un outil puissant pour appréhender la physique de l atmosphl

Transcription

1 La haute résolution r spectrale de IASI: un outil puissant pour appréhender la physique de l atmosphl atmosphère Claudia Stubenrauch 1 & participants au projet IASI-physique du TOSCA: R. Armante 1, J.-L. Attlié 2, V. Capelle 1, J.-P. Chaboureau 2, A. Chédin 1, C. Crevoisier 1, T. Deleporte 3, P. Durand 2, C. Flamant 3, A. Guignard 1, Ph. Ricaud 2, N. Scott 1, C. Tsamalis Mar 2012 colloque de restitution TOSCA, CNES 1

2 Sounders: : TOVS, ATOVS, AIRS, IASI (1,2,3), IASI-NG >1980 / 1995 NOAA, 2002 NASA, 2006 CNES-EUMETSAT EUMETSAT measurements within H 2 O, CO 2, O 3 absorption bands & atm. windows long time series -> climate studies retrieval day & night increasing spectral resolution: -> increasing vertical resolution (H 2 O & T profiles) -> decreasing noise (surface temperature, aerosols & cirrus) A-Train synergy (AIRS-CALIPSO-CloudSat): retrieval method validation Surface characteristics: temperature & spectral emissivity Dust aerosols: optical depth,, altitude, eff. particle size of coarse mode Clouds: global climatologies, cirrus microphysics Water vapour & temperature: Ice supersaturation & cirrus GSICS validation activities & radiative transfer: P138 P138 Le projet 4A-OP OP Scott et al. Mar 2012 colloque de restitution TOSCA, CNES 2

3 Retrieval strategies: : surface & aerosols clear sky aerosol cloud -> thermodynamic state of atmosphere select TIGR atmosphere (proximity recognition) by comparing TB s surface characteristics: estimate T surf : select 3 λ s with good transmittance and small variability in ε surf ( 0.97 for 11-12µm) multi-spectral detection determine ε surf spectrum (3.7 14µm; 0.05 µm resolution): first guess: database from 165 laboratory measurements of typical earth surfaces (selected from MODIS, ASTER) physically retrieved from the RTE for 101 IASI channels dust aerosol properties: 12 km spatial resolution 6 IASI channels sensitive to T, H 2 O, sampling of troposphere 8 IASI channels sensitive to aerosol properties simulate Look Up Tables of atmosphere + aerosols depending on aerosol optical depth & altitude by 4A DISORT radiative transfer (SSPs from OPAC) essential for earth radiative budget long-term monitoring of continental surfaces (Sahel) altitude, opt. depth(10µm) & coarse mode eff. particle radius complementary to UV-VIS-NIR -> transport studies Mar 2012 colloque de restitution TOSCA, CNES 3

4 Surface characteristics climatologies: T surf & ε surf (λ) AIRS ( ): Péquignot et al., 2008, IASI( 2007): Capelle et al., latitude x 1 longitude, monthly means, 30 N-30 S Time series Precision: ~1-1.5% at µm and ~4% at µm DESERT at 8.55 µm SAHEL Validation with MODIS seasonal variations depend on surface type ε surf increases with moisture (vegetation) new development with ARIES airborne measurements (MetOffice) at IASI resolution (12 km) 23 ARIES IASI-MSM good agreement with MODIS low resolution ε surf local maximum at 8.65µm for sand as observed in laboratory Flight track 50 km Mar 2012 colloque de restitution TOSCA, CNES

5 Dust climatologies: AIRS & IASI 1 AIRS: Pierangelo et al., 2005; Peyridieu et al., 2010 Tsamalis IASI: Peyridieu et al., in prep.; AOD Maximum in July for both optical depth & mean altitude Decrease of magnitudes with transport away from sources Comparison with MODIS AIRS r~0.8 IASI r~0.9 1 x 1, monthly means altitude Comparison with CALIPSO difference < 500 m Mar 2012 colloque de restitution TOSCA, CNES 5

