Some artefacts have been reported in the initial part of some of the RED spectra of the UVES Echelle pipeline-processed data.
In the first 10-20 nm of many RED settings, corresponding to the initial 2-4 echelle orders, there is the danger of spectral artefacts in the data taken until 2007-03-31. These are due to those echelle orders being hit by bad CCD columns, in combination with a too high value used for the kappa clipping parameter in the optimal extraction (20 instead of 10). The background then might contain artificial "absorption" features which are subtracted from the signal and cause spurious emission features. This effect shows up often (not always) in the early years (until 2007-03-31). We have found these features in particular in the 564, 580 and 860 nm settings.

If this effect is strong, then the artefacts can be recognized easily and there is no other choice than ignoring these spectral regions. In unfavourable cases, the artefacts have an amplitude and a spectral shape that can be mistaken as an astrophysical signal. Unfortunately there is no way to safely predict under which circumstances this effect occurs. It can be strong on one day and absent the next day. If in doubt, always check the background signal (column #4 in the spectral table) against the extracted and background-subtracted signal (column #2). If both have an anti-correlated "signal", this is an artefact. Ideally the background signal should be smooth.

We plan a re-processing of the data from these early years. A new announcement will follow once ready.

Read more...