| 1 | | I found that ASAP fails to align two datasets if they are observed at |
| 2 | | different frequencies. Such situation often happens if two datasets observed well apart |
| 3 | | in time are averaged together and hence is a critical issue for all projects which |
| 4 | | require long integrations (and observed with the doppler tracking on, which is |
| 5 | | standard at Mopra). Investigating the issue I found the following |
| 6 | | |
| 7 | | * scantable.freq_align fails on a merged dataset because it reckons that the dataset is already aligned in LSRK although the reference values are different for the first and the second scan |
| 8 | | * An attempt to align the merged dataset after changing its frequency frame to BARY is successful, but the two scans still have different reference values after alignment |
| 9 | | * Averaging all scans together in the merged dataset gives double or blurred lines if the reference value doesn't match (i.e. the spectra are just averaged channel by channel and the offset is ignored). |
