Astron. Astrophys. 318, L5-L8 (1997)

3. Nontidally and tidally split comets

The relative contributions from the differential deceleration and the separation velocity to the rate at which two components of a split comet drift apart appear to be an important criterion for discriminating between the tidally and the nontidally split comets, as shown below.

An updated list of the nontidally split comets is presented in Table 1. With no exception, the observed fragment configurations show that the principal nucleus is always the leading component, with all the companions trailing behind. These configurations imply that deceleration effects clearly prevail over separation-velocity effects. The differential decelerations attain values typically between a few and 500 units of the solar attraction. All companions vanish before does (if ever) the principal nucleus. The duration of a companion's visibility was found in Paper 1 to be generally correlated with its deceleration: the lesser the deceleration, the longer the lifetime.

Table 1. List of known nontidally split comets.

The tidally split comets are listed in Table 2. Three were observed to have broken up into more than two components: two at Jupiter (D/Shoemaker-Levy 9 and 16P/Brooks 2) and one at the Sun (1882 R1 = the Great September Comet). Comparison with the nontidally split comets indicates that an average tidal-disruption event generates a significantly larger number of fragments.

Table 2. List of known tidally split comets.

Numerous investigations of D/Shoemaker-Levy 9, the most extensively studied tidally split comet, firmly established that the most massive components - G, K, and L - were all near the middle of the nuclear train, while the leading nucleus A was much less conspicuous and obviously less massive (e.g., Hammel et al. 1995). This evidence is supported by the results from the orbital determinations (Chodas and Yeomans 1996) for the comet's 21 components, which yielded excellent solutions without the need to incorporate nongravitational terms in the equations of motion. A more recent, extensive study of discrete secondary-fragmentation episodes (Sekanina et al. 1996), which were found to have occurred over a period of many months following the comet's encounter with Jupiter in July 1992, implies the absence of any detectable differential decelerations except for the motion of the component P₁ that disintegrated entirely before reaching Jupiter in July 1994.

The only other comet known to have split tidally near Jupiter is 16P/Brooks 2. The closest approach, to 2.0 Jovian radii from the planet's center, took place in July 1886. Unlike Shoemaker-Levy 9, Brooks 2 was perturbed by Jupiter into a slightly hyperbolic post-encounter jovicentric orbit, which brought the object to 1.95 AU from the Sun in 1889. Barnard's (1889) drawing (also cf. Fig. 1 of Sekanina 1996) made eight weeks before perihelion shows the principal nucleus A (the component that is still surviving today) to be trailing the companion nuclei. Only the companion C was positively identified to have separated from A at Jupiter. Solving for both the deceleration and the separation velocity as unknowns, I ascertained that the deceleration was indeterminate. Solving for the separation velocity only offered a better fit than all the other models that incorporated the deceleration (Sekanina 1978). The third component, B, was found to have separated from C nearer the Sun, about 19 months after the comet's encounter with Jupiter (Sekanina 1977, 1982). This episode may have been either a secondary-fragmentation event (similar to those observed for Shoemaker-Levy 9) or, less probably, an independent nontidal splitting.

The nucleus of the brightest member of the sungrazing comet group, 1882 R1, was observed after perihelion to consist of up to six separate components, arranged - like the fragments of Shoemaker-Levy 9 - in a rectilinear train immersed in a sheath of nebulous material (Kreutz 1888). However, useful orbital information is available for only the four components nearest the Sun. The two brightest and longest surviving components were the second and the third from the sunward end of the train, so that once again the leading component was not the principal nucleus. The solutions that included the deceleration and those in which was replaced with a transverse component of the separation velocity fitted the data equally well (Sekanina 1977). This equivalence was explained as due to an extremely steep decrease in the deceleration (assumed to vary inversely as the square of heliocentric distance) near the perihelion point of a sungrazing orbit (Sekanina 1978, 1982). From the virial theorem, the relationship between the two quantities for the orbit of 1882 R1 is , where is in m/s and in units of the solar attraction. If the separation velocity can be interpreted as an approximation to the equatorial rotational velocity, the minimum effective diameter of the parent nucleus can be calculated from , where is a critical rotation period and is the range of for the components located at the train's ends. Only a lower limit to this quantity can be derived from the available results for the first and the fourth components (Sekanina 1977): m/s. For an assumed nucleus bulk density of 0.3 g/cm³, for example, hr and km, a plausible value.

Another tidally split sungrazer, 1965 S1 (Ikeya-Seki), displayed only two nuclear components. Even though the principal (and systematically the brighter) nucleus was the leading component, the derived differential deceleration for the companion is very small and outside the range of values indicated by the nontidally split comets (Sekanina 1978, 1982). This circumstance suggests that, once again, one deals here with a disguised separation-velocity effect, in which case one now obtains m/s and, with the same critical rotation period as above, km. Thus, the leading position of the principal nucleus presented a signature of the direction of nuclear rotation rather than of the companion's differential deceleration.

I thus find that among the three tidally split comets that displayed more than two nuclear fragments, the principal nucleus was never the leading component and that the leading position of the principal nucleus of the two-component tidally split comet Ikeya-Seki should not be interpreted as an effect of a deceleration. It can safely be concluded that the motions of the tidally split comets are essentially determined by effects of the separation velocity acquired by the components at the time of their splitting. The physical significance of this fundamental difference between the two categories of the split comets is briefly discussed in Sec. 5.

© European Southern Observatory (ESO) 1997

Online publication: July 8, 1998
helpdesk.link@springer.de