Droplet spectra broadening in cumulus clouds. Part II : Microscale droplet concentration heterogeneities

Cloud samples of narrow spectra observed in adiabatic cores of cumulus clouds are selected and the droplet spatial distribution is examined in order to document microscale heterogeneities of the droplet concentration. Counting statistics and the Fishing test are applied and compared to the properties of a random spatial distribution, that is, to Poisson statistics. These tests suggest that microscale heterogeneities of the concentration are not significant in adiabatic cores. A conceptual model is then developed to estimate the lifetime that would be required for the observed heterogeneities to generate the observed spectra broadening. The model implies that the lifetime of the heterogeneities should reach unrealistic values to significantly contribute to the observed broadening. The statistical properties of direct numerical simulations of the droplet inertial coupling with turbulence are then compared to the properties of the actual samples. It appears that the properties of the actual samples are closer to the Poisson reference than those of the fields generated by the models. It is concluded that the microscale observations of the droplet spatial distribution in adiabatic cores do not support the hypothesis that inertial coupling is a significant source of spectra broadening.