GEOPHYSICAL RESEARCH LETTERS, VOL. 22, NO. 21, PAGES 2933-2936, NOVEMBER 1, 1995 Aerosolsurfaceareas deduced from early 1993 SAGE II data and comparisons with stratospheric photochemistry, aerosols, and Dynamics Expedition measurements G. K. Yue, 1 L. W. Thomason, andJ. E. Dye 1 L. R. Poole, 1 P.-H. Wang, D. Baumgardner 3 Abstract. Stratospheric Aerosol and Gas Experiment II (SAGE II) multi-wavelength stratospheric aerosolextinc- tion measurements are used to estimate near-global dis- tributions of aerosol surface area density for early months of 1993. A comparison of monthly contour plots shows that the aerosol surface area density above the tropopause gradually decreased from about 15 /•m2cm -a in January 1993 to about 10 /•m2cm -3 in May 1993. Aerosol surface area density profiles measured by the Forward Scattering Spectrometer Probe (FSSP-300) during the Strato- spheric Photochemistry, Aerosols and Dynamics Expedi- tion (SPADE) campaign in May 1993 are compared with estimated SAGE II surface area density profiles obtained at nearby locations. FSSP measurements are in general, slightly lower than the SAGE II measurements. The possi- ble sources of this discrepancy are discussed.When 4-day averageFSSP measurements are comparedwith the SAGE II zonal mean profile for May 1993 and 35øN-40øN at al- titudes above 16 km, the agreement is good. In addition, SAGE II aerosol surfacearea density profiles calculated by assuming a lognormal size distribution and by using prin- cipal componentanalysisare compared,good agreementis shown at high altitudes. However, at low altitudes their differences may be as high as 25%. Introduction Holmann andSolomon [1989] proposed that heteroge- neous reactions on the surfaces of stratospheric aerosol particles can play a role in the partitioning of reactive nitrogen species and the depletion of ozone.They sug- gestedthat the negative ozone anomaliesobserved in the year after the 1982 E1 Chichon eruption could be the result of chemical reactions on the surface of sulfate aerosols, although Chandra and Stolarski [1991] have argued that most of the observed anomalies could be ex- plained by the quasibiennial oscillation (QBO) in ozone. The eruption of Mr. Pinatubo on June 15, 1991, is re- garded as the largest eruption in decades [McCormick and Veiga,1992]. The more than one order of mag- nitude increase in surface area available for chemical 1Atmospheric Sciences Division,NASA LangleyResearch Center, Hampton, Va 23681 2Science and Technology Corp.,Hampton, Va 23666 3National Center for Atmospheric Research,Boulder. Co 80307 Copyright 1995 bythe American Gex)physical Union. Paper numb• 95GL02941 0094-8534/95/95GL-02941 $03.00 reactions produced by this eruption provides a good opportunity to study the influence of aerosolparticles on the dynamicsand chemistryof the stratosphere. The multiwavelength aerosol extinction measurements by the StratosphericAerosol and Gas Experiment II (SAGE II) have beenused previously to deduce aerosol sizeproperties [Yue et at., 1986; Wanget at., 1989]. Estimated aerosol surface area density for the years 1991 and 1992 have been used by several investiga- tors to study the influence of Pinatubo volcanicaerosols [Rodriguez et at., 1994]. One of the main objectives of the present paper is to provideestimatedglobal aerosol surfacearea densities for the early months of 1993 for use in photochemicalmodels to estimate the influence of the decaying Pinatubo aerosol layer on stratospheric chemistry. Sincemany chemicalspecies were measured during the Stratospheric Photochemistry, Aerosols and Dynamics Expedition (SPADE) in April-May 1993,and resultsfrom a photochemical model has been usedto in- terpret and compare ER-2 measurements of hydrogen, nitrogen, and chlorine radicals made during SPADE [Salawitch et at., 1994], the aerosol surface area densi- ties presented in this paper can be a usefulinput source for the model and for the interpretation of SPADE data. Another objective of this paper is to compare surface area density profilesmeasured by the Forward Scatter- ing Spectrometer Probe (FSSP-300)[Baumgardner et at., 1992] on the ER-2 during SPADE with those esti- mated from nearby SAGE II measurements and with the SAGE II monthly zonal mean profiles. Severalmethodshave been suggested for estimating aerosol size distribution and surfacearea density from the multi-wavelengthSAGE II extinction data. In one method [Yueet at., 1986], a lognormal size distribution is assumed and the aerosol number concentration and size parameters in the lognormal size distribution are first deduced.The aerosol surface area densityis subse- quently calculated usingthe inferred lognormalparam- eters. The methodologydeveloped by Thomason, as discussed in Thomason and Osborn [1992] andThoma- sonand Poole [1993], uses principal component anal- ysis to determine a constrained linear combination of the aerosol extinctions at differentwavelengths for esti- mating aerosol size characteristics suchas surface area density. The third objectiveof this paper is to compare aerosol surface area densityprofiles estimatedby these two methods. 2933