As part of the Service Mode package, QC1 reports about SCIENCE reduction are delivered. For echelle data reduced with optimum extraction, a set of three plots exists for each CCD (BLUE/REDLower/REDUpper) :

• a comprehensive QC1 report about the reduction process, including a quick-look of the result spectrum and its errors,
• a plot with the positions of the numeric extraction slit used during optimum extraction,
• a plot with the FWHM determined for the Gaussian fit.

The QC1 report for image slicer data is similar to the report for optimum extraction but excludes those quantities not provided by average extraction.

All plots come in gif format (for quick-look purpose) and in compressed postscript format (for printouts).

Check out here the description of the QC1 report. There are also descriptions of the position and FWHM plots and a tutorial on position and FWHM plots.
 

• top panel
• plot 1: central row and column (raw data)
• plot 2: cross-dispersion profile
• plot 3: spectral windows
• plot 4: reduced spectrum, errorbars, compressed view
• plot 5: histograms raw/reduced
• plot 6: signal position and FWHM (highly condensed version of position and FWHM QC 1 reports)
• bottom panel

Complete sample plot

Top panel. The name of the reduced spectrum, its setting values, and the names of the line table and master flat used for the reduction are indicated.

The central plot shows a sketch of the slit (width and length drawn to scale), and the average object position (OBJ) plus/minus the seeing observed.

The right top panel has alarm buttons for temperature and object centering criteria. Alarms are indicated red if

• the temperature difference between SCIENCE exposure and wavelength calibration is larger than 2 degree C,
• the object position is closer than 20% to the slit ends,
• the nominal S/N value is higher than a certain threshold value.
   

1. Central row and central column (raw data). This plot gives an indication about some sorts of irregularities and about the flux level. The red line shows the nominal BIAS level (taken from the bias QC1 plots).

2. Cross-dispersion profile (on a logarithmic flux scale). It has been derived by collapsing all columns in a subwindow 100x100 pixels wide (inclination of orders is neglected here).

3. Three spectral windows (6 A wide) to show single bins of the result spectrum. They are marked in plot 4 by the red broken lines and numbered as a-c.Horizontal broken line indicates zero flux.

The 390 nm settings show the spectrum twice: in addition to the normal one, there is a scaled version (factor of 5) in order to treat with the rather steep blue flux gradient showing up in many objects.

4. Reduced spectrum, errors. The full spectrum is shown here, with a 5A smoothed version in red. Y scale is in extracted ADU (modifed by flat field). A compressed version is also shown, to provide a quick look of potential ripple patterns. Nominal positions of some prominent spectral lines are marked blue. The 390 nm settings show the spectrum twice: in addition to the normal one, there is a scaled version (factor of 5) in order to treat with a steep blue flux gradient.

The errors shown are in the same format. They have been derived from the ERRORBAR file. The starting wavelengths per order are marked by blue broken lines, as well as some selected order numbers. The 390nm settings have the ERRORBAR file shown twice (with a version scaled by a factor 2).

5. Histograms. A histogram for the raw data is shown around the background (100...400 ADU; log frequency). The BIAS level is marked by the red line. Any shift of the maximum away from the BIAS level would indicate contributions from dark current or sky background.

A similar histogram is shown for the reduced data. It may be useful for estimating the quality of background subtraction. 0 level is marked red.

6. Signal position and width. These boxes provide condensed QC1 information about the signal position within the extraction slit, and the width of the extracted signal. Both are determined by the optimal extraction routine on approx 50-100 chunks per order. They are shown in full details in separate QC1 plots.

The narrow plot (left) visualizes the slit, with the signal position averaged across all orders (red) and the nominal centre position (black). Mean values per order are shown in the central plot (blown up to show the central 5 pixels). The right plot shows the FWHM per order.

More details are found in the order position and FWHM QC1 plots. There is also a tutorial on position and FWHM plots.

Bottom panel. Here are recorded some relevant observational parameters (exposure time, airmass, seeing, mean value of FWHM, slit width, bin width, start wavelength), plus QC parameters measured by the pipeline (mean values of raw frame, reduced spectrum, errorbar file).

The SCIENCE reduction QC1 report has the following components which are described below:

• top panel
• plot 1: central row and column (raw data)
• plot 2: cross-dispersion profile
• plot 3: spectral windows
• plot 4: reduced spectrum, errorbars, compressed view
• plot 5: histograms raw/reduced
• bottom panel

Complete sample plot

Top panel. The name of the reduced spectrum, its setting values, and the names of the line table and master flat used for the reduction are indicated.

The right top panel has an alarm button for the temperature. An alarm is indicated red if the temperature difference between SCIENCE exposure and wavelength calibration is larger than 2degree C.

1. Central row and central column (raw data). This plot gives an indication about some sorts of irregularities and about the flux level. The red line shows the nominal BIAS level (taken from the bias QC1 plots).

2. Cross-dispersion profile (on a logarithmic flux scale). It has been derived by collapsing all columns in a subwindow 100x100 pixels wide (inclination of orders is neglected here). The plot is useful for an indication about the slicer cross-dispersion profile. The example was observed with slicer No. 3 which has five slices in total.

3. Three spectral windows (6 A wide) to show single bins of the result spectrum. They are marked in plot 4 by the red broken lines and numbered as a-c.Horizontal broken line indicates zero flux.

The 390 nm settings show the spectrum twice: in addition to the normal one, there is a scaled version (factor of 5) in order to treat with the rather steep blue flux gradient showing up in many objects.

4. Reduced spectrum, errors. The full spectrum is shown here, with a 5A smoothed version in red. Y scale is in extracted ADU (modifed by flat field). A compressed version is also shown, to provide a quick look of potential ripple patterns. Nominal positions of some prominent spectral lines are marked blue. The 390 nm settings show the spectrum twice: in addition to the normal one, there is a scaled version (factor of 5) in order to treat with a steep blue flux gradient.

The errors shown are in the same format. They have been derived from the ERRORBAR file. The starting wavelengths per order are marked by blue broken lines, as well as some selected order numbers. The 390nm settings have the ERRORBAR file shown twice (with a version scaled by a factor 2).

5. Histograms. A histogram for the raw data is shown around the background (100...400 ADU; log frequency). The BIAS level is marked by the red line. Any shift of the maximum away from the BIAS level would indicate contributions from dark current or sky background.

A similar histogram is shown for the reduced data. It may be useful for estimating the quality of background subtraction. 0 level is marked red.

Bottom panel. Here are recorded some relevant observational parameters (exposure time, airmass, seeing, slit width, bin width, start wavelength, image slicer number), plus QC parameters measured by the pipeline (mean values of raw frame, reduced spectrum, errorbar file).