2 An Echelle Spectrum

In ‘traditional’ spectrographs a dispersing element—typically a diffraction grating or prism—is used to produce the spectrum. This results in a single spectrum which can be imaged using a CCD or other type of camera. The data can then be extracted using a suitable program. The recordable part of the wavelength range covered by this type of spectrograph is limited by the size of available image sensors, i.e., CCDs.

A quick inspection of a CCD image of such a spectrum will also reveal that much of the detector area away from the spectrum itself is unused. One method of optimising the use of the available detector area is to use an échelle spectrograph.

An échelle is a diffraction grating in which the rulings are much further apart than usual. This leads to spectra of very high dispersion, but only over a short wavelength range in each order. As well as being ‘short’, the high orders will overlap. To overcome this effect a cross-dispersing element is used to produce an order separation. Figure 1 shows a small part of such an échellogram recorded with a CCD camera. You can see a short part of three orders which run from the top to the bottom of the image at a slight angle. In the order to the right you can see a couple of absorption features. Several cosmic-ray events (bright spots) are also visible.

Figure 1: Echelle Image: Part of an échelle image produced with UCLES and a CCD camera.

Echelle spectrographs for astronomy are designed so that the wavelength coverage in one order will overlap the coverage of the adjacent orders. (That is at least for the middle orders in the full échellogram—there may be some gaps at the extremes of the image.) Using a suitable detector–usually a CCD—these spectral orders can be recorded.

The extraction of an échelle spectrum from a set of images is, in principle, similar to the extraction of a single-order spectrum. Additional complexity arises because:

• There are more data to extract.
• The orders can have a more complex shape than those from a single-order instrument.
• The high dispersion used can make it difficult to distinguish between true spectral features and cosmic-ray events.
• Flat fielding the data can be difficult.
• In some cases adjacent orders overlap slightly in the spatial direction (there can be several reasons for this) making accurate background subtraction difficult.

Fortunately, there are several dedicated software packages available which address these specific features of échelle data reduction.