Specifying the Appropriate Background

There are several potential types of background and noise that can affect the observations; the main contributors being:

  1. Earthshine
  2. Zodiacal light
  3. Geo-coronal emission Lines (UV)
  4. Atmospheric helium (IR)
  5. Thermal background (IR)
  6. Dark current
  7. Read noise (CCDs and NICMOS)

The sum of the External Background components is presented on the Results page. Note that each of the components of External Background can be switched off (set to None or No airglow) so that one of the components can be selected for display on the Results page. All components should be included when simulating observations, as artificially turning off the sky does not represent actual observing conditions.

External Background

Earthshine

Earthshine can vary strongly depending on the limb angle, on the fraction of the earth that is sun-lit, and on the fraction of cloud cover. At low limb angles, it can dominate the zodiacal light, which varies by only a factor of ~3 throughout the sky available to HST (See Giavalisco et al. WFC3 ISR 2002-12).

The table below lists the relative levels of Earthshine selectable in the ETC.

Table 1: Earthshine Background Contributions Relative to “High”

Level: None Average High Extremely High
Earthshine 0% 50% 100% 200%
Limb angle N/A 50 deg 38 deg 24 deg

The phase II special requirement LOW-SKY ensures that the limb angle will be greater than 40 degrees, i.e. that it will avoid the limb angle of the High Earthshine case by at least 2 deg. This requirement greatly reduces scheduling opportunities, so should be requested only when critical to the science and should be well justified in the phase I proposal.

Zodiacal Light

The Zodiacal light values given in Table 2 are in ecliptic co-ordinates \beta (ecliptic latitude) and \lambda-\lambda_{\odot} (helio-ecliptic longitude or elongation). See the following figure.

Zodiacal Light Angle Illustration

When a user request to calculate the contribution of the Zodiacal light with position, first the Equatorial coordinates are converted to Ecliptic coordinates. The relation between these set of coordinates is:

\sin(\beta) = \sin(\delta) \cos(\epsilon) - \cos(\delta) \sin(\epsilon) \sin(\alpha) \ \ (i) \\
\cos(\lambda) = \cos(\alpha) \cos(\delta) / \cos(\beta) \ \ (ii) \\
\sin(\lambda) = [\sin(\delta) \sin(\epsilon) + \cos(\delta) \sin(\alpha) \cos(\epsilon) ] / \cos(\beta) \ \ (iii)

or

\sin(\beta) = \sin(\delta) \cos(\epsilon) - \cos(\delta) \sin(\epsilon) \sin(\alpha) \ \ (iv) \\
\tan(\lambda) = [\tan(\delta) \sin(\epsilon) + \cos(\epsilon) \sin(\alpha) ] / \cos(\alpha) \ \ (v)

where:
\alpha = right ascension
\delta = declination
\epsilon = obliquity of the ecliptic with \epsilon = 23.446^o for equinox 1950.

Since we know \epsilon we can then use (i) and (ii) to calculate \beta and \lambda.

To solve for the ambiguity of the sign you should calculate the equations (iii) and (v) and

  • if \lambda_1 < 0 and \lambda_2 > 0 and \lambda_3 > 0 then \lambda = \lambda_1 + 180
  • if \lambda_1 > 0 and \lambda_2 < 0 and \lambda_3 > 0 then \lambda = \lambda_1 + 180
  • if \lambda_1 < 0 and \lambda_2 < 0 and \lambda_3 > 0 then \lambda = \lambda_1 + 360
where:
\lambda_1 = \tan^{-1}(\tan(\lambda))
\lambda_2 = \sin^{-1}(\sin(\lambda))
\lambda_3 = \cos^{-1}(\cos(\lambda))

The ETC then calculates the helio-ecliptic logitude or elongation (sun angle between sun-earth and earth particle) to determine the brightness along the line of sight:

\lambda' = \lambda - \lambda_{\odot}

where:
\lambda_{\odot} is the equatorial longitude of the Earth

\lambda_{\odot} is derived as follows:

\lambda_{\odot} = 360 \times \Delta / 365.25 \ \ (vi)

where:
\Delta is the number of days from the Vernal Equinox of the selected year (March 20).

