Using Collocated AVHRR Imager Background and Motivation - PowerPoint PPT Presentation

AVHRR IASI E. Maddy et al. Using Collocated AVHRR Imager Background and Motivation Measurements to Constrain Cloud Cleared Collocation Radiances from IASI Cloud Clearing E. S. Maddy 1 , 2 , T. S. King 1 , 2 ,H. Sun 1 , 2 , W. W. Wolf 2 , C.D. Barnet 2 ,A. Heidinger 2 , Z. Cheng 2 , A. Gambacorta 1 , 2 ,K. Zhang 1 , 2 ,C. Zhang 1 , 2 , M. Goldberg 2 1 Dell, 2 NOAA/NESDIS/STAR email: eric.maddy@noaa.gov November 4, 2010 1 / 25

AVHRR IASI IASI Only Cloud-Clearing E. Maddy et al. Background and Motivation Basic Idea Collocation Cloud Clearing MEAN(∆T RET − ECMWF ) = − 1 . 901K %CASES ACCEPTED = 60 . 07% STD . DEV(∆T RET − ECMWF ) = 3 . 027K %CASES ≥ | 3K | ABOUT MEAN = 22 . 78% CORREL(T RET , T ECMWF ) = 0 . 9368 2 / 25

AVHRR IASI Background and Motivation E. Maddy et al. • Smith, et al. (2004) and Li, et al. (2005) showed that collocated AIRS IR sounder and Aqua MODIS imager measurements enable direct calculation Background and of high quality cloud-cleared radiances without the use of a forward model to Motivation Basic Idea estimate clear sky radiances. Collocation • Their methods rely on the high spatial resolution MODIS measurements and cloud-mask to estimate clear-sky measurements Cloud Clearing in MODIS IR spectral bands spatially collocated and averaged onto the AIRS footprints. • The use of IR spectral bands covering the spectral domains sampled by the AIRS instrument enables direct comparison of the clear-sky MODIS measurements to AIRS and therefore does not require a priori assumptions about the geophysical state (i.e., surface properties, trace gas concentrations and/or water vapor abundances ) to enable calculation of clear-sky radiances. • Noise amplification was not considered; however, the reported accuracies of the cloud-cleared radiances for successfully cleared cloudy scenes (30% of all cloudy scenes) were 0 . 5 K or better. In what follows we apply the single formation cloud-clearing equations in the η formulation (enables multiple cloud-formations) to solve for cloud-cleared radiances from IASI and AVHRR. 3 / 25

Top: CLAVR-X AVHRR Ch.4 BTs Sample IASI Spectrum and AVHRR AVHRR IASI (courtesy A. Heidinger) Spectral Response Functions Bottom: AVHRR collocated onto IASI E. Maddy et footprints (courtesy H. Sun) al. Background and Motivation Basic Idea Collocation Cloud Clearing Spectral convolution of IASI to AVHRR Spatial convolution of AVHRR to IASI resolution footprints We want to exploit the high spatial resolution of the multispectral AVHRR data to improve and/or enhance IASI retrievals in two ways: 1. QA cloud-cleared radiances using spectrally convolved IASI and spatially convolved subpixel clear AVHRR to compare apples to apples. 2. Utilize subpixel ( ≈ 1 km AVHRR versus ≈ 12 km IASI) /multispectral (visible/NIR) information about clouds from AVHRR to improve/validate cloud-clearing, improve our ’clear-estimate’, and/or develop syngeristic products thereby enhancing other retrievals. 4 / 25

Top: CLAVR-X Cloud Mask (courtesy A. Sample IASI Spectrum and AVHRR AVHRR IASI Heidinger) Spectral Response Functions Bottom: AVHRR collocated onto IASI E. Maddy et footprints (courtesy H. Sun) al. Background and Motivation Basic Idea Collocation Cloud Clearing Spectral convolution of IASI to AVHRR Spatial convolution of AVHRR to IASI resolution footprints We want to exploit the high spatial resolution of the multispectral AVHRR data to improve and/or enhance IASI retrievals in two ways: 1. QA cloud-cleared radiances using spectrally convolved IASI and spatially convolved subpixel clear AVHRR to compare apples to apples. 2. Utilize subpixel ( ≈ 1 km AVHRR versus ≈ 12 km IASI) /multispectral (visible/NIR) information about clouds from AVHRR to improve/validate cloud-clearing, improve our ’clear-estimate’, and/or develop syngeristic products thereby enhancing other retrievals. 5 / 25

AVHRR IASI Methodology to Test Collocations and E. Maddy et Instrument Synergy al. Background and Motivation • Use a single days worth of data Oct. 3, 2010 (performed analysis on various days in 2010 and found similar results). Collocation Testing and • Collocate AVHRR measurements for channels 4 (928.15cm − 1 ) and 5 Validation (833.25cm − 1 ) to IASI fields-of-view (FOVs) in 2 × 2 array forming the IASI Stats All Stats Clear field-of-regard (FOR). Discussion • Use CLAVR-X cloud-mask to aggregate AVHRR clear (and/or all-sky) pixels Cloud Clearing onto IASI FOVs using the IASI spatial response or integrated point spread function IPSF n clr AV HRR R clr X IPSF l R clr,l A i = (1) A i l =1 • Use the AVHRR SRF (NOAA KLM User’s Guide) to spectrally convolve the IASI radiance to AVHRR spectral resolution. X R A i = SRF i,ν R ν (2) ν 6 / 25

