Atmospheric Chemistry and Physics (May 2012)
Does precipitation susceptibility vary with increasing cloud thickness in marine stratocumulus?
Abstract
The relationship between precipitation rate and accumulation mode aerosol concentration in marine stratocumulus-topped boundary layers is investigated by applying the precipitation susceptibility metric to aircraft data obtained during the VOCALS Regional Experiment. A new method to calculate the precipitation susceptibility that incorporates non-precipitating clouds is introduced. The mean precipitation rate <i>R</i> over a segment of the data is expressed as the product of a drizzle fraction <i>f</i> and a drizzle intensity <i>I</i> (mean rate for drizzling columns). The susceptibility <i>S</i><sub>x</sub> is then defined as the fractional decrease in precipitation variable <i>x</i> = {<i>R</i>, <i>f</i>, <i>I</i>} per fractional increase in the concentration of aerosols with dry diameter >0.1 μm, with cloud thickness <i>h</i> held fixed. The precipitation susceptibility <i>S</i><sub>R</sub> is calculated using data from both precipitating and non-precipitating cloudy columns to quantify how aerosol concentrations affect the mean precipitation rate of all clouds of a given <i>h</i> range and not just the mean precipitation of clouds that are precipitating. <i>S</i><sub>R</sub> systematically decreases with increasing <i>h</i>, and this is largely because <i>S</i><sub>f</sub> decreases with <i>h</i> while <i>S</i><sub>I</sub> is approximately independent of <i>h</i>. In a general sense, <i>S</i><i>f</i> can be thought of as the effect of aerosols on the probability of precipitation, while <i>S</i><sub>I</sub> can be thought of as the effect of aerosols on the intensity of precipitation. Since thicker clouds are likely to precipitate regardless of ambient aerosol concentration, we expect <i>S</i><sub>f</sub> to decrease with increasing <i>h</i>. The results are broadly insensitive to the choice of horizontal averaging scale. Similar susceptibilities are found for both cloud base and near-surface drizzle rates. The analysis is repeated with cloud liquid water path held fixed instead of cloud thickness. Simple power law relationships relating precipitation rate to aerosol concentration or cloud droplet concentration do not capture this observed behavior.
