Building the master calibration images

The HRT system takes several calibration images: Bias, dark and flat fields, before and after the observations.

Bias: Image with closed shutter and exposure time zero.

Dark: Image with closed shutter and exposure time not equal to zero. This image is used

Error during the combination

End of the reduction process Next reduction step

If the number of calibration images < 3

Checking the total number of images and the stability of the mean

Saving the master calibration image and creating a log file Combining the single images to a master calibration image image values and rejecting outliers

Figure 4.5: Flow chart with the main steps for building the master calibration images

to eliminate the dark current of the CCD camera.

Flat field: Spectrum of a continuum lamp. The flat field is used to define the order positions and the quantum efficiency correction.

The respective images are averaged and then used as master calibration images. The scientific images have to be corrected by bias and dark. All pixels in the image have a bias value which has to be corrected. In Fig. 4.4 (left image), one sees how the corrections influence the scientific images. In the left image, the raw image is shown. The spectral orders are only slightly visible. This is a contrast effect because the difference between the pixel values is not very high. After the bias correction, Fig. 4.4 (middle image), the orders are clearly visible. However, the image shows individual white pixel so that the image looks like grainy. This is caused by the dark noise. After the dark correction (Fig.

4.4 (right image)), the spectral orders are clearly visible and the dark noise is mainly eliminated. The average dark noise in the images is 20.01±0.10 elec/600 sec. This high dark noise is caused by the fact that the CCD camera is only cooling to -25 C^◦.

The main steps to build the master image are similar for bias, dark and flat field. The flow chart in Fig. 4.5 shows the main steps of this part. The first step is to check the variation in the calibration images. The percentage of relative variation rv_i between the arithmetic mean of the single images himii and the median of this values is calculated.

The equation is:

rvi=

himii −median(him1i...himni median(him₁i...him_ni

· 100. (4.1)

For the calculation of rv_i for the darks, the single images are corrected by bias and for the rv_i for the flats by bias and dark. The rv_i is plotted in a log file.

The calibration images are split in two lists. If the same number of images was taken before and after the observation, the images taken at the start of the observation are collected in the first list and those taken after the observation in the second list. Then, the percentage of relative variations of the images are checked for both lists. Normally,

4.2 Reduction pipeline 31

the total number of images in both lists is greater or equal to 3. The standard deviations of the percentage of the relative variation of images (Eq. 4.1) is calculated for both lists and used as thresholds to build the master calibration image (Sect. 4.4.1). If the standard deviation is less than 3.5, then it is reset to the minimal threshold of combination 3.5.

Ifrv_iis greater than a threshold (Sect. 4.3), the corresponding image is not used for build-ing the master calibration image. If the total number of images in one list is less than 3, an error message is obtained and the images are collected in a new list. The content of this new list is checked. Now there are 3 possibilities:

1. The total number of images is< 3: The reduction ends, because one needs at least 3 calibration images to build a master calibration image.

2. The total number of images < 6: The new list is not split in two lists and the REDUCE proceduresumfits.pro(Sect. 4.4.2), is used to build the master calibration image. The standard deviation of the percentage of relative variation is calculated for the list and used as threshold for the proceduresumfits.pro. If the standard devi-ation is less than 3.5, then the value is reset to the minimal threshold of combindevi-ation 3.5. The procedure sumfits.pro (Sect. 4.4.2) adds the images directly. Thereafter, the summed image has to be normalised by the number of images. This image is trimmed and rotated with the REDUCE function clipnflip.pro (Sect. 4.4.4). It is used as master calibration image.

3. The total number of images≥6: The new list is split in two lists and the procedure sumimage h.pro (Sect. 4.4.1), is used to build the master calibration image. This possibility is similar the normal case, the only difference being is that the calibration images are not split in calibration images before and after the observations.

The single calibration images are averaged with the procedure sumimage h.pro (Sect.

4.4.1).

As next step, it is checked if an error arose during the combination of the images and the keyword ’’ERR’’ (Sect. 4.4.1 and 4.4.2) was set. If the keyword set, then the pipeline stops and an error message is written to a file.

When building the master dark and flat field the image dimensions of average dark and flat field are compared with the image dimensions of the master bias. If the sizes are not equal, then the pipeline stops and an error message is written to a file.

The master bias is subtracted from the average dark and flat field and additionally, the flat field is corrected from the dark contribution. The dark correction is not performed if the arithmetic mean of dark is less than the threshold because in this case a dark contribution is negligible. After the subtraction of the master bias, the dark is time normalised.

The results are saved as master calibration images (hereafter, bias, dark and flat field). To monitor long-term changes the arithmetic mean of the single images and the corresponding standard deviation are saved in a log file. Furthermore, the standard deviation of the both lists and the arithmetic mean, the median and the standard deviation of the master calibration image are saved.

Finding and labeling the several orders

Saving the coefficients of the positions fits

End of the reduction process Next reduction step

If the number of single order = 0

Figure 4.6: Flow chart with the main steps of the order definition