Cloud optical thickness and liquid water path – does the <i>k</i> coefficient vary with droplet concentration?

Abstract

Abstract. Cloud radiative transfer calculations in general circulation models involve a link between cloud microphysical and optical properties. Indeed, the liquid water content expresses as a function of the mean volume droplet radius, while the light extinction is a function of their mean surface radius. There is a small difference between these two parameters because of the droplet spectrum width. This issue has been addressed by introducing an empirical multiplying correction factor to the droplet concentration. Analysis of in situ sampled data, however, revealed that the correction factor decreases when the concentration increases, hence partially mitigating the aerosol indirect effect.Five field experiments are reanalyzed here, in which standard and upgraded versions of the droplet spectrometer were used to document shallow cumulus and stratocumulus topped boundary layers. They suggest that the standard probe noticeably underestimates the correction factor compared to the upgraded versions. The analysis is further refined to demonstrate that the value of the correction factor derived by averaging values calculated locally along the flight path overestimates the value derived from liquid water path and optical thickness of a cloudy column, and that there is no detectable relationship between the correction factor and the droplet concentration. It is also shown that the droplet concentration dilution by entrainment-mixing after CCN activation is significantly stronger in shallow cumuli than in stratocumulus layers. These various effects are finally combined to produce the today best estimate of the correction factor to use in general circulation models.

Cloud droplet size distribution broadening during diffusional growth: ripening amplified by deactivation and reactivation

“…Results confirm that the CDSD is broadened during diffusional growth due to Ostwald ripening and associated droplet deactivation and reactivation, which is consistent with previous studies (e.g., Korolev, 1995;Çelik and Marwitz, 1999). In this study, we investigate (1) what are the relative roles of the solute and curvature effects on CDSD broadening, and (2) what other factors can affect this broadening?…”

“…Korolev (1995) argued that stronger turbulent fluctuations of supersaturation would result in a broader CDSD. This is contrary to Çelik and Marwitz (1999), who found that supersaturation fluctuations are not responsible for CDSD broadening and the formation of large droplets. The curvature and solute effects on Ostwald ripening, activation and deactivation have been topics of study in recent years (e.g., Wood et al, 2002;Arabas and Shima, 2017;Chen et al, 2018;Sardina et al, 2018) but, to our knowledge, the relative roles of the curvature effect and solute effect on CDSD broadening have not been investigated.…”

“…In recent years, this model has been modified and applied to investigate various microphysical problems (e.g., Feingold and Kreidenweis, 2000;Xue and Feingold, 2004;Ervens and Feingold, 2012;Yang et al, 2012Yang et al, , 2016Li et al, 2013). In the current version of the parcel model, air pressure (p), parcel height (h), air temperature (T ), water vapor mixing ratio (q v ) and radii of haze and cloud droplets (r i ) are prognostic variables, which are calculated using the variable-coefficient ordinary differential equation solver (VODE) (Brown et al, 1989). Specifically, p is calculated from the hydrostatic equation and h depends on the vertical velocity (w).…”

Cloud Shading Effects on Characteristic Boundary-Layer Length Scales

“…Slingo 1990; Vilà-Guerau de Brenguier et al 2011;Lohou and Patton 2014). For a tropical shallow Cu case, as studied here, observational evidence shows that r e = 10 −5 m is an acceptable approximation of the modal droplet radius (McFarlane and Grabowski 2007;Brenguier et al 2011). This surface shading introduces two effects: firstly a domain-averaged decrease in surface fluxes, and secondly a surface heterogeneity because the actual decreases in surface fluxes are local.…”

“…This is a widely used method for calculating τ (e.g. Slingo 1990; Vilà-Guerau de Brenguier et al 2011;Lohou and Patton 2014). For a tropical shallow Cu case, as studied here, observational evidence shows that r e = 10 −5 m is an acceptable approximation of the modal droplet radius (McFarlane and Grabowski 2007;Brenguier et al 2011).…”