Once the Helioeclipic coordinates are calculated, the ETC performs a bilinear interpolation using the values in Table 2 in order to derive the Johnson V magnitude for the Zodiacal contribution. The result is then fed to the pysynphot task renorm to renormalize the Zodiacal light spectrum model (currently zodiacal_model_001.fits) to the derived V magnitude.

Table 2 provides the approximate values of the Zodiacal light background (V_{mag} / arcsec^2) in ecliptic latitude (\beta) and helio-ecliptic longitude (\lambda' = \lambda - \lambda_{\odot}). The SE values correspond to the solar exclusion zone. This table has been updated for cycle 24 using a finer grid to provide better estimates of the zodiacal light as a target approaches the 50 degree HST-to-target limiting angle. It was again modified for ETC 29.1 to correct typos in a few positions (marked with ‘*’).

Table 2: Zodiacal Light Background Contributions

\lambda'
(degrees)
\beta (degrees)
0 5 10 15 20 25 30 35 45 50 60 75 90
0 SE SE SE SE SE SE SE SE SE 22.0708 22.5136 22.9538 23.2298
5 SE SE SE SE SE SE SE SE SE 22.0816 22.5136 22.9538 23.2298
10 SE SE SE SE SE SE SE SE SE 22.1033 22.5210 22.9538 23.2298
15 SE SE SE SE SE SE SE SE SE 22.1454 22.5360 22.9538 23.2298
20 SE SE SE SE SE SE SE SE SE 22.2004 22.5743 22.9649 23.2298
25 SE SE SE SE SE SE SE SE 22.0808 22.2586 22.6141 22.9762 23.2298
30 SE SE SE SE SE SE SE SE 22.1578 22.3237 22.6554 23.0107 23.2298
35 SE SE SE SE SE SE SE 21.9203 22.2350 22.3924 22.7071 23.0224 23.2298
40 SE SE SE SE SE SE 21.8257 22.0287 22.3181 22.4628 22.7522 23.0343 23.2298
45 SE SE 21.0810 21.3356 21.5717 21.7872 21.9545 22.1379 22.3948 22.5232 22.7801 23.0707 23.2298
50 20.8432 21.0663 21.3194 21.5397 21.7408 21.9486 22.0833 22.2472 22.4715 22.5837 22.8080 23.1071 23.2298
60 21.1844 21.3356 21.5842 21.7872 21.9859 22.1525 22.2937 22.4437 22.6304 22.7237 22.9104 23.1212 23.2298
75 21.6258 21.6965 21.8737 22.0611 22.2180 22.3621 22.4989 22.6319 22.7801 22.8542 23.0024 23.1607 23.2298
90 21.9155 21.9768 22.0660 22.2350 22.3948 22.5284 22.6470 22.7699 22.9104 22.9807 23.1212 23.2016 23.2298
105 22.1315 22.1419 22.2124 22.3686 22.5136 22.6387 22.7614 22.8912 22.9990 23.0529 23.1607 23.2298* 23.2298
120 22.2639 22.2757 22.3305 22.4844 22.5980 22.7071 22.8186 22.9646 23.0707 23.1237 23.2298 23.2736 23.2298
135 22.3181 22.3243 22.3948 22.5284 22.6304 22.7339 22.8483 22.9224* 23.0707 23.1237* 23.2298* 23.2885 23.2298
150 22.3181 22.3243 22.4014 22.5210 22.6060 22.6896 22.7801 22.8639 22.9990 23.0665 23.2016 23.3037 23.2298
165 22.2180 22.2407 22.3181 22.4014 22.4989 22.5743 22.6554* 22.7435 22.9104 22.9938 23.1607 23.3037 23.2298
180 22.0418 22.1315 22.2236 22.3243 22.4216 22.5210 22.6304 22.7348 22.8998 22.9823 23.1473 23.3037 23.2298

Derived from Leinert, et al. (1998), Table 17.

This is used to normalize the Zodiacal Light spectra used by the ETC and shown in the figure below.

