The issue is due to an anomalous offset calculation during the Level 1 processing, as it can be deduced looking
at the wave in the baseline in band A at high altitude.
The problem is that in some anomalous cases, the offset from the auxiliary file is used for the processing
instead of the one present in the input Level 0 product
(there is an offset measurement every four limb scan in nominal instrument operation).
The issue was recognized by the processor developers as a weakness in the algorithm baseline, since
the offset of the input Level 0 product (if not corrupted) should be always preferred with respect
to the one provided in the auxiliary file. This feature will be modified in future version of the Level 1
The corrective action consists in re-moving these files from the D-PAC ftp server.
The re-processing will be done with an updated version of the IPF (V5.02)