Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions Spectral and Radiometric Issues for Level 1C L. Larrabee Strow and Scott Hannon Atmospheric Spectroscopy Laboratory (ASL) Physics Department and Joint Center for Earth Systems Technology University of Maryland Baltimore County (UMBC) AIRS Science Team Meeting Nov. 3-5, 2010, Greenbelt, MD ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions Overview Science using AIRS is pushing below the 0.03K level! Is the instrument and the AIRS Radiative Transfer Algorithm up to this? Examine AIRS biases relative to ECMWF versus viewing angle. Absolute errors remain uncertain, but may highlight other errors. Assimilation community has been doing this forever... ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions 861 cm − 1 Window Channel Bias vs Viewing Angle Left: Versus scan angle, Right: Versus secant of viewing angle 861.5 cm − 1 Channel − 0.2 − 0.22 − 0.24 AIRS AIRS IASI IASI − 0.25 − 0.26 − 0.28 − 0.3 − 0.3 Bias in K Bias in K − 0.32 − 0.34 − 0.35 − 0.36 − 0.38 − 0.4 − 0.4 − 0.42 − 0.45 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 − 50 0 50 Scan Angle or Equivalent (degrees) Viewing Secant Window channel, don’t expect bias linear in secant. Note asymmetric behavior of AIRS. IASI flat bias until very high angles. AIRS asymmetry related to polarization? ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions 732 cm − 1 Opaque Channel Bias vs Viewing Angle Left: Versus scan angle, Right: Versus secant of viewing angle ~731.9 cm − 1 732 cm − 1 Region Channels 0.12 0.12 AIRS − ch1 AIRS − ch2 0.1 AIRS − ch1 0.1 IASI AIRS − ch2 0.08 0.08 IASI 0.06 0.06 0.04 0.04 Bias in K Bias in K 0.02 0.02 0 0 − 0.02 − 0.02 − 0.04 − 0.04 − 0.06 − 0.06 − 0.08 − 0.08 − 50 0 50 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 Scan Angle or Equivalent (degrees) Viewing Secant AIRS ch1 and ch2 are on different sides of a spectral line AIRS has both symmetric and anti-symmetric behavior IASI almost symmetric AIRS asymmetry is xtrack frequency dependence ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions Fitting Function for Bias Bias ( xtrack ) = a 1 + a 2 × ( secant − 1 ) + a 3 × ( secant − 1 ) 2 + a 4 × scan _ angle where the Bias is relative to ECMWF for a large statistical set of clear ocean spectra (0-25 Deg. North), secant is the secant of the satellite zenith angle, and scan_angle is the AIRS scan mirror angular position. a 1 is the nadir bias (scan_angle == 0). 1 a 2 , 3 account for viewing angle bias, mostly spectroscopy, 2 symmetric a 4 introduces a linear asymmetry term to the bias 3 a 1 − a 2 should only contain instrument and profile errors, no spectroscopy errors (for opaque channels). Note: a 3 is small. ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions a 4 Term: Linear asymmetric bias term 0.3 0.25 0.2 Max Asymmetry (K) 0.15 0.1 0.05 0 − 0.05 − 0.1 − 0.15 1000 1500 2000 2500 Wavenumber (cm − 1 ) ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions a 4 Term: Linear asymmetric bias term 0.3 0.25 Polarization Asymmetries? 0.2 Max Asymmetry (K) 0.15 0.1 0.05 0 − 0.05 Mostly Frequency − 0.1 Shifts − 0.15 1000 1500 2000 2500 Wavenumber (cm − 1 ) ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions a 4 Term, Now include IASI 0.3 AIRS 0.25 IASI 0.2 Max Asymmetry (K) 0.15 0.1 0.05 0 − 0.05 − 0.1 − 0.15 1000 1500 2000 2500 Wavenumber (cm − 1 ) ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions a 4 Term, Now include IASI, ZOOM 0.3 AIRS 0.25 IASI 0.2 Max Asymmetry (K) 0.15 0.1 0.05 0 − 0.05 − 0.1 − 0.15 700 800 900 1000 1100 1200 1300 1400 1500 1600 Wavenumber (cm − 1 ) ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions What is causing these asymmetric biases? Clue: Not seen in IASI! Larger biases in window regions probably related to polarization. AIRS has aluminum overcoat mirror, IASI mirror is gold Closer examination of opaque channel biases reveals that this “hash” is a frequency shift! Evan Manning quickly postulated this was a Doppler shift, earth’s rotation relative to AIRS changes sign at nadir. IASI does not exhibit the Doppler shift due to METOP “yaw steering” and image motion compensation? All previous frequency calibration measurements used granule averages! ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions Re-examination Frequency Calibration Now separate calibration by xtrack position. Orbit Phase 350 0.3 300 0.2 Freqency Calibration (um) 250 0.1 200 0 150 � 0.1 100 � 0.2 50 � 0.3 0 00:00 06:00 12:00 18:00 00:00 Time This is a one-day frequency calibration with orbit phase encoded by color. Effect largest at equator (Orbit phase == 0, 180, 360). ASL Consistent with Doppler effect.
