RECENT VALIDATION RESULTS FOR THE ATMOSPHERIC CHEMISTRY EXPERIMENT (ACE)

Size: px

Start display at page:

Download "RECENT VALIDATION RESULTS FOR THE ATMOSPHERIC CHEMISTRY EXPERIMENT (ACE)"

William Stokes
5 years ago
Views:

1 RECENT VALIDATION RESULTS FOR THE ATMOSPHERIC CHEMISTRY EXPERIMENT (ACE) Kaley A. Walker (1), Chris Boone (1), Randall Skelton (1), Sean D. McLeod (1), Peter F. Bernath (1), Cora E. Randall (2), Charles R. Trepte (3), Kimberly Strong (4), and C. Thomas McElroy () ABSTRACT (1) Department of Chemistry, University of Waterloo, Waterloo, Ontario, CANADA (2) Laboratory for Atmospheric and Space Physics, University of Colorado, Boulder, Colorado, USA (3) NASA Langley Research Center, Hampton, Virginia, USA. (4) Department of Physics, University of Toronto, Toronto, Ontario, CANADA () Meteorological Service of Canada, Environment Canada, Toronto, Ontario, CANADA The Atmospheric Chemistry Experiment (ACE) is a Canadian scientific satellite mission to perform remote sensing measurements of the Earth's atmosphere. One of the primary goals of ACE mission is to improve our understanding of the chemical and dynamical processes that control the distribution of ozone in the stratosphere and upper troposphere. In order to do this, the accuracy and reliability of the ACE measurements of ozone and other trace gases involved in ozone chemistry (such as NO, NO 2, N 2 O, HN, HCl, and ClONO 2 ) has to be established. This is done by comparisons with measurements from satellite-borne, ground-based, and balloon-borne instruments. This paper outlines recent satellite and ozonesonde comparison results for the ACE-FTS version 2.2 and 2.2 ozone update ozone data products. 1 INTRODUCTION The Atmospheric Chemistry Experiment (ACE), also known as SCISAT-1, is a Canadian scientific satellite to perform remote sensing measurements of the Earth's atmosphere [1]. The satellite was successfully launched by NASA into low Earth orbit on 12 August 23. The 6 km altitude, 74 degree circular orbit provides the mission with global coverage though the focus is on the Arctic and Antarctic regions. The primary goal of the ACE mission is to measure and to understand the chemical and dynamical processes that control the distribution of ozone in the upper troposphere and stratosphere, with a particular emphasis on the Arctic region [2, 3]. The primary ACE instrument is a high-resolution (.2 cm-1) Fourier Transform Spectrometer (ACE-FTS) operating between 7 and 44 cm-1. A passive cooler, radiating to deep space, is used to cool the ACE-FTS detectors. Typical operating temperatures are lower than 9 K. It also houses two filtered imagers that measure atmospheric extinction due to clouds and aerosols at.2 and 1.2 microns. The ACE-FTS was built by ABB-Bomem of Quebec City and the imagers were provided by the Belgian government. The secondary instrument on-board SCISAT-1 is a dual UVvisible-NIR spectrophotometer called ACE-MAESTRO (Measurements of Aerosol Extinction in the Stratosphere and Troposphere Retrieved by Occultation) which extends the wavelength coverage to the nm spectral region. ACE-MAESTRO was built by the University of Toronto, the Meteorological Service of Canada, and EMS Technologies. A feedback controlled sun tracking mirror is used to keep both instruments pointed at the Sun centre during each measurement. The principal measurement technique for the ACE instruments is solar occultation. The ACE-FTS and ACE- MAESTRO have been making spectroscopic measurements of the Earth s atmosphere since February 24. During each sunrise or sunset seen by the satellite, the instrument measures a series of spectra at different ray paths through the atmosphere. These measurements are self calibrating. Transmittance spectra are calculated using exo-atmospheric solar spectra as the reference thereby removing solar and instrumental features. The transmittance spectra from the sunrise and sunset measurements are used to determine atmospheric profiles of trace gases, temperature, pressure and atmospheric extinction [4]. This paper presents recent validation comparisons for the ACE-FTS ozone data product as well as a mission status update for the ACE satellite. Initial ACE-FTS validation comparison studies were done using results from other satellite instruments, including SAGE III [], POAM III [], OSIRIS [6], GOMOS [7], HALOE [8], and EOS-MLS [9]. For these first ACE-FTS ozone retrievals (using version 1. and 2.1), the agreement was typically within 1% with other satellite measurements with the ACE-FTS measuring less ozone near the profile maximum. The ozone retrievals have been further refined and comparisons with this updated version of the ACE-FTS ozone retrieval is presented in this

