Location: Hydrology and Remote Sensing Laboratory
Title: A spatial-temporal seamless evapotranspiration product based on TSEB-SM model and DNN-ETo method across ChinaAuthor
![]() |
ZHAO, GENGLE - Anhui Academy Of Agricultural Sciences |
![]() |
SONG, LISHENG - Anhui Academy Of Agricultural Sciences |
![]() |
Kustas, William |
![]() |
LIU, SHAOMIN - Beijing Normal University |
![]() |
YIN, GAOFEI - Jiaotong University |
![]() |
ZHAO, LONG - Southwest University |
|
Submitted to: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 12/5/2025 Publication Date: 12/11/2025 Citation: Zhao, G., Song, L., Kustas, W.P., Liu, S., Yin, G., Zhao, L. 2025. A spatial-temporal seamless evapotranspiration product based on TSEB-SM model and DNN-ETo method across China. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 19:2300-2313. https://doi.org/10.1109/JSTARS.2025.3642952. DOI: https://doi.org/10.1109/JSTARS.2025.3642952 Interpretive Summary: A reliable daily evapotranspiration (ET) product is crucial for quantifying the water and energy balance from field to regional scales. Currently, there is no reliable daily ET product over China. In this study, a daily ET dataset for China from 2001 to 2020, with spatial resolution of 1 kilometer was generated by the two-source energy balance model coupled to soil moisture (TSEB-SM) and a novel reconstruction method using deep neural networks and reference evapotranspiration. This dataset showed satisfactory agreement with ground measurements across various landcover types. In addition, the spatial distributions and interannual variations in ET generated by the model showed similar performance to previous studies, while providing more reasonable values in the arid and semi-arid regions. This dataset generated by TSEB-SM model provides information on ET and its components of evaporation and transpiration over China from an energy balance perspective and has potential in providing useful information for water resources management and climate change research. Technical Abstract: A reliable daily evapotranspiration (ET) product is crucial for quantifying the water and energy balance from field to regional scales. Currently, there is no reliable daily ET product over China. In this study, we developed a daily ET dataset (also including transpiration, T and evaporation, E) for China from 2001 to 2020, with spatial resolution of 0.01°. This dataset is generated through the two-source energy balance model coupled to soil moisture (TSEB-SM) and a novel reconstruction method using deep neural networks (DNN) and reference evapotranspiration (ETo) (DNN-ETo). This dataset showed satisfactory agreement with ground measurements across various landcover types with an average root mean square error (RMSE) value of 1.02 mm/day and a bias of -0.17 mm/day. However greater scatter with observations was observed for daily ET > 2 mm/day which will require future research to address larger model error at higher ET rates. Regarding spatial distributions and inter-annual variations, TSEB-SM ET exhibits similar performance to previous studies, while providing more reasonable values in the arid and semi-arid regions. The dataset provides information on ET and its components of E and T over China from an energy balance perspective and has potential in providing useful information for water resources management and climate change research. |
