1 / 29

Aerosol retrievals from AERONET sun/sky radiometers: Overview of

Aerosol retrievals from AERONET sun/sky radiometers: Overview of - inversion principles - aerosol retrieval products - advances and perspectives.

chelsey
Download Presentation

Aerosol retrievals from AERONET sun/sky radiometers: Overview of

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Aerosol retrievals from AERONET sun/sky radiometers: Overview of - inversion principles - aerosol retrieval products -advances and perspectives O. Dubovik1,2, A. Sinuyk2, B.N. Holben2andAERONET team1 - University of Lille, CNRS, France2 - NASA/GSFC, Greebelt, USA The Second International Conference of Aerosol Science and Global Change August, 18-21, 2009, Hangzhou, China

  2. AERONET Inversion Forward Model: Single Scat: ensemble of polydisperse randomly oriented spheroids (mixture of spherical and non-spherical aerosol components) Multiple Scat: (scalar) Nakajima and Tanaka, 1988, or (polarized) Lenouble et al., JQSRT, 2007 t(l), I(l,Q),P(l,Q) Optimized Numerical inversion: - Accounting for uncertainty (F11; -F12/F11 !!!) - Setting a priori constraints aerosol particle sizes, complex refractive index (SSA), Non-spherical fraction

  3. Accounting for multiple scattering effects ASSUMPTIONS in the retrievals: • - cloud-free atmosphere; • horizontal homogeneous atmosphere; • - assumed gaseous absorption and molecular scattering; • vertically homogenous atmosphere (assumed profile of concentration !? ) • bi-directional surface reflectance assumed from MODIS observations • - accounting for polarization effects!?!

  4. AERONET model of aerosol AERONET model of aerosol Dubovik et al., 2006 spherical: Randomly oriented spheroids : (Mishchenko et al., 1997)

  5. Aerosol single particle scattering: ASSUMPTIONS in the retrievals: • EACH AEROSOL PARTICLE • sphere or spheroid (!!!); • homogeneous; • 1.33 ≤ n ≤ 1.6 (1.7- ???) • - 0.0005 (0 - ???) ≤ k ≤ 0.5 • n and k spectrally dependent(but smooth)

  6. Aerosol particle size distribution Trapezoidal approximation (Twomey 1977) • ASSUMPTIONS: • dV/dlnr - volume size distribution of aerosol in total atmospheric column; • size distribution is modeled using 22 triangle size bins (0.05 ≤ R ≤ 15 m); • size distribution is smooth

  7. Mixing of particle shapes retrieved C  + (1-C)  Aspect ratio distr. • ASSUMPTIONS: • - dV/dlnr - volume size distribution is the same for both components; • non-spherical - mixture of randomly oriented polydisperse spheroids; • aspect ratio distribution N(is fixed to the retrieved by Dubovik et al. 2006

  8. Single Scattering using spheroids spheroidkernels data basefor operational modeling !!! Input:wp(Np =11), V(ri) (Ni =22 -26) K- pre-computed kernel matrices: Input: n and k • Basic Model by Mishchenko et al. 1997: • randomly oriented homogeneous spheroids • w(e) - size independent shape distribution Time: < one sec. Accuracy: < 1-3 % Range of applicability: 0.012 ≤ 2pr/l ≤ 625 (41 bins) 0.3 ≤ e ≤ 3.0 (25 bins) 1.3 ≤ n ≤ 1.7 0.0005 ≤ k ≤ 0.5 Output:t(l), w0(l), F11(Q), F12(Q),F22(Q), F33(Q),F34(Q),F44(Q)

  9. INPUT of Forward Model AERONET retrievals are driven by 31 variables : dV/lnr - size distribution (22 values); n() and k() - ref. index (4 +4 values) Cspher(%) - spherical fraction (1 value) Complex Refractive Index at l = 0.44; 0.67; 0.87; 1.02 mm Particle Size Distribution: 0.05 mm ≤ R (22 bins) ≤ 15 mm Desert Dust Maritime Smoke

  10. Inversion • Statistically Optimized Minimization- Fitting • (Dubovik and King, 2000) weighting Lagrange parameters • Measurements: • i=1 - optical thickness • i=2 - sky radiances • their covariances • (should depend on l and Q) • -lognormal error distributions a priori restrictions on norms of derivatives of: i=3 -size distr. variability; i=4 -n spectral variability; i=5 -k spectral variability; i=6 - limiting dV/dlnr for Rmin consistency Indicator

  11. A priori restirctions • A priori restrictions on smoothness • (Dubovik and King, 2000) Strength of constraint Most unsmooth KNOWN size distribution norms of derivatives Meaning : m=1 -constant straight line: V(lnr)= C; m=2 -constant straight line: V(lnr)= B lnr +C; m=3 -parabola: V(lnr)= A(lnr)2 + B lnr +C;

  12. AERONET retrieval products: - V1 - V2 - V3 • Directly retrieved parameters: • dV/dlnR - size distribution; (- dynamic errors ) • -C(t,f,c), Rv(t,f,c), (t,f,c), Reff (t,f,c) - integral parameters of dV/dlnR • n() and k () at 0.44, 0.67, 0.8, 1.02 m; (- dynamic errors ) • -Cspherical- fraction of spherical particles (- dynamic errors ) • Indirectly retrieved/estimated parameters: • popular: •  - at 0.44, 0.67, 0.8, 1.02 m; (- dynamic errors ) • P11()(- dynamic errors ) and <cos()> ; • P12() and P22()- ???(- dynamic errors ) • - FTOA() and FBOA() - down ward spectral fluxes • - FTOA() andFBOA() - upward spectral fluxes • not well-known / under-developed: - S() - lidar backscattering-to-extinction ratio; (- dynamic errors ) -() - lidar depolarization ratio ; (- dynamic errors ) - FTOA and FBOA - down ward broad-band (visible) fluxes; - FTOAandFBOA- upward broad-band (visible) fluxes; - ∆FTOA and ∆FBOA - radiative forcing -∆FEffTOAand ∆FEffBOA- radiative forcing efficiency

