ANALYSIS OF TROPOSPHERIC OBSERVATIONS FROM GOME AND TOMS
ANALYSIS OF TROPOSPHERIC OBSERVATIONS FROM GOME AND TOMS Randall Martin, Daniel Jacob, Jennifer Logan, Paul Palmer Harvard University Kelly Chance, Thomas Kurosu Harvard-Smithsonian Center for Astrophysics
HOW DO COLUMNS OF TROPOSPHERIC NO 2 FROM GOME COMPARE WITH TRADITIONAL BOTTOM-UP NOx INVENTORIES? GOME/SCIAMACHY Tropospheric NO 2 column ~ ENOx ~ 2 km BOUNDARY LAYER hn (440 nm) NO NO 2 NO/ NO 2 O 3, RO 2 NOx lifetime 1 day Emission HNO 3 Deposition NITROGEN OXIDES (NOx) WITH ALTITUDE
RETRIEVAL OF TROPOSPHERIC NO 2 FROM GOME (errors e in 1015 molecules cm-2) GOME SPECTRUM (423 -451 nm) Fit spectrum e 1 = 0. 6 -0. 8 O 3, O 4, H 2 O, Ring, Undersampling, Common Mode SLANT NO 2 COLUMN Remove stratospheric contribution, diffuser plate artifact e 2 = 0. 4 Use Central Pacific GOME data with: • HALOE to test strat zonal invariance • PEM-Tropics, GEOS-CHEM 3 -D model to treat tropospheric residual TROPOSPHERIC SLANT NO 2 COLUMN Apply AMF to convert slant column to vertical column Quantitative retrieval in partly cloudy scene e 3 = 0. 5 -3. 2 Use radiative transfer model with: • local surface albedos from GOME • local vertical shape factors from GEOS-CHEM global model • local cloud info from GOMECAT TROPOSPHERIC NO 2 COLUMN
GEOS-CHEM MODEL • • Assimilated Meteorology (GEOS) 2 ox 2. 5 o (4 ox 5 o) horizontal resolution, 26 layers in vertical 24 tracers, 120 solved species, ~400 reactions describe tropospheric O 3 -NOxhydrocarbon chemistry Heterogeneous chemistry (with off-line aerosol fields) Photolysis: Fast-J including aerosol scattering Emissions: – Fossil fuel: GEIA (NOx), Logan (CO), Piccot (NMHCs) – Biosphere: modified GEIA (hydrocarbons) & Yienger/Levy (soil NOx) – Lightning: Price/Rind/Pickering, GEOS convective cloud tops – Interannually varying biomass burning (Logan, Duncan et al. 2002) Deposition: modified Wesely (dry), Liu/Mari (wet) Cross-tropopause transport: SYNOZ RECENT AND CURRENT APPLICATIONS: • Tropospheric ozone : global budget, Asian outflow, U. S. air quality, Middle East, transatlantic transport, tropics (TOMS) • Carbon monoxide: budgets, interannual variability • Studies of Aerosols, Carbon dioxide, and Organics • Satellite retrievals, inversions, data assimilation: CO, CO 2, O 3, HCHO, NO 2 • Chemical forecasting: TRACE-P, NOAA 2 K 2
GEOS-CHEM MODEL CAPTURES REGIONAL VARIATION IN NO GEOS-CHEM Aircraft Observations NO 2 number density
RETRIEVAL OF TROPOSPHERIC NO 2 FROM GOME SPECTRUM (423 -451 nm) SLANT NO 2 COLUMN Remove stratospheric contribution, diffuser plate artifact Use Central Pacific GOME data with: • HALOE to test strat zonal invariance • PEM-Tropics, GEOS-CHEM 3 -D model to treat tropospheric residual TROPOSPHERIC SLANT NO 2 COLUMN TROPOSPHERIC NO 2 COLUMN
GEOS-CHEM MODEL IDENTIFIES FAVORABLE REGIONS TO DETERMINE STRATOSPHERIC COLUMN
BIAS THAT WOULD RESULT FROM THE ASSUMPTION OF ZERO TROPOSPHERIC NO 2 OVER THE PACIFIC GEOS-CHEM Aircraft Observations Comparison with PEM-T observations of NO from aircraft suggests small model bias NO 2 number density
TROPOSPHERIC NO 2 COLUMN FROM GOME AFTER REMOVING STRATOSPHERE AND DIFFUSER PLATE ARTIFACT, AND CORRECTING FOR THE PACIFIC BIAS 1996
RETRIEVAL OF TROPOSPHERIC NO 2 FROM GOME SPECTRUM (423 -451 nm) SLANT NO 2 COLUMN TROPOSPHERIC SLANT NO 2 COLUMN Apply AMF to convert slant column to vertical column Quantitative retrieval in partly cloudy scene Use radiative transfer model with: • local surface albedos from GOME • local vertical shape factors from GEOS-CHEM global model • local cloud info from GOMECAT TROPOSPHERIC NO 2 COLUMN
IN SCATTERING ATMOSPHERE, AMF CALCULATION NEEDS EXTERNAL INFO ON SHAPE OF VERTICAL PROFILE RADIATIVE TRANSFER MODEL Io sigma ( ) IB ATMOSPHERIC CHEMISTRY MODEL dt( ) “a-priori” Shape factor EARTH SURFACE NO 2 mixing ratio CNO 2( ) Scattering weight ( ) is temperature dependent cross-section Tabulate w( ) as function of: • solar and viewing zenith angle • surface albedo, pressure • cloud optical depth, pressure INDIVIDUAL GOME SCENES
