Spectral and Radiometric Issues for Level 1C L. Larrabee Strow and - - PowerPoint PPT Presentation

ASL

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

−50 50 −0.45 −0.4 −0.35 −0.3 −0.25 −0.2 Scan Angle or Equivalent (degrees) Bias in K 861.5 cm−1 Channel AIRS IASI 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 −0.42 −0.4 −0.38 −0.36 −0.34 −0.32 −0.3 −0.28 −0.26 −0.24 −0.22 Viewing Secant Bias in K AIRS IASI

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

−50 50 −0.08 −0.06 −0.04 −0.02 0.02 0.04 0.06 0.08 0.1 0.12 732 cm−1 Region Channels Scan Angle or Equivalent (degrees) Bias in K AIRS−ch1 AIRS−ch2 IASI 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 −0.08 −0.06 −0.04 −0.02 0.02 0.04 0.06 0.08 0.1 0.12 ~731.9 cm−1 Viewing Secant Bias in K AIRS−ch1 AIRS−ch2 IASI

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) = a1+a2×(secant−1)+a3×(secant−1)2+a4×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.

1

a1 is the nadir bias (scan_angle == 0).

2

a2,3 account for viewing angle bias, mostly spectroscopy, symmetric

3

a4 introduces a linear asymmetry term to the bias a1 − a2 should only contain instrument and profile errors, no spectroscopy errors (for opaque channels). Note: a3 is small.

ASL

Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions

a4 Term: Linear asymmetric bias term

1000 1500 2000 2500 −0.15 −0.1 −0.05 0.05 0.1 0.15 0.2 0.25 0.3 Wavenumber (cm−1) Max Asymmetry (K)

ASL

Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions

a4 Term: Linear asymmetric bias term

1000 1500 2000 2500 −0.15 −0.1 −0.05 0.05 0.1 0.15 0.2 0.25 0.3 Wavenumber (cm−1) Max Asymmetry (K) Polarization Asymmetries? Mostly Frequency Shifts

ASL

Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions

a4 Term, Now include IASI

1000 1500 2000 2500 −0.15 −0.1 −0.05 0.05 0.1 0.15 0.2 0.25 0.3 Wavenumber (cm−1) Max Asymmetry (K) AIRS IASI

ASL

Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions

a4 Term, Now include IASI, ZOOM

700 800 900 1000 1100 1200 1300 1400 1500 1600 −0.15 −0.1 −0.05 0.05 0.1 0.15 0.2 0.25 0.3 Wavenumber (cm−1) Max Asymmetry (K) AIRS IASI

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.

00:00 06:00 12:00 18:00 00:00 0.3 0.2 0.1 0.1 0.2 0.3 Time Freqency Calibration (um) 50 100 150 200 250 300 350 Orbit Phase

This is a one-day frequency calibration with orbit phase encoded by color. Effect largest at equator (Orbit phase == 0, 180, 360). Consistent with Doppler effect.

ASL

Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions

Another View: Frequency Calibration vs Orbit Phase

50 100 150 200 250 300 350 0.4 0.3 0.2 0.1 0.1 0.2 0.3 0.4 Orbit Phase Freqency Calibration (um)

Equator South Pole Equator North Pole

ASL

Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions

Examine other Bias Coefficients: a2

Linear secant term: Spectroscopy Errors.

1000 1500 2000 2500 −1 −0.8 −0.6 −0.4 −0.2 0.2 0.4 0.6 0.8 1 Wavenumber (cm−1) Linear in Secant Bias (K) AIRS IASI

ASL

Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions

Examine other Bias Coefficients: a2, ZOOM

2390 2395 2400 2405 2410 −0.35 −0.3 −0.25 −0.2 −0.15 −0.1 −0.05 0.05 0.1 Wavenumber (cm−1) Linear in Secant Bias (K) AIRS IASI

ASL

Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions

a1 − a2 Removes Spectroscopy, Asymmetric Errors

This term contains instrument and ECMWF profile errors.

800 1000 1200 1400 1600 1800 2000 2200 2400 2600 −1.5 −1 −0.5 0.5 1 1.5 Wavenumber (cm−1) Nadir Bias − Secant Dependence (K) Bias with Spectroscopy Errors and Asymmetry Removed

ASL

Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions

a1 − a2, now with IASI

IASI and AIRS similar, implies profile errors.

700 800 900 1000 1100 1200 1300 1400 1500 1600 −1.5 −1 −0.5 0.5 1 Wavenumber (cm−1) Nadir Bias − Secant Dependence (K) Bias with Spectroscopy Errors and Asymmetry Removed AIRS IASI

ASL

Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions

a1 − a2, Zoom in Window Region

A/B Calibration variation shows up, 850-900 cm−1.

800 850 900 950 1000 −0.4 −0.3 −0.2 −0.1 0.1 0.2 0.3 0.4 0.5 Wavenumber (cm−1) Nadir Bias − Secant Dependence (K) Bias with Spectroscopy Errors and Asymmetry Removed AIRS IASI

ASL

Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions

a1 − a2, Zoom in 4.3 µm R-branch

Very different form than secant error. More consistent with profile error.

2385 2390 2395 2400 2405 2410 2415 −0.5 0.5 1 1.5 Wavenumber (cm−1) Nadir Bias − Secant Dependence (K) Bias with Spectroscopy Errors and Asymmetry Removed AIRS IASI IASI Fringing Fixed in Day2 Processing

ASL

Introduction Scan Angle Biases Doppler Effect Biases IASI vs AIRS Biases Conclusions

IASI Secant Bias over NH Land

Note HNO3 and dip near 790 cm−1.

780 800 820 840 860 880 900 920 0.4 0.3 0.2 0.1 0.1 0.2 0.3 0.4 Wavenumber (cm1) Secant Bias (K/sec) Secant Bias HNO3

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 (CO2, for example).