NICMOS Instrument Handbook for Cycle 11

Photometry

---

Ground-based near-infrared observations are limited to a set of transparent atmospheric windows, while NICMOS suffers no such restrictions. For this reason, there are no suitable faint flux standards with continuous, empirical spectrophotometry throughout the 0.8 µm < < 2.5 µm range. The absolute flux calibration of NICMOS, therefore, has been calculated using observations of stars for which reliable spectral models, normalized by ground-based photometry, are available. Two types of flux standards have been observed: pure hydrogen white dwarfs, and solar analog stars. Grism sensitivity is determined directly from flat-field corrected spectra of these stars using their known spectral energy distributions. Filter sensitivities are calculated from imaging measurements according to the synthetic photometry procedure detailed in Koornneef and Coole (1981, ApJ, 247, 860). Since the pipeline calibration cannot utilize color information, the headers of reduced data contain the calibration constant that specifies the equivalent count rate for a spectral energy distribution that is constant with wavelength. For convenience, this calibration constant appears twice, once in Jansky units and once in erg/s/cm²/Angstrom units. Color transformations could be defined for post-pipeline analysis.

Solar Analog Absolute Standards

For calibration using solar analogs, a reference spectrum of the Sun is normalized to the flux levels of the NICMOS standards using ground-based photometry of the standard stars in the J, H and K bands. This continuous spectral model is then integrated through the total system throughput function for a given bandpass (including filter, detector, instrument and telescope optics), and the integral flux is compared to the measured count rate from the star in observations through that filter to derive the flux calibration constants. The absolute flux accuracy achieved by this method relies on two assumptions:

1. that the absolutely calibrated reference spectrum of the sun is known with an uncertainty of a few percent (Colina, Bohlin and Castelli, 1996), and
2. that the near-infrared spectra of the solar analogs are nearly identical to that of the sun.

In the past, this method has been used to determine the absolute calibration of near-infrared photometry at ground-based observatories. In these cases, the absolute calibration accuracy was estimated to be at least 5%, and for some bands 2% to 3% (Campins, Rieke and Lebofsky, 1985).

Ground-based photometry by Persson et al. (1998, AJ, 116, 2475) of several solar analog stars used in the NICMOS calibration program has shown that the stars P330E and P177D (see Colina & Bohlin 1997, AJ, 113, 1138; Colina, Bohlin & Castelli 1996, AJ, 112, 307) are most closely matched to the colors of the Sun, and are thus most suitable for NICMOS photometric calibration. P330E is the primary NICMOS solar analog standard for photometric calibration.

White Dwarf Absolute Standards

Pure hydrogen white dwarfs are useful calibration standards because their spectral energy distributions can be accurately modeled from the UV through the near-IR (Bohlin, Colina & Finley 1995, AJ, 110, 1316; Bohlin 1996, AJ, 111, 1743). The star G191B2B has therefore served as a primary calibration standard for several HST instruments, and was selected for NICMOS observation along with another star, GD153. At present, only G191B2B has ground-based JHK photometry of suitable quality (Persson, private communication) to verify absolute photometry in the near-infrared. It was found to be necessary to renormalize the G191B2B spectral model by a few percent relative to its optical calibration in order to get good agreement with the ground-based JHK photometry. It is believed that this reflects uncertainties in the white dwarf spectral model at the Paschen limit. Once this renormalization is made, the white dwarf and solar analog flux calibrations for NICMOS agree with one another at the 1% to 3% level. The largest remaining discrepancies are in the K-band, and are presently being investigated.

Photometric throughput and stability

Overall, NICMOS throughput (i.e. photoelectrons per second detected from a source with given flux) is generally within 20% of pre-launch expectations in all observing modes. The photometric stability of NIC1 and NIC2 was monitored once a month since August 1997, and more frequently near the end of the NICMOS Cryogen lifetime. Observations of the solar analog P330E were taken through a subset of filters (5 for NIC1, 6 for NIC2) covering the entire wavelength range of the NICMOS cameras, and dithered through three or four pointings. NIC3 has also been monitored in a similar fashion, although only two filters were used for part of the instrument's lifetime. For most filters and cameras the zeropoints have been stable to within 3% throughout the lifetime of the instrument, although there may be some evidence for a slow secular drift which depends on temperature.

Intrapixel sensitivity variations

The response of a pixel in the NICMOS detectors to light from an unresolved source varies with the positioning of the source within the pixel due to low sensitivity at the pixel's edges and dead zones between pixels. This effect has no impact on observations of resolved sources, and little effect on well-sampled point sources (e.g. observations with NIC1 and NIC2 through most filters). However in NIC3, point sources are badly under-sampled, especially at short wavelengths where the telescope diffraction limit is much smaller than the NIC3 pixel size. Object counts may vary by as much as 30% depending on the positioning of a star within a pixel. Well dithered exposures will average out this effect, but NIC3 observations of stars with few dither positions can have significant uncertainties which may limit the achievable quality of point source photometry. This problem, as well as various possible post-processing solutions, is discussed in Storrs et al. (1999, NICMOS ISR-99-005) and in Lauer (1999, PASP, 111, 1434).

