Dr. Jeff Snider, UWyo
Improved understanding of the aerosol indirect effect is resulting from the collaborative efforts of experimentalists and climate modelers. One aspect of this interaction is efforts aimed at using sub-grid parameterizations of updraft velocity (w) to estimate cloud droplet concentration (Nd). Such calculations require a description of the aerosol (i.e., size distribution and composition) that is unavailable in most global climate models, but under consideration as an improvement to future algorithms. The intent of this work is to attempt to validate Nd predictions which are based on measurements of w and aerosol physicochemical properties. We compare these predictions to in-situ measurements of Nd. On a related issue we also examine the observed and predicted response of sub-cloud aerosol size to relative humidity.
The Second Aerosol Characterization Experiment (ACE-2) data set was used for this analysis. For both closure studies we used aerosol physicochemical measurements made at the Punto del Hidalgo (PDH) site, located on the north coast of Tenerife, and concurrent data obtained from the Meteo France Merlin. The PDH data was used to initialize a simulation of the vertical evolution of the wet aerosol size distribution, first as a function of relative humidity in the sub-cloud environment, and second as a function of updraft velocity to yield predictions of Nd. One closure study consisted of comparisons of the predicted wet aerosol size distribution to sub-cloud measurements made onboard the Merlin with a FSSP-300 (D > 0.4 um). The other closure study involved comparisons between the predictions of Nd and in-cloud measurements obtained from the Merlin Fast-FSSP (D > 2 um).
These comparisons show that wet aerosol medium size and the cloud droplet concentration values, both derived from first principles using the PDH data set, are substantially larger than the airborne measurements. This discrepancy could imply that the PDH measurements are not representative of the environment sampled by the Merlin, 100 km north of the PDH, that there are large systematic errors in the independent data sets used to make the comparisons, or that there are deficiencies in the model. All three possibilities will be explored.