MEASURING TRACE GAS PROFILES FROM SPACE

Transcription

1 MEASURING TRACE GAS PROFILES FROM SPACE Caroline Nowlan Atomic and Molecular Physics Division Harvard-Smithsonian Center for Astrophysics Collaborators: Kelly Chance, Xiong Liu, Gonzalo Gonzalez Abad, Zhaonan Cai, Jay Kar, Kaley Walker, Chris Sioris EPS 238, Harvard University 10 April 2014

2 Profiling the Atmosphere Active LIDAR RADAR Passive Occultation Nadir backscatter Nadir thermal emission (IASI, TES) Limb scattering (OSIRIS, SCIAMACHY) Limb thermal emission (MIPAS)

3 Profiling the Atmosphere Active LIDAR RADAR Passive Occultation Nadir backscatter Nadir thermal emission (IASI, TES) Limb scattering (OSIRIS, SCIAMACHY) Limb thermal emission (MIPAS)

4 Passive Measurement Geometries

5 Nadir Backscatter Instruments Instrument Dates Profiles retrieved from nadir? TOMS GOME O 3 (Liu et al., 2005) SCIAMACHY OMI 2004 O 3 (Liu et al., 2010) SO 2 (Yang et al., 2010) GOME-2A GOME-2B GOME-2C OMPS 2011 TROPOMI 2015 Sentinel TEMPO 2018 GEMS 2018 O 3 (Cai et al., 2012) SO 2 (Nowlan et al., 2011)

6 Remote Sounding: Inverse Problem y = F(x) + ε We have y (spectral data) We want x (profile & other fitted parameters) Requirements for inversion Forward model (F) Retrieval algorithm

7 Nadir Backscatter Global coverage Limited altitude information, and only for certain molecules (Ozone, volcanic SO 2 ) Almost always need a priori information on state of atmosphere (i.e., ozone profile climatology) Altitude information on ozone and SO 2 has so far been derived from OMI, GOME, and GOME-2 using the UV

8 Optimal Estimation Approach Combine a priori knowledge with measurements Iterate until convergence Δx = (K T S ε -1 K + S a -1 ) -1 [K T S ε -1 Δy S a -1 (x x a ) ] x = current guess of retrieved parameter y = measurements S ε = measurement error covariance matrix K = dy/dx x a = a priori profile (climatology or from a model) S a = a priori error covariance matrix

9 K. Chance, CfA

10 GOME-2 Spectra, Channel 2

11 GOME-2 Ozone Profile (Backscatter) x A = ˆ x (Z. Cai, Chinese Academy of Sciences)

12 OMI Tropospheric Ozone X. Liu (CfA)

13 GOME-2 and OMI SO 2 Mt. Kasatochi Alaska 9 August 2008 A priori SO 2 altitude = 10 km Uncertainty = 2 km

14 Occultation Measurements Self-calibrating High vertical resolution Sparse global coverage

15 Occultation Instruments Instrument Dates Spectral Region Source ATMOS (shuttle) infrared Sun HALOE NIR, infrared Sun SAGE I SAGE II SAGE III SAGE III ISS UV, visible, NIR Sun GPS technique 1995 radio GPS ILAS ILAS II NIR, infrared Sun SCIAMACHY UV infrared Sun and moon GOMOS UV, visible, NIR Stars ACE-FTS 2003 infrared Sun ACE-MAESTRO 2003 visible, NIR Sun

16 Atmospheric Chemistry Experiment ACE is on SCISAT satellite Launched August 12, 2003 Two primary instruments on-board ACE-FTS (Fourier Transform Spectrometer)! INFRARED ACE-MAESTRO! VISIBLE-NIR ACE-FTS measures: H 2 O, O 3, N 2 O, NO, NO 2, HNO 3, N 2 O 5, H 2 O 2, HO 2 NO 2, N 2, HCl, HF, ClONO 2, CFC-11, CFC-12, CFC-113, COF 2, COCl 2, COFCl, CF 4, SF 6, CH 3 Cl, CCl 4, HCFC-22, HCFC-141b, HCFC-142b, CO, CH 4, CH 3 OH, H 2 CO, HCOOH, C 2 H 2, C 2 H 4, C 2 H 6, OCS, HCN, ClO, acetone, PAN, aerosols

17 Solar Occultation Measurements E x o a t m o s p h e r i c R a y 3 R a y 2 S u n R a y 1 S C I S A T

18 ACE Occultation Coverage: 2004 SUNRISE SUNSET Kar et al., 2007

19 MAESTRO Optical Depth Spectra 7 6 O 3 O 2 O 2 5 O 2 optical depth H 2 O wavelength (nm)

20 Occultation Retrievals Traditional method: onion peeling Newer approach: global fitting Simultaneous fitting of every spectrum in an occultation Arrange all spectra into one giant measurement vector Δx = (K T S ε -1 K + S a -1 ) -1 [K T S ε -1 Δy S a -1 (x x a ) ] x = current guess of retrieved parameter y = measurements S ε = measurement error covariance matrix K = dy/dx x a = a priori profile (climatology or from a model) S a = a priori error covariance matrix

21 ACE-MAESTRO: Ozone Profiles Kar et al., 2007

22 ACE-MAESTRO: Mt. Kasatochi Aerosols Sioris et al., JGR, 2010

23 Density Averaging Kernels from O 2 Bands A = ˆx x

24 ACE-FTS Spectra

25 ACE-FTS CO 2 line (near 61 km) K. Walker, U of Toronto

26 ACE-FTS: Canadian Biomass Burning Tereszchuk et al., ACP, 2013

27 Halogen-containing Species Trends CFC-11 (CCl 3 F) Tropical troposphere averages (30N-30S) HCFC-22 (CHClF 2 ) ACE-FTS meas. (~7-16 km) SLIMCAT model (~7-16 km) AGAGE meas. (surface) A. Brown et al., JQSRT, 112, (2011)

28 Distribution of COClF Carbonyl chlorofluoride is a product of chlorofluorocarbon (CFC-11 mainly) decomposition Previously studied by aircraft instruments (5-12 km) First global picture obtained from ACE-FTS D. Fu et al., JQSRT, 110, (2009)

29 Exoplanet atmospheres from occultation? Credit: NASA/Tim Pyle

30

31 The End