About This Article
In this STAN, we provide information on a new COS FUV lifetime position for Cycle 34, Phase I submission best practices, new ETC normalizations for Gaia fluxes, and an update to the COS Gain Sag Table Reference File.
CalloutContents
- New Lifetime Position (LP 11) for G140L in Cycle 34
- New Gaia Exposure Time Calculator (ETC) Normalizations
- Policy on the Maximum Usage of Lifetime per Observing Program Still In Effect
- Continued Enforcement of the COS2025 Rules for the G130M Grating
- Special Requirements Must Be Justified in the Phase 1 Proposal
- New COS Gain Sag Table Reference File
End callout
New Lifetime Position (LP 11) for G140L in Cycle 34
The COS FUV detector experiences gain sag, in which the response of a given detector location decreases with accumulated exposure to light. To mitigate this effect and preserve sensitivity, COS periodically introduces new lifetime positions (LPs), shifting spectra in the cross-dispersion direction to less-used regions of the detector. Beginning in Cycle 34, COS will introduce Lifetime Position 11 (LP11) for G140L observations. All G140L cenwaves will move from the current LP3 configuration to the new LP11.
As with previous lifetime-position changes, the impact on science observations is minimal. The spectral resolution and sensitivity of G140L vary only slightly with lifetime position, and users do not need to handle LP11 observations any differently than at previous LP3 while preparing their Phase I proposals. The COS team will communicate any Phase II changes in an upcoming STAN. Performance and calibration details for LP11 will be documented in forthcoming COS Instrument Science Reports.
New Gaia Exposure Time Calculator (ETC) Normalizations
Section 4 of the COS ETC allows users to normalize an input spectral distribution by various fluxes and bandpasses. As of ETC version 34.1.1, users may now select Gaia bandpasses for normalization. It is important to note that Gaia photometry is in the Vega system, and users will need to select "Vega Magnitude" if they are using magnitudes reported in Gaia DR3. Users are reminded to regularly check the STScI ETC News and Known Issues or the ETC Release Notes for the latest information on the ETC.
Policy on the Maximum Usage of Lifetime per Observing Program Still In Effect
The COS FUV detector usage over the last few years exceeded historical levels, increasing the risk of degraded performance before 2030. To ensure that COS and its capabilities remain available into the 2030s, STScI has implemented a policy to preserve the COS FUV detector, capping the usage to 2% of the lifetime at a Lifetime Position (LP) for an observing program.
STScI now requires users to provide an estimate of their lifetime usage for given modes in their Phase I justification. Additionally, lifetime usage beyond 1% for a program should be justified in the Phase I to address whether other modes with lower lifetime impact could be used and why observations could not be obtained with STIS. Furthermore, since Cycle 32, users are not allowed to obtain a signal-to-noise ratio (SNR) on a single target beyond the limit set by fixed pattern noise for a given mode.
All COS users should determine from the table below the approximate maximum number of targets observable under the lifetime usage cap of 2% for a given SNR and instrumental mode. If the number is within a factor of two of the limit (i.e., beyond 1% lifetime usage), users should perform a more careful calculation and consider optimizing the operational setup to bring the lifetime usage down prior to submission (contact scientists of such approved programs will work with the PIs to minimize the lifetime usage). Details of this policy and how to estimate lifetime usage are given in Table 1.
The COS bright-object safety screening limits (Section 10.2 of the IHB) for the FUV were chosen to protect the detector against overlight events that can occur over short time periods. However, very bright continuum sources, emission lines, or moderately bright objects observed for long periods can still deplete a significant fraction of an LP's lifetime without violating the screening limits. Approximately 27,000 counts can fall on a pixel before the reduced gain at that location results in significant flux loss. FUV observational programs must keep the estimated total counts in the brightest pixel integrated over different targets for a given lifetime position under 2% of a pixel's lifetime, or 540 counts in a given pixel. Users in these situations should carefully consider whether sacrificing wavelength coverage by turning off segments might help keep their science under the lifetime caps. Note also that the 2% restriction refers to a single lifetime position. As of Cycle 34, LP7 is the default for G130M 1055, 1096, 1222, and 1291 cenwaves, with G130M cenwaves greater than 1291 remaining at LP5. G160M cenwaves reside at LP10, and G140L cenwaves are planned to be at LP11.
How to Determine the Lifetime Usage of Your Observation
Proposers will need to evaluate whether their observations exceed this limit if their total targets are within a factor of two of the numbers in Table 1. Users can do one calculation for their highest SNR target and apply it to all of their targets; if they still exceed the cap they will have to do more detailed calculations. The procedure is as follows:
- Perform an ETC Calculation: Run a COS ETC calculation for the science target and determine the brightest pixel count rate for each segment. If observations include airglow lines such as geocoronal Lyman-alpha or OI, use the "No airglow" option from Question 5c of the ETC to estimate the brightest pixel’s count rate. If unsure, the airglow feature can be turned off regardless of whether it is covered.
