Abstract
An neural network model of data mining is used to identify error sources in satellite-derived tropical sea surface temperature (SST) estimates from thermal infrared sensors onboard the Geostationary Operational Environmental Satellite (GOES). By using the Back Propagation Network (BPN) algorithm, it is found that air temperature, relative humidity, and wind speed variation are the major factors causing the errors of GOES SST products in the tropical Pacific. The accuracy of SST estimates is also improved by the model. The root mean square error (RMSE) for the daily SST estimate is reduced from 0.58 K to 0.38 K and mean absolute percentage error (MAPE) is 1.03%. For the hourly mean SST estimate, its RMSE is also reduced from 0.66 K to 0.44 K and the MAPE is 1.3%.
Highlights
Sea Surface Temperature (SST) is an important factor in ocean processes with a major impact on weather and climate [1]
Multi-channel Sea Surface Temperature (MCSST) is the first algorithm to derive SST from satellite infrared sensors. It is derived from radiances collected by the Advanced Very High Resolution Radiometer (AVHRR) sensor carried onboard the National Oceanic and Atmospheric Administration (NOAA) polar orbiting satellites
Results of the Back Propagation Network (BPN) model successfully improve the accuracy of Geostationary Operational Environmental Satellite (GOES) SST and reveal that air temperature and relative humidity are the two main factors contributing to GOES SST bias
Summary
Sea Surface Temperature (SST) is an important factor in ocean processes with a major impact on weather and climate [1]. Multi-channel Sea Surface Temperature (MCSST) is the first algorithm to derive SST from satellite infrared sensors. It is derived from radiances collected by the Advanced Very High Resolution Radiometer (AVHRR) sensor carried onboard the National Oceanic and Atmospheric Administration (NOAA) polar orbiting satellites. The NLSST accounts for a minor non-linearity in water vapor by including a first-guess SST that is a surrogate for total water vapor amounts [7,8,9,10]. By using these algorithms, NOAA has been continuously providing research quality SST data since 1981. The higher system noise levels of SST derived from geostationary satellites are larger than those derived from polar orbiting satellites [11]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
