On 20th of December 1999, a new version (v8.7) of the ISO Off-Line Processing
(OLP) software was released for the ISO Data Archive users. The main item
of this release was a major upgrade of the LWS processing including the
Year 2000 (Y2K) compatibility. As a consequence of the processing
improvements LWS04 is now SCIENTIFICALLY VALIDATED. Below additional details
separately for the FP and grating modes.
FP:
The OLP 8.7 LWS FP products from AOTs LWS03 and LWS04 should show a marked improvement in the flux calibration in comparison to their OLP 7 counterparts. In OLP 7 the FP LSAN file gave the flux per spectral resolution element (in units of Watts cm^-2). In OLP 8 the units will be Watts cm^-2 um^-1, where the throughput correction (i.e. the FP transmission multiplied by the FP resolution element) has been performed using Mars as the calibrator. The other noteworthy change in OLP 8.7 is that fixed dark currents are subtracted for all FP observations, rather than those measured during the observations themselves. These fixed set of detector dark currents originate from a series of dedicated measurements made in revolution 650.
Despite these significant advances, some problems still remain to be addressed. These include corrections for line asymmetry, velocity shifts and the grating profile removal. Each of these effects is briefly described below and, where possible, a quantitative estimate of the resulting uncertainty is given.
1) Line Asymmetry
a) The line profiles observed using AOT LWS04 are generally asymmetric, with the long wavelength wing being higher than the short wavelength one. This phenomenon is partially attributed to transients (also known as detector memory effects). Transients are also present in grating spectra. Various methods for removal of transient effects are being investigated. The effect on line flux depends on the line-to-continuum ratio. For example, for a line with no continuum, the line flux can change by as much as 30%.
The effect on the velocity shift is relatively small at about +/- 3 km/s.
b) It is also possible that some of the profile asymmetry observed for FPL may well be due to loss of parallelism as the FPL is scanned.
2) Velocity Shifts
There appears to be a discrepancy in the measured and expected velocity shifts in LWS04 lines of up to +/- 30 km/sec, depending on the actual line being observed. At present the cause for this discrepancy is not known and is under investigation.
3) Grating profile
In OLP 8.7 the underlying grating profile is not removed. For LWS04, where the FP scans are quite short, the effect should be very small. For LWS03, however, there may be some effect on the slope of the continuum, depending on the extent of the FP scan at each grating position.
Removal of the grating response profile is complicated by the fact that the grating position is not known sufficiently accurately. This can result in a skewed spectrum. The user is recommended to refer to FP_PROC, the LWS Interactive Analysis (IA) FP processing tool which removes the grating profile. For LWS04 spectra the user can remove the skewness interactively, thereby deducing the actual grating position. For LWS03 observations, a least-squares fitting algorithm which minimizes the residuals in the overlap regions of spectra at adjacent grating positions is employed.
The effect of not removing the grating profile on FP line fluxes from
LWS04 observations is likely to be very small. However, for continuum fluxes
in LWS03 observations the effect may not be negligible and is expected
to be a function of the FP scan extent.
Grating:
The model for Uranus has changed between OLP version 7 and OLP version 8. The effect of the new model is that for a source decreasing in flux from short to longer wavelengths, the spectrum of this source will now be `flatter' in detectors SW1 and SW2.
With the new model, both the RSRF and photometric calibration have been re-derived.
The photometric calibration of SW1 has been slightly degraded in the centre of the detector i.e. this is 10% higher than before as more of the detector was used to set the level. The photometric level of SW1 will be updated soon to replace it back to the OLP7 level. The photometric levels of the other nine detectors remains the same as for OLP 7.
On checking all observations of Uranus against the model, a 'problem'
has been found in the data. This has the effect that detectors SW1 and
SW2 appear tilted down at short wavelengths and up at long wavelengths.
This tilt has been noted for all observations after the LWS was switched
off in the period, revolutions 380-442. Before this period the slope follows
the model. This may indicate a change to the RSRF of these detectors during
the switch off period but further investigation is needed before definitive
conclusions can be made.
The ISO Data Archive with OLP 8.7 can be accessed at http://www.iso.vilspa.esa.es/ in the Data Archive section.
Compiled from UK NDC (RAL) inputs by Timo Prusti, ISO Data Centre, 20/12/1999