- Multiply by the Total Exposure Time: This assessment depends on whether the brightest pixel is dominated by a continuum or an emission line:
- Continuum-Dominated Observations: The total exposure time per grating should be used, irrespective of the number of FP-POS settings. Therefore, multiply by the total exposure time per grating per visit
- Emission Line-Dominated Observations: The total exposure time per FP-POS should be considered since the emission line will land on different regions of the detector at each FP-POS setting. Therefore, multiply by the exposure time per FP-POS per visit.
- Sum Across All Visits
- Check Against the 540-Count Limit: If the total exceeds 540 counts at a given LP, adjustments will be necessary, such as using a different mode, turning off a segment, reducing exposure time, or moving some observations to STIS.
Example Calculations
Example 1: Continuum Source
A bright A3V star is observed with G160M for a total exposure time of 11,590 s over five visits. ETC results show peak counts/s/pixel of 0.128 for Segment B.
Calculation: Since this exceeds the 540-count limit by a factor of 2.75, a less sensitive mode would be required, such as STIS/E140M. The ETC calculation for this example is: COS.sp.1895424
Example 2: Emission Line Source
Epsilon Eridani is observed with G130M/1222 for 12,000 s, using 4 FP-POS. ETC results indicate peak counts/s of 0.071 for Segment B.
Calculation: Since the total per FP-POS is below 540 counts, the observation remains within limits. Additionally, the total per FP-POS is below 270 counts, so this observation will not trigger a review or optimization. The ETC calculation for this example is: COS.sp.1923915
Example 3: Optimizing a program for a large number of targets
A program plans to observe 40 WDs at SNR=27 with G130M/1291. Looking at the table above, the estimated allowed number of targets in this mode at that SNR is only 22. The program could either move to G130M/1222 or consider moving some of the observations to STIS. The proposing team should still perform detailed ETC calculations to confirm that they will stay under the 2% cap and justify any lifetime usage beyond 1%. An example ETC calculation for G130M/1291 is: COS.sp.1954076 and for G130M/1222 is: COS.sp.1954080
For questions or assistance with estimating detector impact, please contact the HST Help Desk.
Continued Enforcement of the COS2025 Rules for the G130M Grating
The COS2025 rules, designed to extend the lifetime of the COS FUV detector, are of particular interest to scientists who want the full two-segment wavelength coverage offered by the G130M grating or who want to observe Ly α in zero-redshift targets. The COS2025 rules will continue to apply in Cycle 34. The rules limit the locations where damaging Ly α airglow can fall on detector segment FUVB: 1) G130M cenwaves greater than 1291 cannot be used when segment B is on, and 2) G130M/1291 is limited to two FP-POS when both segments are on. Please see the COS2025 page for further details. A consequence of these rules is that, for zero-redshift astronomical targets, Ly α falls on gain-sagged regions of the detector at the default LP7. Users who wish to observe Ly α in such targets should instead use LP3. This option must be requested and justified in the Phase I proposal.
Special Requirements Must Be Justified in the Phase I Proposal
All special requirements must be requested and justified in the "Special Requirements" section of Phase I proposals. A list of such requirements appears in the Call for Proposals. For COS, the most common requirements are explicit scheduling requirements (e.g., for planetary transits), waivers of the FP-POS requirements, requests to use non-default lifetime positions, and ORIENT constraints.
ORIENT constraints are commonly used for extended sources (i.e., sources with FWHM > 0.6"). Observing these at different position angles can result in changes in the cross-dispersion and dispersion profiles on the COS detector. Such changes can lead to differences in spectral resolution and flux calibration that may prove problematic when combining spectra across visits or performing science on separate spectra taken at different position angles. This can be mitigated by implementing an ORIENT constraint across visits within APT, although users should also be aware of ORIENT constraining schedulability in Reduced Gyro Mode. As with other special requirements, these must be requested and justified in the Phase I proposal.
New COS Gain Sag Table Reference File
The COS team will soon deliver a new gain sag reference table (GSAGTAB) that includes updated flagging of pixels primarily on the bluest wavelength side of segment FUVB at lifetime position (LP) 5 with G130M/1291 (< 1150 Angstroms), though a small number of pixels at other LPs will also be updated. Although the default location for G130M/1291 moved to LP7 for Cycle 33, some Cycle 32 observations continue to be taken at LP5. Upon delivery of the new GSAGTAB file, users with data taken at LP5 and segment FUVB after ~ November 2025 are encouraged to download their data from MAST to receive the most up-to-date data products. Additional details regarding how pixels are flagged due to gain sag can be found in the COS Instrument Handbook in Section 5.12.2.