CLOUDS SIGNIFICANTLY AFFECT SENSITIVITY OF GOME Clear-sky scattering weights Cloudy-sky scattering weights Shape factor
CLOUD REFLECTIVITY (Rc) AND CLOUD FRACTION (f) HAVE A LARGE INFLUENCE ON THE AMF Solar Zenith Angle Cloud Optical Thickness
JULY 1996 Clear-sky AMF Fraction of I From Clouds (GOMECAT and LIDORT) Actual AMF accounting for clouds
VERTICAL COLUMNS LARGELY CONFINED TO REGIONS OF SURFACE EMISSIONS NO/ NO 2 WITH ALTITUDE NOx lifetime ~1 day
GOME RETRIEVAL OF TROPOSPHERIC NO 2 vs. GEOS-CHEM SIMULATION (July 1996) GEIA & Logan emissions scaled to 1996
MODELS AND SATELLITE OBSERVATIONS: THE ODD COUPLE SATELLITE SPECTRA “L 1 DATA” IN SITU OBSERVATIONS (“L 1 DATA”) A PRIORI INFORMATION profile shape, Concentration range, Correlations… RETRIEVAL ATMOSPHERIC CONCENTRATIONS “L 2 DATA” MODELS INCEST? EVALUATION ASSIMILATION INCREASED KNOWELDGE SCIENTIFIC ANALYSIS “L 4 DATA”
DIAGNOSE MODEL CONTAMINATION OF RETRIEVAL BY CORRELATING AMF WITH VERTICAL COLUMN Little relationship between AMF and enhanced NO 2 columns r = -0. 65 r = -0. 14 Negative correlation implies that AMF conversion to vertical columns will modify the slant column patterns to better fit the model
CAN WE USE GOME TO ESTIMATE NOx EMISSIONS? TEST IN U. S. WHERE GOOD A PRIORI EXISTS Comparison of GOME retrieval (July 1996) to GEOS-CHEM model fields using EPA emission inventory for NOx GOME GEOS-CHEM (EPA emissions) GOME BIAS = +18% R = 0. 78
NO 2 COLUMN FROM LIGHTNING SMALL COMPARED TO RETRIEVAL ERROR Tropospheric NO 2 Column Enhancement from Lightning (6 Tg N yr-1) for July (GEOS-CHEM) Error in Tropospheric NO 2 Column Retrieval 7 -33 x 1014 molecules cm-2
We conclude that GOME is consistent with bottom-up NOx emissions inventories, but interesting differences remain … What can we learn from TOMS about the relative roles of biomass burning, lightning, and dynamics in the distribution of tropical tropospheric ozone? ? Lightning ? Nitrogen oxides (NOx) CO, Hydrocarbons ? hn Ozone (O 3) hn, H 2 O Hydroxyl (OH) TROPICAL TROPOSPHERIC OZONE LARGELY DETERMINES OXIDIZING POWER OF ATMOSPHERE Fires Biosphere Human activity Ocean
MOST LIGHTNING ACTIVITY IS OVER LAND
INTENSE BIOMASS BURNING OVER NORTHERN AFRICA DURING DJF
TROPOSPHERIC OZONE COLUMNS (Sep’ 96 -Aug’ 97) GEOS-CHEM TOMS (CCD) DJF MAM JJA SON R = 0. 66 MODEL BIAS = -0. 5 DU
EL NINO INTERANNUAL VARIABILITY IN OZONE: DIFFERENCE BETWEEN OCT 97 AND OCT 96 Chandra et al. [2002]
OZONE ENHANCEMENT FROM LIGHTNING (GEOS-CHEM) largely explains observed wave-1 pattern in TOMS ozone
SIMULATED OZONE CONCENTRATIONS AND FLUXES AT 300 h. Pa IN JAN 97 Incursion of northern hemispheric ozone over the South Atlantic through the “westerly duct” contributes to the wave-1 pattern
OZONE VERTICAL PROFILES OVER ABIDJAN North African Dec-Feb ozone enhancement from biomass burning Is seen by aircraft observations but not by TOMS GEOSCHEM TOMS (distributed w/assumed standard profile) MOZAIC aircraft data
RAYLEIGH SCATTERING LIMITS SENSITIVITY OF TOMS TO SEASONAL VARIATION IN LOWER TROPOSPHERE TOMS sensitivity to ozone (LIDORT radiative transfer model) TOMS standard profiles S. Atlantic profile Abidjan profile
TOMS UNDERESTIMATES OZONE OVER BIOMASS BURNING REGIONS AND OVERESTIMATES OZONE OVER THE PACIFIC
CORRECTION FOR TOMS RETRIEVAL EFFICIENCY CANNOT EXPLAIN DISCREPANCY OVER ABIDJAN What do GOME and SCIAMACHY observe? TOMS (MR) TOMS (CCD) Corrected CCD Tropopause GEOS-CHEM 200 h. Pa MOZAIC (200 h. Pa) Month
SUMMARY WAVE-1 from Lightning Surface NOx from GOME What does GOME show? Accurate a-priori vertical profile and cloud info essential for nadir retrievals
- Slides: 32