Near Real‐Time Mapping of All‐Sky Land Surface Temperature From GOES‐R Using Machine Learning DOI Creative Commons
Sadegh Ranjbar, Danielle Losos, Sophie Hoffman

et al.

Journal of Geophysical Research Machine Learning and Computation, Journal Year: 2025, Volume and Issue: 2(2)

Published: April 25, 2025

Abstract Land surface temperature (LST) is crucial for understanding earth system processes. We expanded the Advanced Baseline Imager Live Imaging of Vegetated Ecosystems (ALIVE) framework to estimate LST in near‐real‐time both cloudy and clear sky conditions at a five‐minute resolution. compared two machine learning (ML) models, Long Short‐Term Memory (LSTM) networks Gradient Boosting Regressor (GBR), using top‐of‐atmosphere observations from (ABI) on GOES‐16 satellite against hundreds observation sites five‐year period. outperformed GBR, especially coarser resolutions under challenging conditions, with R 2 0.96 (RMSE 2.31K) 0.83 4.10K) across CONUS, based 10‐repeat Leave‐One‐Out Cross‐Validation (LOOCV). GBR maintained high accuracy ran 5.3 times faster, only 0.01–0.02 drop. Feature importance revealed infrared bands were key LSTM adapting dynamically atmospheric changes, while utilized more time information conditions. A comparative analysis physically ABI product showed strong agreement winter, particularly also highlighting challenges summer estimation due increased thermal variability. This study underscores strengths limitations data‐driven models suggests potential pathways integrating ML enhance coverage products.

Language: Английский

Advancing hourly gross primary productivity mapping over East Asia using Himawari-8 AHI and artificial intelligence: Unveiling the impact of aerosol-induced radiation dynamics DOI
Soo Ya Bae, Bokyung Son, Taejun Sung

et al.

Remote Sensing of Environment, Journal Year: 2025, Volume and Issue: 323, P. 114735 - 114735

Published: April 6, 2025

Language: Английский

Citations

0
Rapid changes in terrestrial carbon dioxide uptake captured in near-real time from a geostationary satellite: The ALIVE framework DOI Creative Commons
Danielle Losos, Sadegh Ranjbar, Sophie Hoffman

et al.

Remote Sensing of Environment, Journal Year: 2025, Volume and Issue: 324, P. 114759 - 114759

Published: April 14, 2025

Language: Английский

Citations

0
Near Real‐Time Mapping of All‐Sky Land Surface Temperature From GOES‐R Using Machine Learning DOI Creative Commons
Sadegh Ranjbar, Danielle Losos, Sophie Hoffman

et al.

Journal of Geophysical Research Machine Learning and Computation, Journal Year: 2025, Volume and Issue: 2(2)

Published: April 25, 2025

Abstract Land surface temperature (LST) is crucial for understanding earth system processes. We expanded the Advanced Baseline Imager Live Imaging of Vegetated Ecosystems (ALIVE) framework to estimate LST in near‐real‐time both cloudy and clear sky conditions at a five‐minute resolution. compared two machine learning (ML) models, Long Short‐Term Memory (LSTM) networks Gradient Boosting Regressor (GBR), using top‐of‐atmosphere observations from (ABI) on GOES‐16 satellite against hundreds observation sites five‐year period. outperformed GBR, especially coarser resolutions under challenging conditions, with R 2 0.96 (RMSE 2.31K) 0.83 4.10K) across CONUS, based 10‐repeat Leave‐One‐Out Cross‐Validation (LOOCV). GBR maintained high accuracy ran 5.3 times faster, only 0.01–0.02 drop. Feature importance revealed infrared bands were key LSTM adapting dynamically atmospheric changes, while utilized more time information conditions. A comparative analysis physically ABI product showed strong agreement winter, particularly also highlighting challenges summer estimation due increased thermal variability. This study underscores strengths limitations data‐driven models suggests potential pathways integrating ML enhance coverage products.

Language: Английский

Citations

0