Instrument web page | Data products page

SOPHIE : Final reduction steps

Extracted spectra in wavelength space

This page illustrates some of the final steps in the reduction of SOPHIE spectra. We have selected three orders from the observation of HD 34078 done on 2007-02-10 in HR mode. The first row shows the extracted data for each order in the e2ds file, plotted vs. wavelength, obtained from the associated wave file, while the second row shows the same data after division by the associated blaze file. Note that the standard e2ds data product is not de-blazed.

Order 8, flat-fielded

Order 9, flat-fielded

Order 10, flat-fielded


Order 8, deblazed

Order 9, de-blazed

Order 10, de-blazed

De-blazed, rebinned spectra after order reconnection

The third row shows the same wavelength intervals from the deblazed, reconnected and rebinned s1d file. Note how the reconnection improves the S/N ratio in the intervals corresponding to the order edges, thanks to a substantial order overlap. Note that s1d spectra have not been corrected neither for atmospheric extinction nor for the instrument response. The flux given is relative to the flat-field lamp energy distribution.

Interval for order 8

Interval for order 9

Interval for order 10

Spectral sampling and correction for Earth's velocity

Users should be aware that the s1d data product as delivered has a 0.01 Å wavelength step. This is smaller than the actual sampling done by the CCD, which is 0.019 Å/pixel near 3900Å, 0.024 Å/pixel near 5000Å and 0.034 Å/pixel near 6900Å. Note that the s1d spectra have been corrected for the Doppler shift of the telescope relative to the barycenter of the Solar System, and that the lines are thus not at the same wavelength as in the e2ds spectra. The fourth row illustrates this in the case of HD 34078 for two different interstellar CH absorption lines.

CH[+] line at 4233Å, original sampling

CH line at 4300Å, original sampling

Same, after rebinning and Doppler shifting

Same, after rebinning and Doppler shifting