STScI Logo

Hubble Space Telescope
June 2018 STAN

Older STANs

STScI Analysis Newsletter (STAN)

Cosmic Origins Spectrograph (COS)

June 2018

In this Newsletter we provide information on new central wavelengths offered in the FUV channel, action required for the 800 cenwave for ETC calculations for Phase I proposals, updates to the background levels and throughputs in the ETC, and community generated airglow templates.


Two new FUV central wavelength modes available for Cycle 26: G140L/800 and G160M/1533

Two new central wavelengths (cenwaves) are being offered in Cycle 26 for COS/FUV observations, named G140L/800 and G160M/1533.

Cenwave 800 allows for contiguous coverage of the entire spectral region from 800 to 1950 Angstroms on a single detector segment (FUVA) with a low spectral height below 1150 Angstroms, allowing higher S/N for background-limited observations. This setting will be particularly useful for background-limited observations of faint targets at low wavelengths.

Cenwave 1533 extends coverage at the short-wavelength end of G160M by 44 Angstroms to overlap with the longest wavelengths covered by cenwave G130M/1222, and is otherwise expected to have similar spectral resolution and sensitivity to the existing G160M/1577 cenwave. This allows a broad range of FUV wavelengths to be covered at good spectral resolution by just two cenwave settings (1222+1533) without placing Lyman-alpha on the detector, allowing the full FUV to be covered with an efficient use of orbits.

The coverage of these modes compared with the existing COS modes is shown at: For full details, see the COS Instrument Handbook. The calibration of these new modes is ongoing, and any updates will be posted to the COS website.

ETC 26.1 underestimates S/N with G140L/800

The current version of the ETC (26.1) underestimates the S/N obtained with the new G140L/800 mode, and conversely, overestimates the time required to achieve a certain S/N.

Using the quantities calculated by the ETC it is fairly easy to determine the correct values, as given below. We note that users performing ETC calculations for reference wavelengths above 1150 might want to use instead the G140L/1105 mode, as the ETC calculations for that mode are correct.

The current implementation of the G140L/800 cenwave in ETC 26.1 uses an extraction height of 30 pixels. This is an accurate representation of the extraction provided by the CalCOS pipeline only for wavelengths shorter than 1150 , as at higher wavelengths the extraction will continue to be performed at an extraction height closer to 57 pixels. ETC 26.1 also incorrectly assumes that an extraction box of 30 pixels corresponds to only 59% of the enclosed energy. At wavelengths < 1150 A, 100% of the enclosed energy will be contained in 30 pixels, while at longer wavelengths the 100% enclosed energy is reached only at ~57 pixels.

As a result the S/N returned by the ETC should be adjusted by the following factor:
S/N_actual = S/N_etc * 1.7 * sqrt[ (1.0 + D/S) / (1.7 + D/S) ]

where D is the dark count rate (counts/sec) over the 1 pix x 30 pix extraction region, returned by the ETC and S is the source count rate (counts/sec) also returned by the ETC (column 2 of the output, "Count rate (counts/s)").

For S >> D this means that S/N_actual = 1.3*S/N_etc
For S ~ D this means that S/N_actual = 1.5*S/N_etc

The calculation for the time required to reach a given S/N is correspondingly an overestimate. The correction factor for exposure time is inversely proportional to the square of the S/N correction factor, and is given by:
t_actual = t_etc * (1.0/1.7)2 * (1.7 + D/S) / (1.0 + D/S)

For S >> D this means that t_actual = t_etc*0.59
For S ~ D this means that t_actual = t_etc*0.47

New COS dark rates and throughputs adopted in ETC v26.1

The dark rate for both COS FUV detectors (FUVA and FUVB) is monitored closely. The dark rate in both FUVA and FUVB has been stable with a low scatter since mid-2015, following the solar maximum, as shown in:

As a result of the lower levels, the FUV dark rates adopted by the COS Spectroscopy ETC in version 26.1 have been lowered to 1.67x10-6 (FUVA) and 1.59x10-6 (FUVB) counts/s/pixel. For faint targets, these lower levels reduce the exposure time needed to reach a given S/N level. For spectroscopic target acquisition ETC calculations, the COS dark rates have been changed to 2.89x10-6 (FUVA) and 2.99x10-6 (FUVB) cts/sec/pixel, and for the NUV detector the dark rate adopted in the ETC is 8.77x10-4 cts/sec/pixel. Furthermore, the ETC v26.1 has been updated with the latest throughputs, which are slowly declining with time, as shown here:

PIs are strongly encouraged to ensure that any exposure time calculations included in a Cycle 26 COS GO proposal have been performed with the latest version of the ETC.

Community Generated Airglow Templates available for use

Dr. Vincent Bourrier of Geneva Observatory and his team, used a recent set of COS observations of 55 Cnc, to build observational templates of four airglow (geocoronal emission) lines, and used them to successfully decontaminate the observations. Bourrier et al. have graciously offered to share these templates with the astronomy community through STScI. The available templates are for the OI 1302, OI 1305, OI 1306, and HI 1216 lines and include information about the wavelength, flux, and errors. STScI and the COS Team have not tested these templates and while we encourage the astronomy community to make use of these products, we warn users to do so at their own risk. To read more about these products and to download the templates, please visit the COS Airglow Page.

Unsubscribe COS Help Help Desk

The Space Telescope Science Institute is operated by the Association of Universities for
Research in Astronomy, Inc., under NASA contract NAS5-26555.