In this section we describe some instrument-specific behavior which
must be taken into account when estimating required exposure times. The WWW NICMOS Exposure Time Calculator (ETC) provides the most convenient means of estimating count rates and signal-to-noise ratios (SNRs) for imaging observations. The ETC handles either point sources or extended objects and can be accessed from:
Two NICMOS ETCs are available to observers, an imaging ETC and a
Spectroscopic ETC for grism users. Previously, only the imaging ETC was available. The Spectroscopic ETC was delivered on November 8, 2004 along with an updated imaging ETC.
The imaging NICMOS WWW ETC is documented in an Instrument
Science Report (Arribas, NICMOS ISR-2004-002
). Recent updates are also explained in its help file (accessible from the user interface). The ETC generates either the exposure time required for a specified SNR or the expected SNR for a user-defined exposure time. It will also indicate the read noise and the source and background count rate, as well as the saturation-limited exposure time for the object in question.
|The WWW NICMOS Exposure Time Calculator (ETC) should be
regarded as the tool of choice for estimating integration times for NICMOS observations. The ETC provides the most accurate estimates with the most current information on instrument performance; its reference tables are constantly updated with the most recent values of the instrument’s characteristics as our knowledge of the NICMOS performance under NCS operations improves. Please check the NICMOS Web site for recent ETC updates.
NICMOS performance with the NCS has changed (relative to Cycles 7
and 7N) because the instrument now operates at a different temperature. This has improved the Detector Quantum Efficiency (DQE), but has also increased the dark current. The ETC has been updated to use parameters for Cycle 11 and beyond, with a temperature of 77.15K.
The instrumental characteristics used by the ETC are average
values across the field of view of the camera. The actual sensitivity will vary across the field because of DQE variations (see Chapter 7
for details). A second source of spatial variation not included in the ETC occurs near the corners of the chip because of amplifier glow (see Section 7.3.2
For photon-limited observations (limited either by the target count rate
or the background), SNR increases as the square root of the total observation time, regardless of how the observation is subdivided into individual exposures. However, some NICMOS observations may be significantly affected by read noise, and the net SNR from the sum of several exposures will depend on the relative contributions of read noise and photon noise. Read noise variations that depend on different read sequences are not accounted for in the ETC.
Phenomena such as cosmic ray persistence (Chapter 4
) can degrade sensitivity for faint object imaging by increasing the level of background noise. The impact of cosmic ray persistence is not easily quantified because it is non-Gaussian, correlated noise. It is not possible to predict the extent of this effect at the time the observations are planned, so it is not included in the ETC calculations. Additional information on the ETC and its structure can be found in the Instrument Science Reports (ISRs) posted on the STScI NICMOS Instrument Web page.
The ETC does not
take the count rate non-linearity (Section 4.2.4
) into account. This is especially relevant for faint and/or low surface brightness observations on a low background at wavelengths shorter than 1.6 microns. The effect depends on the count rate in each individual pixel and is therefore hard to model in the ETC. To correct for this non-linearity, for faint objects (mAB > ~20
) one should use a 0.25 mag fainter object in the ETC for F110W in NIC1 and NIC2, and 0.16 mag fainter for NIC3, when one has an average background. The effect is always less at longer wavelengths and brighter backgrounds.
The detector properties which will affect the sensitivity are simply those
familiar to ground-based optical and IR observers, namely dark current and read noise, and the detector quantum efficiency (DQE). The dark current and DQEs measured in Cycle 11 are included in this ETC (see Chapter 7
for more details). The variation of the DQE as a function of wavelength and temperature is also taken into account. The ETC currently does not take the count rate non-linearity described in Section 4.2.4 into account. Faint objects may be affected by as much as 0.25 mag in NIC1 and NIC2, and 0.16 mag in NIC3 at the shorter wavelengths.
NICMOS contains a fairly small number of elements which affect the
sensitivity. These elements are the filter transmission, the pixel field of view (determined by the NICMOS optics external to the dewar, in combination with the HST mirrors), the reflectivities and emissivities of the various mirrors and the transmission of the dewar window.
Filter transmissions as a function of wavelength were measured in the
laboratory and convolved with OTA, NICMOS fore-optics and detector response. The resulting curves are presented in .
NICMOS contains a total of seven mirrors external to the dewar, each of
which reduces the signal received at the detector. The mirrors are silver coated (except for the field divider assembly which is gold coated) for a reflectivity of 98.5%. The dewar window has a transmission of roughly 93%. Therefore, the combination of optical elements is expected to transmit ~84% of the incoming signal from the OTA.
The sensitivity will obviously be affected by the pixel field of view. The
smaller the angular size of a pixel, the smaller the fraction of a given source that will illuminate the pixel, but compensating will be a lower sky background. Finally, the optical efficiency will be degraded further by the reflectivities of the aluminum with MgF2
overcoated HST primary and secondary mirrors.
At long wavelengths (> 1.7 microns) the dominant effect limiting the
NICMOS sensitivity is the thermal background emission from the telescope. The magnitude of this background mainly depends on the temperatures of the primary and secondary mirrors and their emissivities. At shorter NICMOS wavelengths, sensitivities are affected by the zodiacal background. Both sources of background are described in Chapter 4