Highlights•We calculated 80 polarimetric curves in different spectral domains for 20 comets.•Maximum polarization of comets correlates with the slope of polarimetric curves.•There are two groups of comets with low and high polarization degree.•Maximum polarization of comets correlates with dust-to-gas ratio in their comae.

AbstractThe ground-based astronomic observations of comet cannot provide a proper phase angle coverage that is needed to estimate with a reasonable accuracy all of the attributes of comet polarization phase curve. To find the best approximation to the phase polarization dependences observed for comets, we apply a simple empiric formula that has already shown good results when operating with asteroid and lunar polarimetric curves (Shestopalov, 2004; Shestopalov and Golubeva, 2015). From the set of comets present in DBCP (Kiselev et al., 2006), we selected 20, for which the calculation of regular polarimetric curves (i.e. the phase angle – polarization dependences with a low level of nonsteady activity) was possible. Within the phase angle coverage area for these 20 comets, a potential user can reproduce 82 best-fitting polarimetric phase curves in various spectral domains. Then we analyzed the properties of negative and positive polarization of the comets. The interrelation between the averaged polarimetric slope h at the inversion angle and wavelength was found. In general, the parameters of negative branch vary slightly from one comet to another. We found a close correlation between the maximum polarization degree Pmax and the slope of the segment of polarimetric curve bounded by phase angles of 30° and 50°. This finding allowed to adduce the evidence in support of the idea voiced by Chernova et al. (1993) about two types of comet with high and low Pmax. Moreover, we have found direct correlation between the maximum polarization degree of comets and their dust-to-gas ratio. The latter is actually a visual proof of assumptions voiced many years ago about a mutual effect of gas and dust on observed polarization of comets (see, for instance, a historical review in Kiselev et al., 2015). Thus, the polarimetric effect of resonant fluorescence should be completely eliminated from the phase-dependent polarization curve of comet in order to correctly interpret the physical properties of cometary dust. The simple method that we suggest to deconvolve polarimetric response from comets into “pure dusty” and “gaseous” polarimetric curves meets specific difficulties associated mainly with a shortage of data about the polarized luminescence of molecules and ions commonly present in the gaseous component of cometary atmospheres.