1 Introduction

By analysis of the emission structure on high-resolution dust-continuum images, the rotation period of the nucleus and the mean dust outflow velocity may be determined, and constraints may be set on the direction of the axis of rotation. Numerical simulation experiments and hydrodynamic modelling, based on observed structures in the inner coma, is a means of constraining fundamental properties of the emission properties of the nucleus and its dust and gas emission. One of the goals of such models is the determination of the physical properties and the activity pattern of the nucleus on the basis from observations, such as images of the coma structure.

Some steps have been taken towards reproducing by modelling some observed features in the coma of a number of comets, such as jet-like features of Hale-Bopp (Sekanina 1998; Sekanina & Boehnhardt 1998) and Halley (Crifo 1997b and references therein), but determination of detailed nucleus emission properties from imaging observations is currently not possible. Coma morphology has been modelled for comets in general and as specific objects (e.g., Sekanina & Larson 1986 and references therein; Sekanina 1987) and for Hale-Bopp specifically (e.g., Sekanina & Boehnhardt 1988; Sekanina 1998; Samarasinha et al. 1998). These works do not take into account interaction processes between outflows subsequent to emission from the active areas.

It has become evident that the formation of structures in cometary comae and their relation to nucleus activity is very complex and requires gas-hydrodynamic modelling in order to obtain an understanding of the underlying processes. The results of such models, incorporating outflow interaction processes, have been published for pure gas and dusty-gas emission cases and for surface topography (Kömle & Ip 1987; Kitamura 1986, 1987, 1990; Keller et al. 1994; Crifo 1995, 1997a, b). For improvement of any future modelling of these processes, ground-based imaging data of high resolution, such as the data set presented here for comet Hale-Bopp, will be of considerable importance.

C/1995 O1 (Hale-Bopp) reached perihelion on 1997 April 1.137 UT at a heliocentric distance of $0.914\ \rm AU$ (Marsden 1997). The geometry during the period of observation was such that the phase angle of $\sim$$35\hbox{$^\circ$}$ made the morning hemisphere of the nucleus unobservable. As the north pole of the nucleus was visible, the region of the sub-solar point and evening terminator of the nucleus was located in the general direction of the Earth.

In this paper, we present very high-resolution images of the near-nucleus coma of Hale-Bopp, and determine the nucleus rotation period and mean outflow velocity of the dust ejecta. The SVST has been used for successful planetary imaging on a number of occasions (e.g., Lindgren 1995; Warell 1996; Orton et al. 1996). Given the high surface brightness and coma detail of Hale-Bopp, it was therefore natural to use the instrument for specific studies of the intricate dust morphology of its inner coma. The apparition of this bright comet made it a unique target for studying the coma morphology close to the nucleus, and the high spatial scale obtainable with the SVST at its excellent site made it a very good instrument for such an investigation.