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions Another View: Frequency Calibration vs Orbit Phase 0.4 North Pole Equator 0.3 Freqency Calibration (um) 0.2 0.1 0 � 0.1 South Pole Equator � 0.2 � 0.3 � 0.4 0 50 100 150 200 250 300 350 Orbit Phase ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions Examine other Bias Coefficients: a 2 Linear secant term: Spectroscopy Errors. 1 AIRS 0.8 IASI 0.6 Linear in Secant Bias (K) 0.4 0.2 0 − 0.2 − 0.4 − 0.6 − 0.8 − 1 1000 1500 2000 2500 Wavenumber (cm − 1 ) ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions Examine other Bias Coefficients: a 2 , ZOOM 0.1 AIRS 0.05 IASI 0 Linear in Secant Bias (K) − 0.05 − 0.1 − 0.15 − 0.2 − 0.25 − 0.3 − 0.35 2390 2395 2400 2405 2410 Wavenumber (cm − 1 ) ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions a 1 − a 2 Removes Spectroscopy, Asymmetric Errors This term contains instrument and ECMWF profile errors. Bias with Spectroscopy Errors and Asymmetry Removed 1.5 1 Nadir Bias − Secant Dependence (K) 0.5 0 − 0.5 − 1 − 1.5 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 ASL Wavenumber (cm − 1 )
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions a 1 − a 2 , now with IASI IASI and AIRS similar, implies profile errors. Bias with Spectroscopy Errors and Asymmetry Removed 1 AIRS IASI Nadir Bias − Secant Dependence (K) 0.5 0 − 0.5 − 1 − 1.5 700 800 900 1000 1100 1200 1300 1400 1500 1600 Wavenumber (cm − 1 ) ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions a 1 − a 2 , Zoom in Window Region A/B Calibration variation shows up, 850-900 cm − 1 . Bias with Spectroscopy Errors and Asymmetry Removed 0.5 AIRS 0.4 IASI Nadir Bias − Secant Dependence (K) 0.3 0.2 0.1 0 − 0.1 − 0.2 − 0.3 − 0.4 800 850 900 950 1000 Wavenumber (cm − 1 ) ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions a 1 − a 2 , Zoom in 4.3 µ m R-branch Very different form than secant error. More consistent with profile error. Bias with Spectroscopy Errors and Asymmetry Removed 1.5 AIRS IASI Fringing Fixed in IASI Nadir Bias − Secant Dependence (K) Day2 Processing 1 0.5 0 − 0.5 2385 2390 2395 2400 2405 2410 2415 Wavenumber (cm − 1 ) ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions IASI Secant Bias over NH Land Note HNO 3 and dip near 790 cm − 1 . Secant Bias 0.4 HNO 3 0.3 0.2 Secant Bias (K/sec) 0.1 0 � 0.1 � 0.2 � 0.3 � 0.4 780 800 820 840 860 880 900 920 Wavenumber (cm � 1 ) ASL
Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions Conclusions AIRS radiative transfer algorithm (RTA) has secant angle biases of up to 0.6K. Needs to be fixed. AIRS has instrumental asymmetric cross-track biases of up to ∼ 0.1-0.2K. Probably polarization? Doppler effect also contributes to biases, this one should be easy to fix. Examination of the secant dependence of the AIRS biases relative to ECMWF has proven very fruitful. Biases that we can attribute to AIRS or the RTA are significant relative to the more demanding AIRS applications (CO 2 , for example). ASL
Recommend
More recommend
