PHOTOMETRIC ACCURACY

The absolute photometric accuracy to be achieved with ISOPHOT ultimately depends upon the accuracy of the celestial standards. There are several calibration programmes from the ground and from the KAO aimed at establishing a network of primary and secondary standards. For wavelengths less than 50 m, stars will be the primary standards. At far infrared wavelengths, asteroids and planets are primary calibrators. As secondary standards, planetary nebulae will be used. The goal is to establish the fluxes of these objects to an accuracy of 10 to 20%.

All ISOPHOT single-pointing measurements involve an FCS calibration for each detector used. Each map or scan requires two FCS calibrations both at the beginning and the end of the measurement. The S/N of the internal calibration measurement depends on the requested S/N of the astronomical measurement, e.g. high calibration accuracy will not be sought when the source is only marginally detected. The internal calibrations will serve as transfer measurements to the celestial photometric standards. Different absolute detector responsivities, perhaps due to e.g. a different environment in irradiation by high energy particles on various orbit positions can be removed in this way.

One significant factor for the relative accuracy - the reproducibility of a certain signal level - has been described in Section 2.8 and some guidelines for optimising the accuracy have been given. It can be assumed that the FCS calibrations reach, in general, the same level of accuracy as the astronomical measurement so that relative photometric accuracies with respect to sky standards of about 5% could be achieved at best.

Furthermore, the determination of the sky background is essential for accurate photometry of fainter sources. Background gradients or background structure can strongly influence the flux level of the object signal. In particular, the cirrus confusion at wavelengths beyond 50 m should be carefully considered (see ISO Observer's Manual for details). It is advised to measure several background fields either by triangular or sawtooth chopping or by sparse mapping. The most accurate procedure but also the most expensive in terms of observing time is mapping.

For standard single-pointing photometry of point sources a correction factor must be applied to correct for truncation of the point spread function by the aperture field of view.

Photometry of extended sources will yield somewhat worse accuracy than for point sources. Beside the uncertainties listed above the flatness of the beam profile, which is known to about 5%, is another factor affecting the accuracy.

The design of the PHT AOTs already makes some provisions to improve the photometric accuracy of certain observing modes:

• Chopping should be used for faint sources to cancel out slow detector baseline drifts.
• The PHT-C mapping AOT uses the chopper for differential measurements in order to assess the baseline drift, a feature which is not possible with a simple repetition of staring mode observations.
• The PHT-C mapping AOTs use sampling patterns in which each sky position of a map is viewed by all pixels of the array in order to optimise the flatfield correction.
• Polarimetric observations are very sensitive differential measurements with different polariser settings, so each observation includes non-polarised photometry for consistency checks. For very long integration times per polariser, the time is divided into several blocks during which all three polariser directions are set up. These time blocks, which are at least 64 seconds long, are long enough to permit assessment of short term drifts due to flux changes introduced by the wheel rotations but still short enough to cancel out any long-term drift. The ability to measure polarisation levels below 3% may be improved if the measurement is repeated some time later (a few months) when the orientation of the focal plane has rotated relative to the sky; such differential methods can be used to determine the instrumental polarisation to high accuracy. The repetition of observations may be limited, however, by visibility restrictions due to spacecraft orientations relative to the Sun (see ISO OBSERVER'S MANUAL).
• Measurements of absolute sky brightness can be improved in accuracy by using the PHT internal fine calibration sources as reference.
• The dark current of a detector can be determined by measuring against the cold, i.e. switched off, fine calibration source.