2 report. These comparisons focus primarily on the March 24 period when the ACE-FTS measurements were coincident with SAGE III and POAM III, and ozonesondes were launched daily from the Eureka, Nunavut weather station. 2 MISSION STATUS In the six months following the launch of SCISAT-1, the commissioning of the spacecraft bus and instruments was completed and a period of science commissioning was undertaken to prepare for operations. Routine satellite operations began in February 24, in time to make measurements of the Arctic polar vortex. Over the 26 months since then, satellite and instrument operations have been nominal. As of mid-april 26, the ACE satellite instruments had made more than 1, occultation measurements. Fig. 1 shows the global distribution of ACE occultations measured from February 24 April 26. Fig. 1. Plot showing the global distribution of ACE occultation measurements from February 24 to April 26. The blue triangles show the location of the occultation measurements and the red points show measurement sites participating in the ACE Validation Program. Operational processing of the ACE-FTS data, to produce atmospheric profiles, was implemented in late summer 24 and over sunset occultations were processed with version 1. of the retrieval code. Volume mixing ratio profiles for 18 species (H 2 O,, CH 4, N 2 O, NO 2, NO, HN, HCl, HF, CO, CFC-11, CFC-12, N 2 O, ClONO2, COF2, SF6, HCFC-22 and HDO) have been produced for sunsets from February October 24 in version 1. of the ACE-FTS retrievals. Following refinement of the processing code, version 2 retrievals now include both sunrise and sunset occultation results and additional species (HCN, CH 3 Cl, CF 4, C 2 H 2, C 2 H 6, ClO) and isotopologues have been added. The current version of the ACE-FTS retrievals is version 2.2 with updates provided for ozone and HDO. Routine processing of ozone and NO 2 profile data from the two ACE-MAESTRO channels (version 1. VIS and NIR) started in early November 24 and the retrieval code was updated in fall 2. The current version of the ACE-MAESTRO retrievals is version 1.1 and it includes profiles for both and NO 2. 3 ACE-FTS OZONE RETRIEVALS During each sunrise or sunset occultation, the ACE-FTS records a series of spectra at different tangent heights. The vertical resolution of these measurements is 3 to 4 km over most of the altitude range. Atmospheric information is retrieved from the spectra using a two step process [4]. First, the pressure and temperature profiles are determined using spectral lines of carbon dioxide. Second, the trace gas volume mixing ratio (VMR) profiles are obtained for each species by holding the temperature profile fixed and fitting narrow microwindow regions (~. cm -1 ) using a global least

3 squares fitting method. The spectroscopic parameters used in the retrievals are taken from the HITRAN 24 linelist [1]. The profiles are interpolated on to a 1km grid using a piecewise quadratic method to produce the final product. Two versions of the ACE-FTS ozone retrievals were used for these comparisons. The version 2.2 retrivals use microwindows in two spectral regions: 32 microwindows between 122 and 1168 cm -1 and 22 microwindows between 183 and 2121 cm -1. While investigating the discrepancies found in comparisons of ACE-FTS profiles with those from other satellite instruments, it was found that there were inconsistent results obtained from the two different microwindow regions (1 microns vs. microns). Therefore, a new set of ozone retrievals was done using a consistent set of 36 microwindows near ~1 microns (98 to 1129 cm -1 ). This second set of ozone profiles is referred to as the version 2.2 ozone update. All of these ozone data products have an altitude range of ~ to 9 km. 1 8 Version 2.2 Ozone Microwindows Altitude (km) Wavenumber (cm -1 ) 1 8 Version 2.2 Ozone Update Microwindows Altitude (km) original 2.2 MW new windows added Wavenumber (cm -1 ) Fig. 2. Comparison of microwindows used in ACE-FTS ozone retrievals for versions 2.2 and 2.2 update. The upper panel shows the version 2.2 microwindows and the lower panel shows the microwindows used for the 2.2 update. 4 COMPARISON DATA SETS The ACE-FTS version 2.2 and 2.2 update ozone profiles have been compared to measurements made by the SAGE III and POAM III satellite instruments and balloon-borne ozonesondes. The measurements are described below. 4.1 Satellite measurements: SAGE III and POAM III The Solar Aerosol and Gas Experiment (SAGE) III measures profiles of, NO 2, H 2 O, N, OClO and aerosols using both solar and lunar occultation [11]. These measurements are made by a grating spectrometer with 87 spectral channels operating between 28 and 4 nm. The version 3. data products are used for these comparisons. The Polar Ozone and Aerosol Measurement (POAM) III instrument uses solar occultation to measure vertical profiles of, NO 2, H 2 O and aerosol extinction [12]. This instrument has nine filtered channels covering the UV, visible and near infrared regions of the spectrum. POAM III version 4. data are used for these comparisons. The satellites carrying both the SAGE III and POAM III instruments are in sun-synchronous orbits therefore the occultations occur in narrow latitude bands in each hemisphere. In the northern hemisphere, the occultations occur at high latitudes and the POAM III measurements are sunrises while those made by SAGE III are sunsets. The retrieved profiles from both instruments have a vertical resolution of approximately 1 km.

