Drizzle Suppression in Stratus Clouds due to Cloud Processing of CCN

Bimodal aerosol size distributions first observed by Hoppel et al. (1985) under marine stratus were attributed to sulfur dioxide conversion to sulfate that made the larger sizes of the accumulation mode of cloud residual particles after water evaporated. Particles that nucleate activated cloud droplets thus have altered hygroscopic mass that changes their critical supersaturations, S_c, while S_c of particles that did not nucleate cloud droplets are unchanged. This results in an S_c and size gap after cloud passage. Aerosol bimodality has since been numerously reported. Hudson et al. (2015) showed bimodal CCN spectra and higher cloud droplet concentrations (N_c) in stratus and lower N_c in cumuli of clouds associated with and probably grown upon bimodal CCN compared to clouds grown on unimodal CCN spectra that had not been cloud processed. Here we show that CCN modality (spectral shape; i.e., whether bimodal or unimodal) affects all aspects of stratus cloud microphysics and drizzle. Panel A shows mean differential cloud droplet spectra that have been divided according to traditional slopes, k, of the 131 measured CCN spectra in the Marine Stratus/Stratocumulus Experiment (MASE) off the Central California coast. K is generally high within the supersaturation, S, range of stratus clouds (< 0.5%). High k means greater diversity of CCN concentrations with S. Because cloud processing generally decreases S_c of some particles to create the accumulation mode, it reduces k. Thus lower k indicates more cloud-processed CCN that are more bimodal. Panel A shows higher concentrations of small cloud droplets associated with and probably grown upon lower k CCN than clouds grown on higher k CCN. At small droplet sizes the concentrations follow the k order of the legend, black, red, green, blue (lowest k to highest k). Above 13 µm diameter the lines cross and the hierarchy reverses so that blue (highest k) has the highest concentrations followed by green, then red and finally black (lowest k). This reversal of the droplet hierarchy continues up into the drizzle size range (panel B) where the most drizzle drops, N_d, occur in clouds grown on the least cloud-processed CCN (blue), while clouds grown on the most processed CCN (black) have the lowest N_d. The other major determinant of N_c and other cloud microphysics, is vertical wind, W, which in stratus is actually the variation of W, σ_w (Hudson and Noble 2014). It seemed unfortunate that σ_w was anti-correlated with k during MASE, because both lower k and higher σ_w work toward higher N_c, it was therefore difficult to distinguish these two influences on clouds, i.e., CCN shape effects from dynamic effects. Although higher σ_w ought to make more N_c, panel B shows by far the highest N_c in clouds with the 2^nd highest σ_w values, while clouds of the highest σ_w group have only the 3^rd highest N_c. Panel D shows the same reversal of the concentration hierarchy of panel C that was shown for panels B and A. Further analysis confirmed what is shown in this figure, that the influence of CCN spectral shape overwhelms the effect of σ_w on N_c and N_d; i.e., CCN spectral shape interferes with the hierarchy of cloud and drizzle response to σ_w but σ_w does not interfere with cloud response to CCN spectral shape. There were also greater k differences among the cloud groups shown in the legends than there were σ_w differences among these cloud groups. Adjusted coefficients of determination (R²_adj) of double regressions of σ_w and k were greater than these two single regressions with cloud microphysics (e.g., N_c) and especially with N_d. This means that the two factors, σ_w and k, each independently influenced N_c and N_d. The suppression of stratus cloud drizzle by cloud processing in panel B is an additional 2^nd indirect aerosol effect (IAE) that along with the enhancement of 1^st IAE by higher N_c due to cloud processing (panel A) are above and beyond original IAE due to anthropogenic CCN. However, further similar analysis is needed in other cloud regimes to determine if what was found in MASE is typical. Such further research in other cloud regimes could also determine whether the correlation between σ_w and k was coincidental or inherent; i.e., that CCN spectral shape affects σ_w, e.g., through variations of latent heat exchanges due to condensation or evaporation differences among clouds grown on CCN spectra with different shapes due to various levels of cloud processing.

Hoppel, W.A., J.W. Fitzgerald, and R.E. Larson, 1985: Aerosol size distributions in air masses advecting off the East Coast of the United States. J. Geophys. Res., 90, 2365-2379.

Hudson, J.G., and S. Noble, 2014: CCN and vertical velocity influences on droplet concentrations and supersaturations in clean and polluted stratus clouds. J. Atmos. Sci., 71, 312-331. DOI: 10.1175/JAS-D-13-086.1

Hudson, J.G., S. Noble, and S. Tabor, 2015: Cloud supersaturations from CCN spectra Hoppel minima. J. Geophys. Res., Atmos., 120, 3436–3452, doi:10.1002/2014JD022669.