Special Situations

Sources with Extreme Colors

We have carried out tests to establish the likely impact on photometric observations of sources of extreme colors induced by the wavelength-dependent flat field. For each filter, we used two sources with different colors assuming the spectral energy distributions were black-body functions. The first case had a color temperature of 10,000K, and thus is typical of stellar photospheres and the resultant color is representative of the bluer of the sources that will be seen with NICMOS. (It is worth noting that for reflection nebulae illuminated by hot stars, a significantly bluer spectrum is often seen.) The second source had a color temperature of 700K which in ground-based terms corresponds to [J - K] = 5, a typical color encountered for embedded sources, such as Young Stellar Objects (YSOs). (Again, there are sources which are known to be redder. The Becklin-Neugebauer object, for example, has no published photometry at J, but has [H - K] = 4.1, and the massive YSO AFGL2591 has [J - K] = 6.0. YSOs with [J - K] = 7 are known, although not in large numbers.)

An example of a pair of simulated spectra is shown Figure 4.4, for the F110W filter. In this filter an image of a very red source will be dominated by the flat field response in the 1.2 to 1.4 micron interval, while for a blue source the most important contribution will come from the 0.8 to 1.0 micron interval. The results of our study for the most affected filters are shown in Table 4.4. The other filters are better.

• Even for the broadest NICMOS filters the wavelength dependence of the flat field response generates only small photometric errors, typically less than 3% for sources of unknown color. Not surprisingly, the largest errors arise in the 3 broadband filters whose bandpasses include some part of the regions where the flat field response changes most rapidly.
• The same results hold true even for filters at the most extreme wavelengths (e.g., F090M, F222M and F240M) because of their small bandwidth.
• It will probably be difficult to obtain photometry to better than the limits shown in Table 4.4 for the F090M, F110W, F140W, F205W and F240M filters, and observers requiring higher accuracy should contact the Help Desk at STScI for guidance.
• These errors can probably be corrected if more accurate photometry is needed, by taking multi-wavelength observations and using an iterative correction technique.
• For observers requiring high precision photometry, these represent non-trivial limits beyond which it will not be possible to venture without obtaining multi-wavelength images. In order to obtain 1% precision using the F110W filter, for instance, observers should observe at a minimum of one other wavelength. The color information derived from the pair (or group) of images could then be used to construct a more appropriate flat field image, which could then be applied to improve the color information.

Figure 4.4: Detected Source Spectrum. These are for sources with color temperatures of 700K (solid line) and 10,000K (dashed line). It is easy to see that the detected image will be dominated by the flat field response in the 1.2-1.4 µm region for a 700K source, while for a 10,000K source the detected image will be affected by the flat field response throughout the filter bandpass.
 

Table 4.4: Photometric Errors for Selected Filters

Filter	10,000K model. Error (percent)	700K model. Error (percent)
F090M	<0.1	1.9
F110W	1.1	2.9
F140W	0.7	3.1
F160W	<0.1	0.3
F187W	<0.1	0.3
F205W	0.4	2.1
F222W	<0.1	0.1
F240M	1	0.9

 

Extended Sources with Extreme Spatial Color
Variations

Our analysis has been limited to point sources, but some mention should be made of the situation for extended objects. A good example is the YSO AFGL2591. This has an extremely red core, whose [J - K] = 6 and which is entirely undetected optically. However, it also has a large IR nebula which is quite prominent at J and K, and in the red visual region, but much fainter at L, and which is probably a reflection nebulosity. Spatially, the nebula has highly variable color, some parts of it having fairly neutral or even slightly blue colors in the NICMOS waveband, while other parts are extremely red. Obtaining very accurate measurements of the color of such a source would again require the use of images at more than one wavelength and an iterative tool of the kind described earlier. A further example of this kind of complicated object is the prototypical post-AGB object CRL2688, the Cygnus Egg Nebula, which has an extremely blue bipolar reflection nebula surrounding an extremely red core. Techniques which require very accurate measurements of the surface brightness of extended objects, such as the brightness fluctuation technique for distant galaxies, will need to be applied with care given to the photometric uncertainties such as those discussed here.

Creating Color-dependent Flat Fields

NICMOS ISR 99-002 describes two methods for creating color-dependent flat fields, and programs and calibration files for making them are available in the software part of the website. One way of approaching the problem is to make monochromatic flats, by doing a linear least squares fit to several narrowband (and, if necessary for increased wavelength coverage) medium band flats, for each pixel. The slope and intercept images that result from such a fit can be used to determine what the detector response would be to a monochromatic source. Note that this method works best if the desired wavelength is within the range covered by the observed flats; extrapolation with this method gives questionable results.

If the source spectrum is known, a composite flat made from the weighted sum of the narrowband flats in the passband of the observed image can be made. A program to do this, given an input spectrum and the calibration database in STSDAS, is available. If you have a variety of sources in your image you may want to make several flatfields and apply them to regions defined by some criterion, say color as defined by a couple of narrowband images on either side of the broadband image.