Qian et al. (2026) A benchmark dataset for half-hourly evapotranspiration estimation in China from 2000 to 2024

Identification

• Journal: Earth system science data
• Year: 2026
• Date: 2026-09-11
• Authors: Long Qian, Xingjiao Yu, Lifeng Wu, Yaokui Cui, Zhitao Zhang, Junying Chen, Sumeng Ye, Xuqian Bai, Xiaogang Liu, Sien Li, Rangjian Qiu
• DOI: 10.5194/essd-18-6707-2026

Research Groups

• School of Soil and Water Conservation, Jiangxi University of Water Resources and Electric Power
• College of Water Resources and Architectural Engineering, Northwest A&F University
• Institute of RS and GIS, School of Earth and Space Sciences, Peking University
• Faculty of Modern Agricultural Engineering, Kunming University of Science and Technology
• Center for Agricultural Water Research in China, China Agricultural University
• State Key Laboratory of Water Resources Engineering and Management, Wuhan University

Short Summary

This study presents a benchmark dataset for half-hourly evapotranspiration estimation in China from 2000 to 2024. The dataset is based on observations from 50 ChinaFlux sites and uses an automated machine learning approach (AutoML) to fill gaps and prolong time series.

Objective

• To develop a continuous ground-based evapotranspiration benchmark dataset for China covering the period 2000-2024.
• To evaluate the performance of AutoML in filling gaps and prolonging time series of latent heat flux observations.

Study Configuration

• Spatial Scale: Site-level, with 50 ChinaFlux sites across major climate zones and underlying surface types in China.
• Temporal Scale: Half-hourly data from 2000 to 2024.

Methodology and Data

• Models used: Automated machine learning (AutoML) approach based on the H2O AutoML platform.
• Data sources:
  • ERA5-Land reanalysis data for meteorological and hydrological driving variables.
  • MODIS products for remotely sensed vegetation indices (NDVI and LAI).

Main Results

• The constructed dataset provides a temporally continuous, quality-controlled, and physically consistent ground-based evapotranspiration benchmark for China.
• AutoML achieves high accuracy in gap-filling within observation periods and reasonable temporal prolongation beyond measurement intervals.

Contributions

• This study addresses the limitations of existing ChinaFlux observations by developing a comprehensive flux tower-based evapotranspiration observation dataset for China.
• The constructed dataset provides essential data support for evaluating remotely sensed ET products, validating reanalysis datasets and land surface process models, and studying regional hydrological and ecosystem processes.

Funding

• This research was supported by the National Natural Science Foundation of China (grant numbers: 51879207, 52009203).
• The ERA5-Land reanalysis data were provided by the European Centre for Medium-Range Weather Forecasts (ECMWF).

Citation

@article{Qian2026benchmark,
  author = {Qian, Long and Yu, Xingjiao and Wu, Lifeng and Cui, Yaokui and Zhang, Zhitao and Chen, Junying and Ye, Sumeng and Bai, Xuqian and Liu, Xiaogang and Li, Sien and Qiu, Rangjian},
  title = {A benchmark dataset for half-hourly evapotranspiration estimation in China from 2000 to 2024},
  journal = {Earth system science data},
  year = {2026},
  doi = {10.5194/essd-18-6707-2026},
  url = {https://doi.org/10.5194/essd-18-6707-2026}
}