Infrared and Microwave Remote Sensing of Sea Surface Temperature Gary A Wick NOAA Environmental Technology Laboratory January 14 2004 Outline Motivation Basic SST Retrieval Methods Current Multi Sensor Merging Efforts Why SST Boundary Condition Weather Models Estimation of Heat Content and Heat Flux Climate Monitoring and Change Detection Naval Operations Climate Anomalies Courtesy NOAA Climate Diagnostics Center Why Satellites Courtesy R Reynolds NOAA NCDC Desired Accuracy WCRP 1985 Tropics 0 3 K on 2 grid every 15 days Robinson et al 1984 Global SST Monitoring 0 05 K on 5 grid every 15 days NPOESS SST EDR Objectives 0 1 K uncertainty at 4 km resolution Definition of SST Interface SST Skin SST Sub skin SST Near Surface SST or SSTDepth Radiative Transfer Equation Methods for SST Retrieval Thermal Infrared Passive Microwave Infrared Retrievals Strengths High Accuracy High Resolution Long Heritage over 20 years Weaknesses Obscured by Clouds Atmospheric Corrections Required Microwave Retrievals Strengths Clouds Transparent Relatively Insensitive to Atmospheric Effects Weaknesses Sensitive to Surface Roughness Poorer Accuracy Poorer Resolution Spatial Coverage Differences Infrared Retrieval Technique Cloud Detection Atmospheric Correction Multi Channel SST TS T1 T1 T2 Multi Frequency Multiple View Algorithm Refinements Additional path length term NLSST Use of multiple frequencies AND multiple view angles Independent estimate of water vapor content Iterative solution for both SST and Microwave Retrieval Technique Environmental Scenes 42 195 Radiosondes 5 Cloud Models SST Randomly Varied for 0 to 30 C Wind Speed Randomly Varied from 0 to 20 m s Wind Direction Randomly Varied from 0 to 360 Truth Ts W V L Complete Radiative Transfer Model Simulated AMSR TB s Gaussian Noise Added Withheld Data Set Derive Coefficients for Multiple Linear Regression Algorithm Algorithm Coefficients Run Algorithm Retrieved values for Ts W V L Evalulate Algorithm Peformance Performance and Cross Talk Statistics Courtesy Remote Sensing Systems Infrared Sensors AVHRR ATSR GOES Imager MODIS Others GMS SEVIRI VIRS Microwave Sensors TMI AMSR WindSat Multi Sensor Blended SST Current Projects Key Issues Sample Results GODAE High Resolution SST Pilot Project Provide rapidly and regularly distributed global multisensor high quality SST products at a fine spatial and temporal resolution Most promising solution to combine complementary infrared and passive microwave satellite measurements with quality controlled in situ observations from ships and buoys www ghrsst pp org Next Generation SST Created by Hiroshi Kawamura Tohoku University Japan http www ocean caos tohoku ac jp adeos sst Blended SST Issues Different product resolutions Different sensor error characteristics Different sampling times and effective depths Merging techniques Error Characteristics Overall Accuracy Observed Differences Between Infrared and Microwave Products Comparisons between the products show complex spatial and temporal differences Sources of Product Differences Diurnal Warming Effects Skin Layer Effects Courtesy S Castro U Colorado Courtesy P Minnett U Miami Blended Infrared and Microwave SST Using derived corrections the infrared and microwave SST products can be more accurately merged into a new enhanced product Accuracy of Merged Product vs Buoys Diurnal warming effects are aliased into the product if not corrected Strong winds off Somalia cause perceived overcooling and large swath edge effects are visible Bias K RMS K w Adj 0 01 0 61 w o Adj 0 15 0 67 NOAA Environmental Technology Laboratory Analyzed SST Product Analysis Characteristics Daily global 40 N 40 S 0 25 degree Referenced to nighttime predawn value Based on Reynolds and Smith Optimal Interpolation Relative product uncertainties derived from difference analyses Analyzed Product Accuracy Summary Product Bias K RMS K Full Analysis 0 13 0 68 Night Obs Only 0 08 0 58 AVHRR Obs Only 0 01 0 56 TMI Obs Only 0 22 0 74 Refined diurnal corrections are the most needed improvement Summary Complementary infrared and microwave SST products provide the opportunity for cross validation and improved SST Multiple sensor related and geophysical effects lead to complex differences between the products Optimal blending of the products requires careful treatment of the differences Is blending correct