In some cases there is a difference between the sub-spectra from different detectors in their overlapping regions, where they should in principle be at the same flux level. As an example see Figure 6.4 where the SW3 detector is obviously misaligned compared to the rest of the detectors. Each sub-spectrum has an independent calibration and this misalignement is most probably due to uncertainty in the responsivity of the detectors, estimated independently for each observation.
The LWS calibration is based on Uranus, which is a point source within the LWS beam. The output of the automatic processing assumes the observed object is also point like in the LWS beam. Therefore for most point sources there is almost no difference between the levels of the detectors with the small differences being due to the uncertainty in the determination of the responsivity of that detector for that observation. For extended sources the sub-spectra are not expected to match. This is because LWS has a different effective beam size for different detectors due to fringing. A full description of the LWS beam and the fringing along with information on how to correct for these effects can be found in Section 6.2.
Another uncertainty, which can have an impact on the relative levels of the sub-spectra, is that of the determination of the dark current. On this subject, see Section 6.4.
It should be noted that SW1 has the least reliable photometric calibration of the LWS detectors, due to a long time constant which makes it visibly affected by long term transient behaviour (Section 6.9).
From a photometric point of view, it is strongly recommended that users should not attempt to merge the sub-spectra together. Nevertheless, the spectra can be fitted back together within ISAP (see Section 8.2.2).