Data handling is also part of the SCS. The SCS anticipates within the evening twilight which filter is the one with the most urgent need, depending on flat-field history, flat stability, and upcoming fields; a decision that could be refined later during dawn twilight. Once all data have been transferred to STELLA (or GREGOR), a preparatory pipeline assures that, for any given night, the most recent bias and flat-field frames are available. Depending on the secured flat-field stability, a warning is issued if the flats in proximity to the night in question are too far off in time in order to assure a pixel-to-pixel stability exceeding 10 −3 .
The data reduction pipeline itself is based on the STELLA WiFSIP pipeline (Granzer et al. 2015, in prep.). There, the raw frames are bias subtracted, super-flat divided and calibrated for astrometry and photometry using the PPMXL catalog (or the GAIA catalog later on). Then frames are bad-pixel interpolated using a mask obtained from the flat field, object masked and line fitted to subtract background residuals, aligned, and average combined per field visit (three field images per night), using standard routines within the IRAF environment.
For the image combination, the images are flux scaled according to their exposure time and zero-point magnitude, and weighted similarly as described in Erben et al. (2005, AN 326, 432) but with additionally a seeing-weight factor inversely proportional to the square of the Full-Width-At Maximum. A threshold is used to reject broad bad pixel features (such as along the frame borders) and an upper sigma-clipping algorithm is used to reject cosmic rays and other short transients. The combined images are astrometric re-calibrated and re-sampled onto the same world coordinate system. Then, the frame sequence is average combined using the propagated scales and weights to account for the varying depth but without sigma-clipping rejection because image ellipticity and FWHM vary from night to night.