  13. Fine / Coarse modes parameters: Flexible separation: minimum between: 0.194 and 0.576 m 0.45mm • Integral parameters of dV/dlnR: • t - total; f - fine ; c - coarse • C(t,f,c) - Volume Concentration • Rv(t,f,c) - Mean Radius • (t,f,c) - Standard Deviation • Reff (t,f,c)- Effective Radius

  14. Retrieval accuracy and limitations Accuracy summary Sensitivity tests by Dubovik et al. 2000 bias ∆t = ± 0.01 Effective wide angular coverage t(0.44) ≤ 0.2 0.05 80-100% 0.05-0.07 t(0.44) ≥ 0.5 0.025 50% 0.03 Real Part Imaginary Part SSA Size Distribution: Nonsphericity biases Random errors

  15. Error estimates: Important Error Factors: - Aerosol Loading - Scattering Angle Range - Number of Angles (homogeneity) - Number of spectral channels - Aerosol Type etc. New strategy: Errors are to be provided in each single retrievals for all retrieved parameters

  16. Dubovik 2004 Rigorous ERRORS estimates:General case: large number of unknowns and redundant measurements U - matrix of partial derivatives in the vicinity of solution Above is valid: - in linear approximation - for Normal Noise - strongly dependent on a priori constraints - very challenging in most interesting cases

  17. Size distribution Input ERRORS and biases Random (normally distributed with 0 means): - optical thickness: s = 0.015  COS(SZA) - sky-radiances: ssky = 3% - a priori: ssky/ i  100 - 300 % (Dubovik and King, 2000) Biases (constant): - optical thickness:  0.015  COS(SZA) - sky-radiances:  3% + obtained misfit - a priori: 100 - 300 % The error estimates are calculated twice with + and - bias.

  18. Examples of error estimates high loading low loading

  19. Dubovik 2004 ERRORS estimates for thefunctions of the retrieved parameters:0, Pii(), etc. - vector of partial derivatives in the vicinity of solution Above is valid: - in linear approximation - for Normal Noise - strongly dependent on a priori constraints

  20. Statistical variability of SSA errors A. Sinyuk The Second International Conference of Aerosol Science and Global Change August, 18-21, 2009, Hangzhou, China

  21. Statistical variability of errors for sphericity parameter A. Sinyuk The Second International Conference of Aerosol Science and Global Change August, 18-21, 2009, Hangzhou, China

  22. Lidar Ratio CALIOP Data: Lidars are sensitive to: Extinction S=19 S=50

  23. Optics Microphysics Volten et al. Volten et al. 2001

  24. Lidar Ratio from AERONET climatology Cattrall et al., 2005

  25. Size Dependence of Depolarization for Randomly Oriented Spheroids Log-normal monomodal dV(r)/dlnr : sv = 0.5, l = 0.44 mm, n = 1.4, k = 0.005 F22(Q)/ F11(Q) Lidar signal depolarization F22/ F11

  26. AERONET estimated broad-band fluxes in solar spectrum FTOA and FBOA FTOA and FBOA • Integrations details: •  min = 0.2m, max = 4.0 m; • more than 200 points of integration between; Aerosol: • dV/dlnR - retrieved • n() and k() are interpolated/extrapolated; • from n(i) and k(i) retrieved; • Radiative transfer code uses 12 moments for P11() • Surface: • Surface reflection is Lambertian; • Values of surface refelctance are interpolated/ extrapolated from MODIS data values Gases: • Gaseous absorption is calculated using correlated k-distributions implemented by P. Dubuisson Validation studies: Derimian et al. 2008 Garcia et al. 2008 ( FBOA ~ 10% agreement ) Size distribution

  27. Aerosol AERONET estimated aerosol forcing in solar spectrum size Radiative forcing: ∆FTOA = F0TOA - FTOA ∆FBOA = F0BOA - FBOA Radiative forcing efficiency: ∆FEffTOA = ∆FTOA/0.55 ∆FEffBOA = ∆FBOA/0.55 Sol. Radiance,mWcm-2str-1mm-1 Terr. Radiance,mWcm-2str-1mm-1 • Finding by Derimian et al. 2008: • importance of non-sphericity: up to 10% overestimation of ∆FTOA/BOA; Suggested improvementsby Derimian and others: • Use net fluxes: ∆FBOA = (F0BOA-F0BOA) - (FBOA-FBOA) • Estimate daily forcing • Estimates of IR fluxes/forcing Size distribution

  28. Shuster, et al. 2005, 2009 ? n() k() Water+ Soluble+ Insoluble+ +BC m()= m( a1 m1(); a2 m2(); a3 m3())

  29. Improving retrieval products: - releasing dynamic errors; - polishing Flux and Forcing products (ref: Y. Derimian talk) - providing lidar ratios; - providing depolarizations ratios; 2. Updating scattering model: - including surface roughness for spheroids - expanding ranges of n and k 3. New Inversion developments: - inversion of polarized data (ref: Z. Li talk) - AERONET/MODIS/PARASOL (ref: A. Sinuyk talk) - AERONET/CALIPSO (ref: A. Sinuyk work ) - inversion of daily data, combining with PARASOL (ref: O. Dubovik talk ) - deriving composition information (ref: G. Shuster work) Perspectives: The Second International Conference of Aerosol Science and Global Change August, 18-21, 2009, Hangzhou, China

More Related