Astron. Astrophys. 333, 369-373 (1998)

2. Instrumentation and observations

We observed Hale-Bopp by using an imaging spectrometer in the 0.4-1 µm spectral range on the 1.5 m telescope at Sierra Nevada Observatory, Granada (Spain), on 20-24 March 1997. The comet was at a heliocentric distance = 0.94 AU and a geocentric distance = 1.32 AU. The phase angle was . The spectrometer used a Thomson CCD detector, 384 288 pixels, of 23 23µm size, cooled by liquid nitrogen at C. The telescope was configured at f /8, yielding a scale of 0.4 arcsec per pixel. In order to increase the signal-to-noise ratio, the readout electronics allows selection of different operation modes by summing pixels on chip. We have used two modes by binning 2 or 3 pixels spatially and 2 spectrally, giving 0.8 or 1.2 arcsec per pixel along the slit and = 50 Å, respectively. The images were obtained by acquiring the slit image at all wavelengths during the right ascension movement of the comet ( 0.007 arcsec/sec). This produced images with different spatial scales, depending upon the selected instrument operation mode and exposure time. The images were also affected by geometrical distorsion, due to the fact that the angular resolution along slit was fixed and determined by the pixel size (0.8 or 1.2 arcsec) while across slit it was determined by the comet velocity and the exposure time (see Table 1). Details on the instrumentation are found in Bellucci et al. (1997). Here we will discuss only the image cubes taken on March 20 at 18:44 U.T. and March 22 at 18:43 U.T.; the reduction of the other images is still in progress and the results will be presented in the future. Relevant data set information is reported in Table 1. The data cubes consist of 144 monochromatic (bandpass = 50 Å) images with different angular resolutions (1 arcsec = 970 km).

Table 1. Summary of the observations

The slit was oriented N-S during the observations. Spectra were bias-subtracted and flatfielded corrected following standard procedures. A preliminary reduction for removing instrumental and atmospheric features from the comet spectrum has been done by dividing the spectra to that of a standard area named MS2 and located in Mare Serenitatis on the Moon ( N E, McCord et al. 1972). The normalised spectra reasonably approximate the spectral reflectivity of the coma, , defined as the percentage of solar light scattered by the dust grains. However, due to the different airmasses of the comet (z = 3) and the Moon (z = 1.2), the telluric features are not completely eliminated. We have attempted to correct the atmospheric extintion by using a radiative transfer computation (LOWTRAN, Kneitzys et al. 1983). Fig.1 shows (left and center panel) the fraction of solar light transmitted through the Earth atmosphere and computed at two different airmasses. On the right panel their ratio is shown. The reflectivity has then been computed by means of the following relation:

where and are the spectra of Hale-Boop and MS2, is the atmosphere transmission at and , is the reflectivity of MS2 (McCord et al. 1972). The calculations made using the LOWTRAN code should be taken to provide a crude estimation of the decrease of radiant intensity through the atmosphere. This because the code uses average atmospheric profiles of the molecular species and aereosols which can be slightly different from a local situation. However, since we are interested to relative colour differences between portions of the image, residual atmospheric features do not constitute a serious problem. Curvature of spectra caused by atmospheric refraction has been corrected with an appropriate geometrical transformation (Bellucci et al. 1997).

Fig. 1. Left and center panels: Transmission of the Earth atmosphere as computed using the LOWTRAN radiative transfer code at airmasses and . Right panel: Ratio between and .

© European Southern Observatory (ESO) 1998

Online publication: April 15, 1998
helpdesk.link@springer.de