Astron. Astrophys. 335, L46-L49 (1998)

2. Observations and data reduction

In 1996 comet 46P/Wirtanen moved towards perihelion (perihelion passage: March 14, 1997) along the part of its orbit that will eventually be covered by the ROSETTA mission. The comet was monitored from ESO, La Silla, while moving inbound from 3.5 AU to 1.1 AU heliocentric distance, r , by means of broad-band filter imaging and spectrophotometry. The analysis of the imaging was done by Boehnhardt et al. (1997). Here we present the results of the spectrophotometric observations obtained from July to December, 1996 (2.8 AU to 1.6 AU). The spectra were taken with the ESO Faint Object Spectrograph and Camera 2 (EFOSC2) mounted on the 2.2m ESO/MPI Telescope except for Nov. when a similar instrument (EFOSC1) was used at the ESO 3.6m Telescope. In July EFOSC2 was still equipped with its old, less sensitive CCD, whereas for all other EFOSC2 observations a new UV flooded chip was used which led to an increase in sensitivity by a factor of 7 in the blue and 2 in the red. The spectral resolutions are specified in Table 2. The slit width was 2" in July, Sept. and Dec. and 1 in Oct. and Nov. (slit length = 5 The centering of the slit on the comet was evaluated before and after each exposure. Only well-centered spectra were kept for the analysis.

The spectra were reduced using the ESO/MIDAS and NOAO/IRAF standard reduction contexts for long-slit spectra. After bias subtraction, flatfielding and wavelength calibration the spectra were flux calibrated using the spectrophotometric standard stars Feige110, LT1020 and LT 377. As the comet did not fill the field of view of our CCD the sky could be determined on both sides of the comet and was subtracted accordingly. All spectra obtained during each observing run were co-added. 46P/Wirtanen was rather faint and did not show considerable short-term variability as is evident from broad-band imaging photometry (Meech et al., 1997; Boehnhardt et al., 1997). The co-addition of spectra taken on consecutive nights therefore led to an improved signal-to-noise ratio without introducing a measurable error due to possible short-term variations. The production rates were initially determined from fluxes integrated over slit lengths corresponding to the entire detectable comet. As the coma increased in size over the months, we used a common slit length of 4" for our analysis, after confirming that this has insignificant influence on the the resulting production rates. The emission band fluxes in a rectangular aperture (4"  slitwidth) are given in Table 1. For the continuum subtraction we measured the continuum bordering each emission band and approximated the continuum contribution to the band by interpolating between left-hand and right-hand continuum. We also fitted a solar analog to the spectra to demonstrate the overall shape of the continuum (Fig. 1). For CN, C₂, and C₃ the fluxes were converted into column densities using the fluorescence efficiencies applied by A'Hearn et al. (1995) (taking into account the dependences on heliocentric distance and velocity in the case of CN (Schleicher, 1983)). For NH₂ we used unpublished recalculated g-factors which do not differ by more than 10% from those of Tegler & Wyckoff (1989) for the bands observed here. The production rates in Table 2 were determined with the Vectorial model (Festou, 1981) and the lifetimes given by Schulz et al. (1994) for CN, C₂ and C₃. For NH₂ we used  = 5300s and  = 62000s (at 1 AU), average values derived from scale-lengths given by Cochran et al. (1992) and Fink & Hicks (1996). The parent velocity was varied as 0.85 km/s   (Cochran & Barker 1986) while the daughter velocity was arbitrarily set to 1 km/s. If the emission of a particular species was not detectable in a spectrum, the 3 upper limit to its production rate was determined. To check our results we additionally calculated the production rates with the Haser model using the parameters of A'Hearn et al. (1995) for CN, C₂ and C₃. Although the absolute values of the Haser production rates are systematically higher than those of the Vectorial model, all effects described in this paper were confirmed to be present in both cases. The dust production was derived from a region in the spectrum (5200 Å - 5250 Å) which is known to represent clean continuum. It is given in Af , a quantity introduced by A'Hearn et al. (1984) and now widely used to measure the production of dust in comets. The empirical correlation determined by one of us (CA) from published values of Af and of the dust production rate , for various comets near 1 AU indicates that with Af in 10³ cm and in 10⁶ g/s, these two quantities are expressed, approximately, by the same simple number (cf. Arpigny et al., 1998). The production rates are given with the 3 error of the average value computed from the individual spectra.

Fig. 1. Evolution of the spectrum of 46P/Wirtanen. The continuuum is fitted as a dotted line.

Table 1. Photometric fluxes of comet 46P/Wirtanen.
Notes:
3 upper limits from noise in neighbouring continuum
From high-resolution spectra, aperture (4 1

Table 2. Production rates, Q, and abundance ratios.
Notes: Values from Vectorial model. Haser model values in parenthesis (Farnham & Schleicher, 1997 and priv. com.)

© European Southern Observatory (ESO) 1998

Online publication: June 18, 1998
helpdesk.link@springer.de