Colour terms for FORS2 and Paranal exctinction coefficients in the UBVRI passbands are determined by QC Garching via a one-step multilinear least squares fit to the data produced by the Pipeline. This is achieved by using a simplified version of the method described by Harris et al. 1981, PASP 93, 507.
The details on the implemented procedure are decribed here.

Basically the method finds the best fit solution, solving simultaneously for the zero-point, colour term and extinction coefficient. This has the advantage that different standard star fields can be used, without the need of airmass pairs. As a consequence of this, data from many nights can be combined together to give a more reliable solution. This page gives the data for the most recent period. Data for previous periods can be found here

Plot Description: In the upper left box the original magnitude differences (Landolt-Instrumental) are plotted as a function of the standard star colour, e.g. (B-V) for the V band. Triangles represent observations performed at an airmass > 1.2, while circles refer to data taken at an airmass <= 1.2.
After the solution is found, the corrected differences are again plotted (lower plot in the upper left box). Empty symbols represent the data which have been rejected, i.e. those points which deviate more than 2 times the RMS deviation (sigma), which gives an estimate of the measurement error. The distribution of the computed zeropoint deviations from the average value are shown in the upper right box. The dotted line represent the data which have been rejected. The behaviour of the observed magnitude differences (Landolt-Instrumental) after correcting for the colour term (left panel) and the extinction coefficient (right panel) are shown in the lower boxes.

Trending

The following tables report the colour terms and extinction coefficients (in mag/airmass) for FORS2. Only values from photometric nights are used. All estimated uncertainties are one standard deviation. Since P81 the U_SPECIAL filter is no longer monitored.