CCN spectra Hoppel minima and cloud supersaturation

Conventional wisdom stratus cloud supersaturations (S) are < 0.3%.  Bimodal dry particle size distributions often found below maritime stratus are a major reason for the low S estimates.  The fact that minima between these two peaks are usually < 100 nm and the premise that these Hoppel minima (Hoppel et al. 1986) are a result of cloud processing of the aerosol implies low cloud S. However, Hudson et al. (2010; H10) estimated higher stratus cloud S by matching CCN spectra of a Desert Research Institute (DRI) CCN spectrometer with mean cloud droplet concentrations (N_c).  But more recently the DRI instruments have provided another method of estimating cloud S.  The high S resolution and resulting differential critical S (S_c) CCN spectra have revealed bimodal distributions below clouds (Fig. 1).  The minima of these distributions provide more direct estimates of cloud S than particle size distributions because they do not require particle composition assumptions. These newest cloud S estimates show higher values than previous Hoppel minima but not as high as those of H10 (Fig. 2); mean S 0.37% compared to 0.48%.  The discrepancy between these methods may be due to subsequent evaporation of small cloud droplets and/or less participation of small cloud droplets in the processes that decrease CCN S_c and thus cause Hoppel minima. Thus, these chemical (gas-to-particle conversion; e.g., sulfate) and physical (coalescence and Brownian capture) processes may be largely limited to only the larger cloud droplets that had grown on the lower S_c CCN.  Thus, Hoppel minima may provide underestimates of cloud S, but these Hoppel estimates may be more pertinent to aerosol-cloud-climate interactions. 

The DRI CCN spectrometers have not only observed Hoppel minima in stratus (POST; off central California) but also below RICO small cumuli. As expected both estimates of RICO cloud S were higher than POST since vertical wind is higher in cumuli. The same S discrepancy between methods was found in RICO. The S discrepancy and the fact that N_c are well correlated with CCN concentrations (H10; Hudson et al. 2012) suggests a more homogeneous evaporation process so that CCN that experience increased S_c that result in Hoppel minima tend to be only those with initially rather low S_c. The higher S_c CCN produce only small cloud droplets that either do not persist or do not participate in very much cloud processing.  Accurate stratus cloud S estimates are needed to determine the anthropogenic particles that cause the indirect aerosol effect. Higher cloud S implies that smaller more numerous particles are able to produce cloud droplets.  Complications for both methods of estimating cloud S include situations when CCN concentrations are different (especially higher) above the stratus clouds (Fig. 2).        

Hoppel, W.A., G.M. Frick and R.E. Larson, 1986: Effect of nonprecipitating clouds on the aerosol size distribution in the marine boundary-layer. Geophys. Res. Lett., 13(2, 125-128.

Hudson, J.G., S. Noble and V. Jha, 2010:  Stratus cloud supersaturations.  Geophys. Res. Lett., 37, L21813, doi:10.1029/2010GL045197.

Hudson, J.G., S. Noble and V. Jha, 2012:  Cloud droplet spectral width relationship to CCN spectra and vertical velocity.  J. Geophys. Res., Vol. 117, D11211, doi:10.1029/2012JD017546, 2012.