Machine Learning to Retrieve Gap-Free Land Surface Temperature from Infrared Atmospheric Sounding Interferometer Observations DOI Creative Commons
Fabio Della Rocca, Pamela Pasquariello, Guido Masiello

и другие.

Remote Sensing, Год журнала: 2025, Номер 17(4), С. 694 - 694

Опубликована: Фев. 18, 2025

Retrieving LST from infrared spectral observations is challenging because it needs separation emissivity in surface radiation emission, which feasible only when the state of surface–atmosphere system known. Thanks to its high resolution, Infrared Atmospheric Sounding Interferometer (IASI) instrument onboard Metop polar-orbiting satellites sensor that can simultaneously retrieve LST, spectrum, and atmospheric composition. Still, cannot penetrate thick cloud layers, making blind emissions under cloudy conditions, with parameters being flagged as voids. The present paper aims discuss a downscaling–fusion methodology missing values on spatial field retrieved spatially scattered IASI yield level 3, regularly gridded data, using proxy data Spinning Enhanced Visible Imager (SEVIRI) flying Meteosat Second Generation (MSG) platform, geostationary instrument, Advanced Very High-Resolution Radiometer (AVHRR) satellites. We address this problem by machine learning techniques, i.e., Gradient Boosting, Random Forest, Gaussian Process Regression, Neural Network, Stacked Regression. applied over Po Valley region, very heterogeneous area allows addressing trained models’ robustness. Overall, methods significantly enhanced sampling, keeping errors terms Root Mean Square Error (RMSE) bias (Mean Absolute Error, MAE) low. Although we demonstrate assess results primarily also intended for applications follow-on, is, Next (IASI-NG), much more Sounder (IRS), planned fly year, 2025, Third platform (MTG).

Язык: Английский

Machine Learning to Retrieve Gap-Free Land Surface Temperature from Infrared Atmospheric Sounding Interferometer Observations DOI Creative Commons
Fabio Della Rocca, Pamela Pasquariello, Guido Masiello

и другие.

Remote Sensing, Год журнала: 2025, Номер 17(4), С. 694 - 694

Опубликована: Фев. 18, 2025

Retrieving LST from infrared spectral observations is challenging because it needs separation emissivity in surface radiation emission, which feasible only when the state of surface–atmosphere system known. Thanks to its high resolution, Infrared Atmospheric Sounding Interferometer (IASI) instrument onboard Metop polar-orbiting satellites sensor that can simultaneously retrieve LST, spectrum, and atmospheric composition. Still, cannot penetrate thick cloud layers, making blind emissions under cloudy conditions, with parameters being flagged as voids. The present paper aims discuss a downscaling–fusion methodology missing values on spatial field retrieved spatially scattered IASI yield level 3, regularly gridded data, using proxy data Spinning Enhanced Visible Imager (SEVIRI) flying Meteosat Second Generation (MSG) platform, geostationary instrument, Advanced Very High-Resolution Radiometer (AVHRR) satellites. We address this problem by machine learning techniques, i.e., Gradient Boosting, Random Forest, Gaussian Process Regression, Neural Network, Stacked Regression. applied over Po Valley region, very heterogeneous area allows addressing trained models’ robustness. Overall, methods significantly enhanced sampling, keeping errors terms Root Mean Square Error (RMSE) bias (Mean Absolute Error, MAE) low. Although we demonstrate assess results primarily also intended for applications follow-on, is, Next (IASI-NG), much more Sounder (IRS), planned fly year, 2025, Third platform (MTG).

Язык: Английский

Процитировано

0