6 Dust climatologies: IASI new developments 2 Pierangelo et al. GRL 2005; Peyridieu et al., in prep.; Tsamalis eff. particle size of coarse mode slight westward decrease with transport next step: validation with AERONET instantaneous retrievals of AOD MODIS AOD 1:30 PM IASI AOD 9:30 PM First results indicate good correlation Mar 2012 colloque de restitution TOSCA, CNES 6

7 Retrieval Methodology: Cloud properties R m (λ i ) multi-spectral cloud detection µ ε ε(p k, λ i ) coherence N Rm ( ) ( λi ) Rclr ( λi ) pk = R ( p λ ) R ( ) 1, λ i= cld k i clr cloud clearing & T, H 2 O inversion 3I-TOVS NASA-AIRS NOAA-IASI (Scott et al. 1999) (Susskind et al. 2003) (Gambacorta et al.) atmospheric temperature & water vapor profiles, T surf -> thermodynamic state of atmosphere: select TIGR atmosphere (proximity recognition) atm. spectral transmissivities from TIGR + spectral surface emissivities i min of χ w2 (p k ) on spectral cloud emissivites ε cld, p cld (Stubenrauch et al. 1999, 2006, 2008, 2010) cirrus emissivities (8-12 µm) simulated ε(λ,de,iwp) a posteriori cloud detection no assumption on microphysics Mitchell, Baran 4A-DISORT + SSP of ice crystals hex. columns, aggregates De, IWP (CIRAMOSA, Rädel et al. 2003, Stubenrauch et al. 2004, Guignard et al. 2012) Mar 2012 colloque de restitution TOSCA, CNES 7

8 Cloud Assessment co-chairs: C. Stubenrauch, S. Kinne Occurrence of high-level clouds (p cld < 440 hpa) Intercomparison of 12 global global cloud datasets global gridded L3 data (1 x 1 ) : monthly averages, variability, Probability Density Functions lidar, CO 2 sounding IR spectrum IR-VIS imagers solar spectrum global CA 65-70% (+ 5% subvisible Ci) 40 50% of all clouds are high-level clouds uncertainties & biases depend on cloud scene: Stubenrauch et al. WCRP, 2012 CAHR depends on instrument performance to identify thin Ci active lidar > IR sounders > VIS-NIR-IR imagers > multi-angle VIS imagers geographical distributions & seasonal cycles similar vertical sounders : good spectral resolution -> esp. reliable Ci properties (day & night); LMD retrieval better performing than NOAA retrieval (CO 2 slicing) clouds + atmospheric T, H 2 0 profiles (RH) + aerosols July Hilton et al., BAMS 2011 IASI cloud climatology Oct 2008 IASI-LMD AIRS-LMD TOVS-B CALIPSO incl subvis Ci excl subvis Ci ISCCP day Mar 2012 colloque de restitution day + TOSCA, night CNES 8 AIRS IASI

9 A-Train Synergy: evaluation & vertical cloud structure Stubenrauch et al. ACP, 2008, 2010 all clouds, global Evaluation of cloud height LMD --- NASA p cld (AIRS) corresponds to: midlevel of apparent cloud for clouds with diffusive tops: on av. 1.5 km below cloud top A-Train synergy unique opportunity to validate IR sounder retrieval methods at global scale Cloud vertical extent for different cloud types z (thin Ci) < z(ci) < z(hgh op) -> determine climatology of cloud vertical extent per cloud type important input for determination of earth radiation budget Mar 2012 colloque de restitution TOSCA, CNES 9

10 Cirrus mircophysics IR spectra 8-12µm: spectral variation of cirrus emissivities or T B depend on microphysical properties (absorption by ice crystals) 4A/OP-DISORT simulations IASI simulator Ice crystal single scattering properties by Baran (MetOffice) AIRS, cirrus emissivity spectra IASI, T B spectra for cirrus De : cirrus emissivities: Guignard et al. ACP 2012 atmospheric contributions taken out 12µm 10µm Chaboureau slopes increase with decreasing De TB spectra also depend on Best sensitivity for ε IR COD VIS atmosphere and underlying surface Mar 2012 sensitivity: D e colloque <90µm, de IWP<120gm restitution TOSCA, -2 CNES 10

