The dependence of the circumnuclear coma structure on the properties of the nucleus -: III.: First modeling of a CO-dominated coma, with application to comets 46 P Wirtanen and 29 P Schwassmann-Wachmann I

We present the first gasdynamic simulations of the coma formed by the diffusion from a comet nucleus interior of a volatile molecule at large heliocentic distance. The method used is a generalization of that described in J. F. Crifo et al. (1995, Icarus 116, 77-112). The molecule is assumed to be CO. Three homogeneous nuclei with the same mass, but different shapes, are considered. Molecular production rates and nucleus sizes appropriate for Comets 46 P/Wirtanen at 3 AU from the Sun and 29 P Schwassmann-Wachmann I at 6 AU are considered in alternance. The main results are: (1) a fluid region surrounds the whole nucleus (not only the sunward hemisphere) even for production rates in the 10(25) molecule per second range; (2) the densest part of the near-nucleus coma is not necessarily the sunward hemisphere; (3) strong day-to-night asymmetries in the CO velocity are predicted, which are due to the surface temperature distribution, not to differences in surface gas production rate from point to point; (4) even in the case of a near-uniform production of CO over the surface, weak shock structures appear in the coma at the terminator, due to the surface temperature gradient (thus, even a uniform gas production over a spherical nucleus does not produce a uniform coma!); (5) shock structures are also formed in the vicinity of surface concavities, even if the gas production is uniform; (6) the ejection of dust grains up to at least millimeter sizes is possible from both the sunward and the anti-sunward side of the nucleus; (7) the terminal dust grain velocities are comparable to those computed for a H(2)O-dominated coma at a similar gas production rate. The reasons why we believe that CO is the best candidate for forming the distant comas of comets are discussed at length, but we also indicate why the present results are not critically affected if CO is replaced by another candidate volatile molecule (e.g., CO(2), HCN, H(2)CO...), or by a mixture of such molecules. The CO velocities we derive for 29 P Schwassmann-Wachmann I are shown to be in excellent agreement with those experimentally derived for this comet (J. Crovisier et al, 1995, Icarus 115, 213-216; N. River et al. 1997, Science 275, 1915-1918); as regards the experimentally derived temperatures, we show that their interpretation is very complicated and will require, in particular, the incorporation in the present model of a gas translational relaxation scheme. (C) 1999 Academic Press.