This section gives estimates for the beam sizes for the LWS grating mode at the central wavelengths of the ten detectors. The beam sizes were determined using two sets of observations of the planet Uranus. One taken in revolution 158 and one taken in revolution 317. The observation in revolution 158 was a cross-scan with the grating in a fixed position in the RA-Dec frame of reference, whereas the revolution 317 observation was a cross scan with at each point a full grating scan in the spacecraft frame of reference. The beam sizes were derived from a normalized profile of the cross-scan taken the full width at half maximum of these profiles. A Gaussian fit to the profiles cannot be used to derive their width, as the beam profiles are not Gaussian in shape (especially for the SW detectors the profiles are more flat topped). The reason for this is that the instrument is not diffraction limited at the short wavelengths. Therefore at these wavelengths the beam size is limited by the aperture size of the instrument. The beam widths derived from the full grating scans in revolution 317 were derived at the same wavelengths as the widths derived for revolution 158. Please note that the beam size changes with wavelength following the fringe pattern, seen especially on the long wavelength detectors. The derived beam sizes in arcseconds are given in the table below. The wavelengths given are the approximate center wavelengths for the detectors in micron.
Det. Wavel. rev 158 rev 317 _ _ _ _
Y Z Y Z Y dY Z dZ
------------------------------------------- -----------------------
SW1 47 81 74 83 66 82 3 70 3
SW2 56 83 81 83 77 83 3 79 3
SW3 66 83 83 83 81 83 3 82 3
SW4 76 84 82 81 75 82.5 3 78.5 3
SW5 85 82 78 82 73 82 3 75.5 3
LW1 102 77 73 77 69 77 4 71 6
LW2 122 68 68 71 67 69.5 6 67.5 6
LW3 142 60 58 66 59 63 6 58.5 6
LW4 161 72 71 71 65 71.5 8 68 10
LW5 178 69 72 75 70 72 10 71 16
------------------------------------------- ------------------------
The last two columns of the table give the averaged beam size using the two independent determinations. For the SW detectors the estimated uncertainty given is based on the differences between the two determinations. For the LW detectors the amplitude of the beam variations due to the fringing in the spectrum (as determined from the full grating scans) is given as a more realistic uncertainty estimate for the beam size.
Note that these beam sizes are estimates, a better determination of the beam will be made using calibration observations of Mars.
The LWS grating spectra are normally collected with the grating scanning the wavelength range forward and backward alternatingly. This observational procedure has two important consequences both due to detectors memory effects. The first is that there is a shift between line centers as seen in forward and backward scans: due to the relatively slow time response of the LWS detectors to especially raising signals, a difference in the center wavelength of (strong) emission lines between forward and backward scans can be seen. The observer is advised that the trailing edges (signal going down) of the lines give in this case more accurate measurements of the line shape.
The second consequence is that the shape of the spectrum is different between forward and backward scans; depending on shape of the source's spectral energy distribution (SED) and flux intensity this effect may be seen in various detectors. The effect still needs to be fully characterized but the best known case is for detector LW2 on sources with 100 micron fluxes between 200 and 1000 Jy (indicative figures): long ward of 119 micron the spectra taken during forward and backward scans have totally different shapes. Observers are strongly encouraged to look at all the individual scans on top of each other before averaging them in order to check for the presence of this effect and avoid erroneous conclusions from their analysis.
While LW2 is the worst detector in this respect, it is noted that this effect is also present in LW3 and LW4 for the same flux range previously quoted for LW2; observers are also warned about the presence of this effect on the shorter wavelength detectors in case of sources with much stronger fluxes and steeply decreasing SEDs.
ISO Science Operations team
and
LWS Instrument Team
25 August 1997.