What's new in the VIMOS pipeline public version 2.0

What's new in the VIMOS pipeline public release 2.0 ?

Here is an overview of the improvements realised since the previous public release (1.0, July 2004) of the VIMOS pipeline. More details can be found in the related sections of the VIMOS Pipeline User Manual.

The wavelength calibration for MOS spectra is now more robust, and accurate calibrations can be obtained even in the case of masks with a narrow distribution of slits (as those used for the observations of spectro-photometric standard stars). A new ``blind'' arc line identification method has been implemented, based on pattern recognition, that doesn't make use of the first-guess IDS models associated to the input calibration lamps. This option would simplify the reduction of those observations where the initial first-guess models appear to be too far from the real distortions. In addition to that, it is now possible to iterate in a simple way the obtained spectral distortions models, using them as increasingly accurate first-guesses.
The reduction of spectra from curved slits is now supported.
Sky fringing can now be subtracted from the reduced frames obtained from a jittered sequence of exposures, both in MOS and in imaging mode. In MOS mode two different methods have been implemented for this purpose.
More methods for the stacking of jittered MOS spectra have been made available.
A new recipe, vmmoscombine, is now available for stacking spectra from different nights.
MOS and IFU spectra, reduced with the recipes vmifuscience, vmmosobsstare and vmmosobsjitter, can now be flux calibrated using the spectro-photometric table produced by the recipe vmifustandard, and the new recipe vmmosstandard. Another recipe, vmspphot, can be used to apply a flux calibration to already reduced spectra.
In recipes vmmosobsstare and vmmosobsjitter the enforced specification of a CCD table, even when bad pixel correction was not requested, has been eliminated.
An incorrect computation of the RMS of the residuals in IFU fiber tracing and wavelength calibration has been eliminated. This had no effect on the data reduction products, but just on the evaluation of the error of local distortion models fits, while the overall RMS associated to the wavelength calibration was computed correctly. To fix this, one should square-root the RMS values listed in the extraction mask and IDS tables produced with the previous pipeline release.
In the optimal extraction of spectra obtained from the stacking of a jittered sequence of MOS exposures, the computation of the variance didn't keep into account the number of stacked frames. This had the effect of reducing the efficiency of cosmic rays hits rejection. On the other side, this had no effect on the optimal extraction in itself, since modifying consistently the optimal weights by a constant factor doesn't change the final result. Now this problem has been fixed.
The 1D-extracted sky is a new product of the recipes vmmosobsstare and vmmosobsjitter.
The recipe vmifucombine now combines IFU reconstructed images of the field-of-view from different quadrants keeping into account the different reference levels used in the relative transmission correction in each quadrant. This is just cosmetics, and it doesn't affect in any way other results from the data reduction process.
It is now possible to examine the intermediate products of the data reduction steps of the recipe vmmosobsjitter.
The total IFU standard star spectrum is now just summed over the fibers that are really affected by the standard star flux, eliminating severe background contamination, particularly in the case of multiplexed low resolution observations.