A VALIDATION OF BACK TRAJECTORIES OF AIR MASSES BY PRINCIPAL COMPONENT ANALYSIS OF ION CONCENTRATIONS IN-CLOUD WATER

The chemical characteristics of polluted marine, polluted continental, and highly polluted air masses were studied by applying principal com ponent analysis (PCA) to the cloud water data collected during field s tudies at a site located in Mt. Mitchell (2038 m ms1, 35 degrees 44'05 '' N, 82 degrees 17'15 '' W) State Park, NC. The site intercepts air masses arriving from the East (polluted marine), the West (polluted co ntinental), and the Northwest (highly polluted). The PCA was used to s tudy the relationship between the ionic constituents of the cloud wate r and the type of air mass in which the cloud formed. By applying PCA to the cloud water chemistry, a set of highly intercorrelated variable s (ions) was replaced with a set of uncorrelated principal components. Using PCA, we can identify the most significant acids and salts disso lved in the cloud water. For instance, sulfuric acid was found to be m ost influential in clouds formed in highly polluted air masses. Sea sa lt particles were present in the majority of the polluted marine cloud events. Calcium was found to be the most important ion in the pollute d continental cloud events. The type of the air mass was determined by 48-h back trajectory analysis using Hybrid Single-Particle Lagrangian Integrated Trajectories (HYSPLIT) model. The results of the HYSPLIT m odel were confirmed by the PCA of the ionic composition of cloud water collected at the mountain top site. Copyright (C) 1996 Elsevier Scien ce Ltd