11 Cirrus mircophysics: parameterization & evaluation of models Relationship between De & IWC Guignard et al. ACP 2012 IASI simulator Méso-NH IWC profiles Méso-NH De profiles MetOffice ice crystal SSPs Méso-NH T, H 2 O profiles Climatologies ε surf, O 3 Chaboureau Optical properties 4A/OP DISORT IASI TB spectrum case study Darfour July 2007 De & IWP for cirrus from AIRS cloud vertical extent z from Radar-Lidar GEOPROF IWC = IWP/ z (Mace et al.2009) logarithmic increase of De with IWC similar behaviour in tropics & midlat., summer / winter slopes of Méso-NH larger than observed by IASI validation with DARDAR IWC profiles (smaller De / smaller IWP than in reality?) in progress liquid clouds not yet simulated Mar 2012 colloque de restitution TOSCA, CNES 11

12 IASI water vapour: validation by measurement campaigns Deleporte, Flamant Ricaud COPS: Rhine valley, summer 2007 Antarctica, Dome C HAMSTRAD: H2O Antarctica Microwave Stratospheric & Tropospheric Radiometers total column comparison with GPS IASI retrievals based on optimal estimation similar results (LATMOS & DIFA) positive bias (~3 kg/m 2 ) compared to GPS largest bias between 2-4 km (~1g/kg) DIFA: Dipartimento di Ingegneria e Fisica dell Ambiante, Potenza. Mar 2012 colloque de restitution TOSCA, CNES 12

13 How can we detect ice supersaturation (ISS)? Lamquin 2009; Lamquin et al., ACP 2012 IR Sounders retrieve water vapour within atmospheric layers of km s => underestimation of RH ice : AIRS peak for cirrus at 70% (instead of 100%) improved spectral resolution : IASI peak for cirrus at 80-85% ISS often occurs in vertical layers < 500 m saturated region tropics AIRS AIRS IASI IASI RHi (%), radiosondes determine probability of ISS presence in layer by calibration with MOZAIC (commercial aircraft) Mar 2012 colloque de restitution TOSCA, CNES 13

14 Ice Supersaturation and Cirrus : impact of contrails from CALIPSO hpa AIRS frequency of potential contrail situations from AIRS Ci occurrence increases with ISS occurrence stronger increase in tropics than in midlat (different formation mecanism?) Lamquin 2009; Lamquin et al., ACP 2012 validate IASI clear sky identification determine ice supersaturation probability from IASI study time series TOVS extract meteorological situations for potential contrails Increase of thin cirrus over Europe eff. cloud amount 25 22, , ,5 10 Stubenrauch and Schumann GRL 2005 pot. Contrails Europe clr+thin Ci % ( 1.5%) per decade ~0.19% % per decade (all situations) Mar 2012 colloque de restitution TOSCA, CNES 14 7,5 5 87, , , , , , , ,5 95 time (years)

15 Conclusions & Outlook The high spectral resolution of IASI has already proven its capability to provide 1) physical properties at high spectral resolution & with high accuracy (surface emissivities) 2) new physical properties (dust effective radius of coarse mode) 3) a continuation of time-series of essential climate variables (aerosols, clouds) 4) better vertical resolution (water vapour) However, the retrieval is often complex and validation on a global scale difficult -> measurement campaigns & intercomparison of different retrieval methods The A-Train constellation allowed to validate AIRS retrievals for transfer to IASI data processing possible thanks to Ether, Icare and ClimServ centers continuous update and maintenance of spectral databases (GEISA), radiative transfer code (4A/OP), climatological (TIGR) and radiosonde (ARSA) databases necessary for retrieval Mar 2012 colloque de restitution TOSCA, CNES 15