring of the zero points of the COS FUV dispersion solutions during Cycle 28 in program 16325. Select cenwaves were monitored for all FUV gratings. Comparisons to COS monitoring data obtained in previous cycles indicate internal stability within the allowed ranges. Comparisons to FUSE and STIS data indicate satisfactory absolute calibration. All measured offsets are within the established thresholds of 3 pixels for the G130M and G160M cenwaves and 9 pixels for the G140L cenwaves, except for Segment B of the G160M/1623 cenwave, which is slightly outside the specification at −3.8 pixels.

Observations of HST spectrophotometric standard stars show that the COS NUV detector has a time-dependent sensitivity (TDS) that must be monitored and accounted for in flux calibration. Regular observations monitor the changes in sensitivity for three NUV gratings: G230L, G185M, and G225M. Because the sensitivity of the fourth grating, G285M, has become very low, it was removed from the routine monitoring program, and it is now available but unsupported for General Observer programs. Results from the Cycle 28 NUV TDS program show that the G230L and G185M gratings, which are coated in MgF2, exhibit trends consistent with little or no change. On the other hand, the G225M grating, which is bare aluminum, shows a sensitivity decline of −2.85% ± 0.18% per year.

The Cycle 28 COS/FUV spectroscopic sensitivity monitor ran from December 2020 to October 2021. Observations of the standard modes and the new modes (G160M/1533 and G140L/800, introduced in Cycle 26) were obtained at Lifetime Position 4 (LP4), the nominal position for COS starting October 2, 2017, and the blue modes (G130M/1055 and G130M/1096) were obtained at Lifetime Position 2 (LP2). Connection visits to LP5 and LP3 were obtained in preparation for the start of G130M/1291 (and longwards) at LP5, and G140L/800 at LP3. The Time-Dependent Sensitivity (TDS) slopes of all modes ranged from 0% to 3% per year. In this ISR we describe the program and its execution and provide a summary of the analysis and results. Based on the change in the net count rates over a one-year timescale, we find that the FUV TDS does not depend on LP between LP4 and LP5.

Here we summarize the Cycle 28 NUV Dark Monitoring Program for the Cosmic Origins Spectrograph (COS) on the Hubble Space Telescope (HST) covering dates November 2020 to October 2021. We present an overview of the calibration plan and summary for this calibration program, and also provide an update to the calculation of the NUV dark rate reported to the Exposure Time Calculator (ETC). The adopted dark rate for the Cycle 28 ETC was 1.11e-03 counts/pixel/sec.

Here we summarize the Cycle 28 FUV Dark Monitoring Program for the Cosmic Origins Spectrograph (COS) on the Hubble Space Telescope (HST) covering dates November 2020 to October 2021. We provide an overview of the calibration plan and summary for this calibration program. The Cycle 28 spectroscopic and target acquisition dark rates decreased by 14-17% for FUVA and 4% for FUVB compared to Cycle 27.

The Cosmic Origins Spectrograph (COS) aboard HST possesses two internal platinum neon (Pt-Ne) hollow cathode lamps which are used independently for 1) TAGFLASH’d wavelength calibration exposures on COS science exposures and 2) target acquisitions (TAs). Over time, the measured intensity of the spectral output from the wavelength calibration lamp has decreased. We analyze the peak counts across several FUV gratings and monitor their on-orbit lamp performance. This is done to determine minimum requirements for accurate wavelength calibrations and quantify the decrease in lamp counts over time. We calculate current lamp exposure times will generally remain sufficient until the late 2020s, with the exception of the G130M/1291 cenwave requiring reevaluation by the end of 2024, and that increasing exposure time does not pose any risk to the lifetime of the lamp.

Starting in 2021 October, the Cosmic Origins Spectrograph (COS) Far-Ultraviolet (FUV) modes G130M/1291,1300,1309,1318,1327 moved to the new Lifetime Position 5 (LP5), and all the G140L cenwaves moved to LP3. As part of the calibration of these “LP5-Era” positions, observations of the White Dwarf star WD0308-565 were obtained with all the G130M cenwaves moving to LP5, and with G140L/800, which had never been used previously at LP3. The data were used to derive low-order flatfields (L-Flats) for G130M (detector segments B and A) at LP5, and sensitivities for all the observed cenwaves. The existing flatfield for G140L at LP3 was retained, and only the sensitivity for G140L/800 at LP3 was derived. The L-flats and sensitivities were used to obtain the flux calibration for the LP5-Era modes. The relative and absolute flux calibration accuracies are within the COS specifications of 2% and 5%, respectively.