The uncertainty in the monomer size is +-0.01um
For an assumed, fixed, phase function of single-scattered intensity as a function of scattering angle, the monomer size can be con-strained to two significant digits from the blue channel observations. However, due to the relatively high optical depth of the haze in the blue channel (approximately 3) even at the beginning of the measurements, the effect of dilution of the singly scattered light by multiple scattering makes the computed polarization uncertain by some 5% for a 20% change in the relative value of the single scattering near 90°. This makes the derived monomer size uncertain by about 0.01 lm. Thus, we find the radius of the haze monomers to be 0.04 ± 0.01 lm.
The shapes of the single-scattering phase functions for intensity
and degree of linear polarization for the models used in the blue channel are shown in Figs. 21 and 22. These curves are computed as described in Tomasko et al. (2008) with the exception that the single scattering matrix element P11 at scattering angles greater than 9° are taken from Henyey–Greenstein functions smoothly grafted to the forward peak derived for the fractal aggregate model. This was done to fit the intensity profiles measured by the Solar Aureole camera at a range of azimuth angles from the Sun. The adjustment is relatively modest, but it results in intensity profiles significantly closer to those measured by the camera. The polariza- tion curves in Fig. 22 use the same single-scattering polarization functions given by the pure fractal expressions in the appendix of Tomasko et al. (2008). The height of the phase function for scattering near 90° com-
pared to the forward peak is constrained by the Solar Aureole cam- era as well as by the upward-looking visible spectrometer (ULVS) of DISR when looking opposite to the Sun relative to the ULVS mea- surements that include the solar beam. We estimate that the uncertainty in the level of the phase function at 90° relative to the forward peak is approximately 20%. Accordingly, we have com- puted polarization models for comparison with the observations in the blue channel with a 20% change in this value. We find that a 20% change in the value of the single-scattering phase function rel- ative to the height of the forward peak leads to a change in the polarization of some 5% in the maximum polarization at each alti- tude. This corresponds to a change in monomer radius of about 0.01 lm.