330 likes | 485 Views
Validation of the forward/inverse physical scheme -IASI with NAST-I and IMG data. REFERENCE FRAME. IASI project: Infrared Atmospheric Sounding Interferometer Fourier Transform spectrometer with a sampling rate of 0.25 cm -1 spectral coverage: 640 to 2760 cm -1
E N D
Validation of the forward/inverse physical scheme -IASI with NAST-I and IMG data.
REFERENCE FRAME • IASI project: Infrared Atmospheric Sounding Interferometer • Fourier Transform spectrometer with a sampling rate of 0.25 cm-1 • spectral coverage: 640 to 2760 cm-1 • METOP/1, launch late 2005 (ESA/EUMETSAT EPS programme).
IMG: Interferometric Monitoring of Greenhouse Gases (0.05 cm-1 sampling rate):ADEOS/1 platform
CAMEX 3 (Convection and Moisture Experiment 3) • Data acquired during the Atlantic basin tropical cyclone field validation NASA ER-2 flight (during local nighttime, September 13-14, 1998) over Andros Island, Bahamas. • EAQUATE (European AQUA Thermodynamic Experiment) • Data acquired during the Italian field validation Proteus flight (3-11 September, 2004) • NAST-I (NPOESS Airborne Sounder Testbed, infrared interferometer) • Spectral sampling of 0.24 cm-1 • Spectral coverage witch matches that of IASI (645 to 2760 cm-1).
Forward/Inverse ToolsThe φ-IASI package • -IASI: forward model • -IASI: physical inverse scheme • 2-IASI: neural network inversion scheme • -IASI: EOF based regression scheme • -IASI: Cloud Detection Scheme • Masiello et al. JQSRT, Vol 77/2, 131-148, 2003 • Masiello, Serio, Cuomo, Appl. Opt. 43, 2004
Application to NAST-I observations from the CAMEX/3 experiment
σ-IASI/Forward Module • σ-IASI is a line-by-line radiative transfer model designed for fast computation of spectral radiance and its derivatives (Jacobian) with respect to a given set of geophysical parameters. • The model can take into account for both clear and cloudy atmosphere • It is Based on look-up table of monochromatic optical depth + an interpolation procedure.
OD Generation • With LBLRTM 8.1 (HITRAN 2k spectral database), • Water vapour self-broadening is parameterized through a linear approximation.
σ-IASI Consistency Check based on NAST-I scan #298 (Nadir)
Eaquate Campaign NAST-I spectrum recorded over a plant canopy
Transform the inverse solution to PC space trough SVD of the resolving kernel:(details in:Carissimo et al, 2005, EMS, Vol. 20, 1111-1126and EUMETSAT Tech. Rep. EUM/CO/04/1285/RST, Final report, 2004)
Physical scheme:which initialization? Climatology EOF Regression Neural Net
Retrieval Exercise based on NAST-I scan #298 spectrum FG: from EOF Regression; Truth: RAOB2
Retrieval Exercise based on NAST-I scan #298 spectrum FG: from EOF Regression; Truth: RAOB2
Physical retrieval, Temperature(Initialization by EOF Regression) 0° scan angle
Physical retrieval(Initialization by EOF Regression) 0° scan angle
Physical retrieval Performance –computed on the whole set of 1137 clear soundings--T and H2O, Initialization by EOF Regression) Note the truth has been assumed to be the mean RAOB profile
Ozone Retrieval from IMG tropical soundings (details in: Grieco et al JQSRT, Volume 95, Issue 3 (2005) 331-348)
Example of N2O retrieval from an IMG sounding(details in: Lubrano et al Tellus, Volume 56B (2004) 249-261)
Main Conclusions • State-of-art radiative transfer can provide syntethic radiance spectra as good as we need to meet Temperature and H2O IASI mission objective. • The Physical inverse scheme improves T and H2O retrieval independently of the initialization scheme. • The Physical approach is needed to meet IASI mission objective (especially for water vapour)
Next Step:Real-Time Physical Inversion for (T, H2O) • GEOPHYSICAL RESEARCH LETTERS, VOL. 31, L11105, doi:10.1029/2004GL019845, 2004 • Dimensionality-reduction approach to the thermal radiative transfer equation inverse problem • G. Masiello • Istituto di Metodologie per le Analisi Ambientali del CNR, Tito Scalo (PZ), Italy • C. Serio • Istituto Nazionale per la Fisica della Materia, Gruppo Collegato di Potenza, Unitá di Napoli, Potenza, Italy