一种基于遥感变量相关性的蒸散发数据高分辨率重建方法
By using a high-resolution reconstruction method based on the correlation of remote sensing variables, the challenges of high accuracy and high spatiotemporal resolution in existing ET estimation techniques are solved, achieving efficient and accurate evapotranspiration data reconstruction, which is suitable for large-scale regional and long-term series applications.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- INST OF GEOGRAPHICAL SCI & NATURAL RESOURCE RES CAS
- Filing Date
- 2025-10-28
- Publication Date
- 2026-07-17
AI Technical Summary
Existing ET estimation techniques are insufficient to meet the requirements of high accuracy, high spatiotemporal resolution, and strong applicability. Physical energy balance methods suffer from parameterization complexity, computational chain fragility, scale adaptation issues, and timeliness constraints. Remote sensing-nonparametric machine learning methods are limited by data dependence, hyperparameter sensitivity, and lack of physical mechanisms, resulting in limited model generalization ability.
A high-resolution reconstruction method based on the correlation of remote sensing variables was adopted. By collecting and preprocessing various remote sensing data, nine variables related to evapotranspiration were selected. An enhanced spatiotemporal adaptive reflectance fusion model was used for data fusion. Potential evapotranspiration was calculated by combining daily meteorological data, and a univariate-evapotranspiration regression model was established. Finally, high spatial and temporal resolution evapotranspiration data reconstruction was achieved.
It significantly improves the computational efficiency and accuracy of evapotranspiration estimation, simplifies the modeling process, reduces computational resource requirements, is suitable for batch processing of large-scale regions and long-term series, and the estimation results are highly consistent with ground observations, demonstrating good interpretability and applicability.
Smart Images
Figure CN121458540B_ABST