4 4.3 Ozonesondes Balloon-borne ozonesondes provide very high vertical resolution ozone profiles from ground level to approximately 3 km. Almost all of the coincidences between ACE-FTS occultation measurements and ozonesonde profiles were for flights launched from the Eureka, Nunavut weather station (79.99 N, 8.94 W). During the 24 Canadian Arctic Validation of ACE campaign at Eureka [13], these instruments were flown each day at 23h UTC from 24 February to 8 March. The launch time was typically within 3 hours of the ACE-FTS occultation measurement. Additional coincidences were found with ozonesondes launched from the Alert, Nunavut weather station (82. N, W). OZONE COMPARISON RESULTS Pairs of coincident measurements for comparisons were chosen using the following criteria: the measurements had to be within km in distance and ±2 hours in time. This time criterion was relaxed for the ozonesonde comparisons to ±12 hours to increase the number of comparison pairs available. Since most of the measurements occurred in March at high latitudes, maps of potential vorticity were used to verify that the coincident measurements were sampling the same region of polar vortex. The SAGE III and POAM III profiles were linearly interpolated to the 1 km grid of the ACE- FTS profiles and no smoothing was applied to these interpolated profiles. Because the ozonesonde profiles have a much higher vertical resolution than the ACE-FTS profiles, these data were smoothed using a weighting function before being linearly interpolated to the ACE-FTS 1 km grid. Weighting functions were constructed for each profile using the measurement tangent altitudes of the coincident ACE-FTS profile (before interpolation) and a nominal 3 km vertical field of view at the limb. Percent difference profiles were calculated for each pair of coincident profiles and then these were averaged to determine average difference profiles. The individual percent differences were calculated with respect to the average of the two profiles (i.e. (ACE-FTS + other)/2). Conversion of ACE-FTS and POAM III profiles from VMR to concentration units was done using atmospheric densities derived from the ACE-FTS temperature and pressure retrievals and UK Met Office data, respectively. Comparisons of the average ozone profiles and average percent difference profiles for ACE-FTS version 2.2 and 2.2 ozone update with SAGE III, POAM III and ozonesondes are shown in Figs. 3, 4 and, respectively. In all three comparisons, the agreement is better with the version 2.2 update than the original version 2.2 retrievals. Particularly, the agreement near the concentration profile maximum has improved significantly. The agreement in this region is now better than %. Between 1 and 4 km, the ACE-FTS ozone results are within 4% of those from SAGE III and within 7% of those from POAM III. ACE-FTS tends to report higher ozone amounts than POAM III above the profile maximum. Between 1 and 3 km, the ACE-FTS ozone measurements are also within ~% of the ozonesonde observations. The divergence seen at higher altitudes (greater than 27 km) may be because the ozonesonde data gets much sparser at these higher altitudes. 6 ACE-FTS v2.2 SAGE III v ACE-FTS v2.2 Ozone Update SAGE III v concentration / 1 12 molecules cm ACE v2.2 - SAGE III v3. / % concentration / 1 12 molecules cm v2.2 Update - SAGE III v3. / % Fig. 3. Comparison of SAGE III data with ACE-FTS version 2.2 ozone results (left) and with version 2.2 ozone update results (right). For each comparison: (Left panel) the average profiles of ozone number density from ACE-FTS and SAGE III measurements between 1 and 18 March profiles are included in each average. The error bars show 1-σ standard deviation of the distribution of the ACE-FTS measurements at each altitude. The average latitude of the measurements was 77.1 N. (Right panel) The average percent difference profile for the ACE-FTS and SAGE III comparisons is shown with 1-σ standard deviation of the mean given by the dotted lines.

