Dai et al. (2026) Impacts of surface heterogeneity on energy partitioning in paddy ecosystems: A dual eddy covariance study

Identification

• Journal: Agricultural and Forest Meteorology
• Year: 2026
• Date: 2026-04-04
• Authors: Yulong Dai, Bo Liu, Jun Wei, Y. Shi, Qilin Li, Fangping Liu, Yufeng Luo, Yuanlai Cui
• DOI: 10.1016/j.agrformet.2026.111165

Research Groups

• State Key Laboratory of Water Resources Engineering and Management, Wuhan University, Wuhan 430072, China
• College of Hydraulic Science and Engineering, Yangzhou University, Yangzhou 225009, China
• College of Water Conservancy and Hydropower Engineering, Sichuan Agricultural University, Yaan 625014, China
• The National Key Laboratory of Water Disaster Prevention, Nanjing Hydraulic Research Institute, Nanjing 210029, China
• Jiangxi Central Irrigation Experiment Station, Nanchang 330201, China

Short Summary

This study quantifies the impacts of surface heterogeneity on energy partitioning in paddy ecosystems using dual eddy covariance and UAV-based footprint modeling. It reveals that even low heterogeneity significantly biases latent heat flux measurements due to advection and physiological responses, and proposes a two-stage correction framework to rectify these biases.

Objective

• Quantify the impacts of surface heterogeneity on energy partitioning in paddy ecosystems and develop a correction framework to rectify biases in eddy covariance measurements, particularly for latent heat flux.

Study Configuration

• Spatial Scale: Heterogeneous (EC1) and homogeneous (EC2) paddy sites.
• Temporal Scale: Not explicitly defined in the provided text, but implies a period sufficient for synchronous observations.

Methodology and Data

• Models used: UAV-based footprint modeling, CatBoost model.
• Data sources: Synchronous dual eddy covariance observations, UAV-based footprint data, microclimate parameters (temperature, vapor pressure deficit, relative humidity, wind speed), turbulent kinetic energy (TKE) components.

Main Results

• Anthropogenic impervious infrastructure at heterogeneous sites (EC1) elevated sensible heat flux by 25.7 % and suppressed latent heat flux (LE) by 30.2 %.
• This created microclimatic "dry islands" characterized by higher temperature (+1.8 °C), vapor pressure deficit (+50.3 %), and Bowen ratio (+147.3 %), alongside lower relative humidity (–16.2 %) and mean wind speed (–10.9 %).
• Turbulent kinetic energy (TKE) reorganization, specifically increased horizontal and decreased vertical velocity fluctuations, confirmed the presence of local advection.
• Significant LE deficits, reaching –119 W m⁻², were observed even when the permeable footprint coverage exceeded 90 %.
• A two-stage correction framework was developed: an area-based adjustment improved energy closure by 13.4 %, and a CatBoost model successfully retrieved the unadulterated crop signal (coefficient of determination R²=0.82, relative error RE=13.19 %).
• Feature importance analysis indicated that permeable area deficit controlled baseline LE, while turbulence metrics captured advection-mediated deviations.

Contributions

• Provides a quantitative assessment of the significant impact of surface heterogeneity on eddy covariance measurements of energy partitioning in paddy ecosystems, even at relatively low levels of heterogeneity.
• Identifies and decouples dual mechanisms for latent heat flux suppression: eddy covariance measurement biases from uncaptured local advection and physiological reductions in crop transpiration driven by advection-enhanced vapor pressure deficit.
• Develops a novel two-stage correction framework (area-based adjustment combined with a CatBoost machine learning model) to effectively rectify these combined biases.
• Offers a critical pathway to enhance the accuracy of global agricultural evapotranspiration monitoring by addressing heterogeneity-induced biases.

Funding

• Not explicitly mentioned in the provided text.

Citation

@article{Dai2026Impacts,
  author = {Dai, Yulong and Liu, Bo and Wei, Jun and Shi, Y. and Li, Qilin and Liu, Fangping and Luo, Yufeng and Cui, Yuanlai},
  title = {Impacts of surface heterogeneity on energy partitioning in paddy ecosystems: A dual eddy covariance study},
  journal = {Agricultural and Forest Meteorology},
  year = {2026},
  doi = {10.1016/j.agrformet.2026.111165},
  url = {https://doi.org/10.1016/j.agrformet.2026.111165}
}