Zodiacal Light Template

Geo-Coronal Emission Lines

Table 3: Geo-Coronal Emission Line Properties

Line Low Average High
Wavelength Angstrom Flux(erg cm^-2 s^-1) FWHM Angstrom Flux(erg cm^-2 s^-1) FWHM Angstrom Flux(erg cm^-2 s^-1) FWHM Angstrom
Lyman_Alpha 1215.7 6.1e-14 0.04 3.05e-13 0.04 6.1e-13 0.04
O_I 1302 3.8e-16 0.013 2.85e-14 0.013 5.7e-14 0.013
O_I 1356 3.0e-17 0.013 2.5e-15 0.013 5.0e-15 0.013
O_II 2471 1.5e-17 0.023 1.5e-15 0.023 3.0e-15 0.023

The strength of the geo-coronal Lyman alpha varies between about 2 and 20 kilo Rayleighs, depending on the time of observations and the position of the target relative to the Sun, and can be kept low by the special requirement “SHADOW”. For more details, see the corresponding Instrument Handbook.

Atmospheric Helium

Short-wavelength WFC3/IR imaging filters (F105W and F110W) and WFC3/IR grisms (G102 and G141) are subject to excess background emission caused by metastable Helium atoms in the solar-illuminated upper atmosphere. The background is due to He I emission at 10830 Angstrom. Its contribution is negligible in Earth’s shadow but increases sharply when the telescope is outside of shadow (up to 5 e-/s/pix in extreme cases). The He background also increases with decreasing limb angle, but can still be significant as high as 40 degrees above the Earth limb. For more information on this background component, see WFC3/ISR 2014-03 (Brammer et al.)

The contribution of atmospheric helium to the WFC3 background is negligible in Earth’s shadow, corresponding to the “None” default ETC setting. The “Average”, “High” and “Very High” values implemented in the ETC correspond to 0.1, 0.5, and 1.5 e-/sec/pix in F105W and are the 50, 75, and 95 percentile fluxes of the excess flux, per exposure, from the Helium line determined from archival observations. The ETC model of this emission is a gaussian source at 10830 Angstrom with FWHM = 2 Angstrom.

Thermal Background and Noise

The thermal background is negligible below about 8000 Angstrom and increases slowly towards longer wavelengths. For WFC3/IR, the thermal count rate (per unbinned pixel) is calculated by an algorithm that is described in detail in “Thermal Background Limitations for IR Instrumentation Onboard HST”, Sosey, M., Wheeler, T., Sivaramakrishnan, A., 2003, NICMOS ISR 2003-007.

Detector dark current is an intrinsic source of background counts. The dark current rate is dependent upon the detector design and temperature. It is measured in counts per unbinned pixel per second. For CCDs, 1 count = 1 e-; for MAMAs, 1 count results from each charge cloud detected.

CCD detectors and the WFC3 IR detector are subject to noise caused by the process of reading out the charge accumulated by the pixels. The amount of read noise varies by detector and as a function of gain. CCD read noise is measured per binned pixel, per read. WFC3/IR read noise per pixel is measured by a linear fit to a complete sequence of 15 reads. Fewer reads (smaller NSAMP) result in greater read noise, not taken into account in the computation. (See WFC3 ISR 2009-21).

Table 4: Dark Current and Read Noise Values

Instrument Detector Dark Current Read Noise
(counts \cdot s-1 \cdot pixel-1) (counts \cdot pixel-1)
ACS HRC 0.058 4.7 (for gain=2.0)
SBC 8.52e-6 N/A
WFC 0.0161 4.45 (for gain=2.0)
COS FUVB 7.01e-6 (target acquisition) N/A
FUVA 6.23e-6 (target acquisition) "
FUVB 4.21e-6 (spectroscopic) "
FUVA 3.39e-6 (spectroscopic) "
NUV 1.25e-3 N/A
STIS CCD 0.026 (low; top of detector) 6.4 (for gain=1); 8.8 (for gain=4)
0.032 (medium; middle of detector) "
0.038 (high; bottom of detector) "
FUV 3.0e-5 (no glow component) N/A
7.5e-5 (low glow component) [1] "
1.5e-4 (medium glow component) "
4.0e-4 (high glow component) "
NUV 0.0013 N/A
WFC3 IR 0.02 14.6 (for gain=2.5)
UVIS1 0.00213 ADU/s/pix = 0.00319 counts/s/pix 3.0 (for gain=1.5)
UVIS2 0.00222 ADU/s/pix = 0.00333 counts/s/pix 3.0 (for gain=1.5)
[1]Much of the time-variable glow component of the dark rate may be avoided by observing point sources at the D1 position.