A method for the parameterization of cloud optical properites in bulk and bin microphysical models. Implications for arctic cloudy boundary layers

Computationally efficient and numerically accurate methods for computing band-averaged cloud optical properties for radiative transfer interactions with various microphysical parameterizations are described. Parameterizations for bulk microphysical models employing generalized gamma distribution representations of the size spectra and binned representations, in which the size spectra fluctuate with time, are discussed. It is shown that simple exponential fits and look-up tables may be used with minimal computational cost and high accuracy for bulk microphysical models. Binned microphysical representations may be parameterized using mean properties for each bin, if averaged appropriately. The implications for the radiative scheme are discussed in comparison with the computed radiative budget of fall/spring season mixed-phase Arctic stratus clouds (ASC). Compared to liquid clouds of the same water path, mixed-phase ASC absorb and reflect less radiation, and transmit more radiation to the surface. This results in greater cooling (warming) of the surface, by up to 60 W m^-2, in the infrared (solar) by mixed-phase clouds. The radiative properties of mixed-phase clouds show a significant sensitivity to crystal habit for clouds with ice water paths ≱ 25 g m^-2. Surface net fluxes and cloud absorption may vary by up to 15 W m^-2, depending on the ice habit. It is also shown that mixed-phase clouds are more sensitive to the choice of ice effective radius (r_e.i) than liquuid clouds are to r_c. Using values of from the literature, it is shown that the surface net fluxes can vary by as much as 50 W m^-2 depending on the value of r_e.i. Furthermore, it is shown that the sign of the surface net flux (i.e. warming or cooling) may be dependent on the value of r_e.i selected.