GEOPHYSICAL RESEARCH LETTERS, VOL. 22, NO. 8, PAGES 909-912, APRIL 15, 1995 First results from POAM Antarctic Ozone Hole II: The dissipation of the 1993 R. M. Bevilacqua, • K.W. Hoppel, l J. S. Hornstein, l R. L. Lucke, l E. P. Shettle, l T. L. Ainsworth, 2 D. Debrestian, • M.D. Fromm, • S.S. Krigman, • J. Lumpe, • W. Glaccum, 3J. J. Olivero, 4 R. T. Clancy? C. E. Randall, sD. W. Rusch, s E. Chassefibre, 6F. Dalaudier, 6, C. Deniel, 6C. Brogniez, 7 J. Lenoble 7 Abstract. POAM II is a space-borne instrument which uses the solaroccultation technique to measure the verticaldistribu- tion of ozone,aerosols andpolar stratospheric clouds, andother properties of the stratosphere and mesosphere. POAM II was launched aboard the SPOT 3 satellite in time to observe the dissipation of the 1993 Antarcticozonehole. POAM data indi- cates that the Antarctic ozone hole dissipates from the top downward.It also supports the hypothesis that the Antarctic vortex is an effective containment vessel. However the strength of the containment appears to decrease markedly at altitudes below 18 km Introduction Solar occultation experiments haveproved to be very power- ful tools for middle atmospheric research. The second Polar Ozone and Aerosol Measurement (POAM II) visible/near IR photometer is designed to measure the abundances of 0 3, H20, NO2, aerosol extinction, and atmospheric temperature by solar occultation techniques. It was launched aboard the French SPOT 3 satelliteon 26 September, 1993 into a Sun synchronous 98.7 ø inclination orbit. In this orbit, which is similar to that of the NASA Stratospheric Aerosol Measurement (SAM 11) experiment (McCormicket al., 1979), all the occultations occur in the polar regions. Thus, both the measurementcomplement and the orbital coverage make the POAM II instrument ideal for study- ing polar ozonedepletion phenomena. In this paperwe present the first resultsof the POAM 1I instrument, concentrating on ozonemeasurements obtainedduring breakup of the Antarctic ozone holein October-November, 1993. The measurements pre- sented in this Letter should be considered preliminary because the datavalidation study is still in progress. However, the inter- 1 Naval Research Laboratory,Code 7227, Washington, DC 20375-5351 2Computational Physics, Incorporated, Fairfax, VA 22031 3Applied Research Corporation, Landover, MD 20785 4Embry-Riddle Aeronautical University, Department of Physical Sciences, DaytonaBeach, FL 32114 5L.A.S.P., University of Colorado, Boulder,CO 80309-0392 6Service D'Aeronomie, CNRS, BP 3, 91371 Verri•res le Buisson CEDEX, France 7L.O.A., Universit6 desSciences et Technologies de Lille, 59655 Villeneuved'AscqCEDEX, France Copyfight1995 by the American Geophysical Union. Papernumber95GL00535 0094-8534/95/95GL-00535 $03.00 comparisons to dateshow the POAM 1I ozone retrievals to be in generally good agreement withother measurements. Instrument and Data Retrieval The POAM lI instrument uses nine separate narrow-band photometers to measure the different species. Nitrogen dioxide, oxygen, andWater vapor are measured using differential on/off pairs: 448.3 and 442 (NO2),760.8 and 780 (02), 935.5 and920 (H20). Aerosol extinction is mainly derived from five wavelengths: the three"off" channels, andat 353 nm and 1059 nm. The primary ozone channel is at 600 nm. The field of view is limited by a long, narrowslit in the focal plane of the lens which subtends an angleof approximately 0.01 ø in elevation by 1.5 ø in azimuth. For the 833-km altitude of the satellite, the 0.01 ø height of the slit combines with the diffraction and spherical aberration of the lensto givean instantaneous vertical resolution of about 0.7 km at the Earth' s limb. Constituent profiles are retrieved via a three-step process consisting of first calculating the total optical depth for each sci- ence channel, by dividing themeasured signal by theappropriate signal level of the unattenuated sun. The total optical depths are then separatedinto their individual components (aerosol, Rayleigh, ozone, etc.) using an optimal estimation retrieval al- gorithm (Rodgers, 1976)at each altitude. Finally, theindividual- component optical depths, as a function of tangent altitude, are inverted to yield the vertical profiles of the absorbing species or aerosol extinction. The detailsof the retrieval algorithm will be described in future work. Results Figure1 compares a POAM II ozone retrieval obtained for a sunrise event over Archangel, Russia on 29 Nov. 1993 with a nearlycoincident (224 km separation at 20 km tangent height) balloon-borne ECC (electrochemical concentration cell) ozone- sonde measurement obtained by J. Rosen(private communica- tion, 1994) aspart of thePOAM 1I validation effort.The vertical resolution (25 m) of the sonde has been degraded to V2km, to makeit morecomparable to the 0.7 km resolution of POAM 1I. A detailed retrieval erroranalysis has notyet been completed for the POAM II ozone measurements. However, the total error is estimated to be about 10 to (below 20 km) 15%, with the largest part of this error resulting from inaccuracies in the determina- tion of instrument pointing and resultant altituderegistration. The comparison between the POAM 1I andRosen measurements is quitereasonable, with differences generally lessthan 10%. We havemade a number of othercomparisons with balloon and satellite ozone data with generally quite good agreement. As mentioned previously, a validation studyis in progress and the resultswill be reported in the near future. The intent here is 909