AVHRR IASI All Collocations of AVHRR BTs and IASI BTs for 10/03/2010 E. Maddy et Black: All Cases, Red: Clear Cases al. Background and Motivation Collocation Testing and Validation Stats All Stats Clear Discussion Cloud Clearing Stats for all cases Stats for all cases MEAN(∆BT A − I ) = − 0 . 163 K MEAN(∆BT A − I ) = − 0 . 203 K STD . DEV(∆BT A − I ) = 0 . 422 K STD . DEV(∆BT A − I ) = 0 . 417 K CORREL(BT A , BT I ) = 0 . 9998 CORREL(BT A , BT I ) = 0 . 9998 7 / 25

AVHRR IASI All Collocations of AVHRR BTs and IASI BTs for 10/03/2010 E. Maddy et Black: All Cases, Red: Clear Cases al. Background and Motivation Collocation Testing and Validation Stats All Stats Clear Discussion Cloud Clearing Stats for clear cases Stats for clear cases MEAN(∆BT A − I ) = − 0 . 321 K MEAN(∆BT A − I ) = − 0 . 425 K STD . DEV(∆BT A − I ) = 0 . 109 K STD . DEV(∆BT A − I ) = 0 . 117 K CORREL(BT A , BT I ) = 0 . 9999 CORREL(BT A , BT I ) = 0 . 9999 8 / 25

AVHRR IASI Discussion E. Maddy et • We have found excellent agreement between all-sky and clear sky spectrally al. convolved IASI measurement and spatially convolved AVHRR Background measurements. and • There is a small channel dependent bias between AVHRR-IASI with IASI Motivation generally being warmer than AVHRR. Collocation • Standard deviation between the two instruments is < 0 . 5 K for non-uniform Testing and Validation scenes and ≈ 0 . 1 K for uniform clear scenes! Stats All Stats Clear • Differences between AVHRR and IASI are dependent on the scene Discussion brightness temperature (saw this comparing the bias of clear to all-sky data). Cloud • L. Wang and C. Cao, 2008 found a similar result using a different Clearing collocation methodology and data from 2007. • Slopes are generally small ( ≈ 1 K at the cold end compared to − 0 . 3 K at the warm end). • Differences between instruments also has a small < 0 . 1 K scan angle dependence that needs investigated - also reported by L. Wang and C. Cao. In what follows we’ve performed a bias correction to the AVHRR measurements R A i similar to what is done for AMSU and IASI or AMSU and AIRS : R ′ A i = a 0 + b 0 R A i We’ve not attempted to force any scan angle dependent differences to zero. 9 / 25

AVHRR IASI E. Maddy et al. Background and Motivation Collocation Cloud Clearing So we have highly accurate collocations . . . Methodology Clear FOVs How do use these to produce cloud-cleared radiances R cc ν ? Partially Cloudy FOVs Overcast FOVs Algorithm Performance Retrieval Algorithm Performance 10 / 25

AVHRR IASI Case I: Clear (AVHRR CLAVR-X Cloud-Mask) FOVs E. Maddy et - Hole Hunting al. Background and Motivation FOV 3 clear by AVHRR CLAVR-X Cloud Mask Collocation Cloud R cc ν = R F OV 3 Clearing ν Methodology Clear FOVs Partially Cloudy FOVs Overcast FOVs Algorithm Performance Retrieval Algorithm Performance FOV 2 and FOV 4 clear by AVHRR CLAVR-X Cloud Mask ν = 1 R cc 2 ( R F OV 2 + R F OV 4 ) ν ν and so on . . . 11 / 25

AVHRR IASI Case II: Partially Cloudy (AVHRR CLAVR-X E. Maddy et Cloud-Mask) FOVs al. Background and Motivation Sort FOVs to minimize amplification of noise and calculate the cloud-cleared radiance R cc ν using Collocation FOVs j and k Cloud Clearing F OV j “ F OV j ” Methodology − R F OV k R cc ν ( j, k ) = R + η ( j, k ) R ν ν ν Clear FOVs Partially Cloudy FOVs Overcast FOVs with η ( j, k ) Algorithm Performance Retrieval Algorithm F OV j F OV j h i h − R F OV k i P 2 R clr Performance A i − R R i =1 A i A i A i η ( j, k ) = i 2 h F OV j P 2 − R F OV k R i =1 A i A i Note that we chose to perform single η experiments at this time. Future work will extend results to multiple- η ’ s and cloud-formations. 12 / 25

AVHRR IASI Case II: Partially Cloudy (AVHRR CLAVR-X E. Maddy et Cloud-Mask) FOVs al. Background and Motivation Sort FOVs to minimize amplification of noise and calculate the cloud-cleared radiance R cc ν using Collocation FOVs j and k Cloud Clearing F OV j “ F OV j ” − R F OV k Methodology R cc ν ( j, k ) = R + η ( j, k ) R ν ν ν Clear FOVs Partially Cloudy FOVs Overcast FOVs with η ( j, k ) Algorithm Performance Retrieval Algorithm h F OV j i h F OV j i − R F OV k P 2 R clr A i − R R Performance i =1 A i A i A i η ( j, k ) = i 2 h F OV j − R F OV k P 2 R i =1 A i A i We then select FOVs ( j ′ , k ′ ) such that R cc ν ( j ′ , k ′ ) has the minimum amplification factor and agrees best with the clear estimate R clr A i . 13 / 25