Position offset correction
The footprints of the PHT-S pixels were peaked rather than flat-top,
therefore observing a compact source outside the optical axis of the pixel
could change the measured signal. Since the calibration was set up for the
pixel centre, off-centre observations have to be corrected in order to derive
their correct flux values. In addition, because the footprint profiles varied
with wavelength (there were especially large jumps at 3.7 and 8.8 um)
off-centre position lead to the appearance of spectral artifacts, too (an
example is shown in Fig. 1). In order to correct for both the incorrect
flux values and the spectral artifacts, we determined the offset
of the source from the centre by comparing the accurate position of the
object with ISO's coordinates. The 2-dimensional footprint map
(we utilised the ones derived by Leinert et al. 2002) was then sampled
at the offset position, and the measured flux was scaled with
the ratio between footprint values at the centre and at the offset
location. This procedure was repeated for each detector pixel independently.
For accurate source positions we consulted the 2MASS catalog or SIMBAD; if no
coordinates better than 1" were available then no correction was performed.
Fig.1: Example for spectral artifacts introduced by off-centre location of
a point source in the PHT-S beam. The object (Ced 112 IRS\,4,
ISO_id:62501217) was offset by +4.8" and -0.6" in satellite Y-
and Z-direction, respectively. Vertical marks show the wavelengths where the
most obvious artifacts, in the forms of discontinuities, appear.