5 6 ACE-FTS v2.2 POAM III v ACE-FTS v2.2 Ozone Update POAM III v concentration / 1 12 molecules cm ACE v2.2 - POAM III v4. / % concentration / 1 12 molecules cm v2.2 Update - POAM III v4. / % Fig. 4. Comparison of POAM III data with ACE-FTS version 2.2 ozone results (left) and with version 2.2 ozone update results (right). For each comparison: (Left panel) the average profiles of ozone number density from ACE-FTS and POAM III measurements between 16 and 22 March profiles are included in each average. The error bars show 1-σ standard deviation of the distribution of the ACE-FTS measurements at each altitude. The average latitude of the measurements was 67. N. (Right panel) The average percent difference profile for the ACE-FTS and POAM III comparisons is shown with 1-σ standard deviation of the mean given by the dotted lines ACE-FTS v2.2 Ozonesonde O x VMR (ppv) -1-1 ACE v2.2 - Ozonesonde / % ACE-FTS v2.2 Update Ozonesonde O x VMR (ppv) -1-1 v2.2 Update - Ozonesonde / % Fig.. Comparison of ozonesonde data with ACE-FTS version 2.2 ozone results (left) and with version 2.2 ozone update results (right). For each comparison: (Left panel) the average profiles of ozone VMR from ACE-FTS and ozonesondes measurements between 2 February and 22 September 24 (primarily in February and March with only 3 pairs from September). profiles are included in each average. The error bars show 1-σ standard deviation of the distribution of the ACE-FTS measurements at each altitude. The average latitude of the measurements was 79.2 N. (Right panel) The average percent difference profile for the ACE-FTS and ozonesonde comparisons is shown. 6 SUMMARY All of the comparisons show that the ACE-FTS version 2.2 ozone update retrievals are quite reasonable over the altitude region between and 4 km. When compared to the POAM III, SAGE III and ozonesonde results, ACE-FTS agrees well near the profile maximum to better than %. These comparisons also show the improvement in the ACE-FTS ozone product achieved with the version 2.2 update. This version provides a more consistent result that agrees to within 4-7 % over the altitude range from 1 to 4 km. Investigations are ongoing to investigate the differences between the ACE-FTS and other satellite profiles at higher altitudes.

6 7 ACKNOWLEDGEMENTS Funding for ACE is provided by the CSA, the Natural Sciences and Engineering Research Council of Canada, the MSC, and the Canadian Foundation for Climate and Atmospheric Sciences. Support at Waterloo was also provided by the NSERC-Bomem-CSA-MSC Industrial Research Chair in Fourier Transform Spectroscopy. The authors acknowledge Jonathan Davies of the Meteorological Service of Canada for providing the ozonesonde profiles from Eureka, Nunavut and Alert, Nunavut. 8 REFERENCES 1. Bernath, P.F. et al., J., Atmospheric Chemistry Experiment (ACE): Mission Overview, Geophys. Res. Lett. 32 LS1, doi:1.129/2gl22386, Wardle, D. J., Kerr, J. B., McElroy, C. T., and Francis, D. R. (Eds.), Ozone Science: A Canadian Perspective on the Changing Ozone Layer, Environ. Can., Downsview, Canada, World Meteorological Organization, Scientific assessment of ozone depletion: 22, Global Ozone Res. Monit. Proj. Rep. 47, Geneva, Switzerland, Boone, C.D. et al., Retrievals for the Atmospheric Chemistry Experiment Fourier Transform Spectrometer, Applied Optics 44, 7218, 2.. Walker, K.A., Randall, C., Trepte, C., Boone, C., Bernath, P., Initial Validation Comparisons for the Atmospheric Chemistry Experiment (ACE-FTS), Geophys. Res. Lett. 32, L16S4, doi:1.129/2gl22388, Petelina, S.V. et al., Validation of ACE-FTS stratospheric ozone profiles against Odin/OSIRIS measurements, Geophys. Res. Lett. 32, LS6, doi:1.129/2gl22377, Fussen, D. et al., Intercomparison of ozone and nitrogen dioxide number density profiles retrieved by the ACE and GOMOS occultation experiments, Geophys. Res. Lett. 32, L16S2, doi:1.129/2gl22468, McHugh, M. et al., Comparison of atmospheric retrievals from ACE and HALOE, Geophys. Res. Lett. 32, LS1, doi:1.129/2gl2243, Froidevaux, L. et al., R., Early Validation Analyses of Atmospheric Profiles from EOS MLS on the Aura Satellite, IEEE Trans. Geosci. Remote Sens. (in press). 1. Rothman, L.S. et al., The HITRAN 24 molecular spectroscopic database, J. Quant. Spectrosc. Rad. Transfer 96, 139, Thomason, L.W., and Taha, G., SAGE III aerosol extinction measurements: Initial results, Geophys. Res. Lett., 3, 1631, doi:1.129/23gl Lucke, R.L., et al. The Polar Ozone and Aerosol Measurement (POAM) III instrument and early validation results, J. Geophys. Res., 14, 18,78, Kerzenmacher, T.E. et al., Measurements of O3, NO2 and Temperature During the 24 Canadian Arctic ACE Validation Campaign, Geophys. Res. Lett. 32, L16S7, doi:1.129/2gl2332, 2.

Comparisons between ACE-FTS and ground-based measurements of stratospheric HCl and ClONO 2 loadings at northern latitudes

GEOPHYSICAL RESEARCH LETTERS, VOL. 32, L15S08, doi:10.1029/2005gl022396, 2005 Comparisons between ACE-FTS and ground-based measurements of stratospheric HCl and ClONO 2 loadings at northern latitudes E.