Jalilvand et al. (2025) Characterization of irrigation timing using thermal satellite observations, a data-driven approach

Identification

• Journal: Remote Sensing of Environment
• Year: 2025
• Date: 2025-11-20
• Authors: Ehsan Jalilvand, Sujay V. Kumar, Charles Truong, Erin Haacker, Sarith Mahanama
• DOI: 10.1016/j.rse.2025.115153

Research Groups

• Hydrological Sciences Laboratory, NASA Goddard Space Flight Center, USA.
• Science Applications International Corporation (SAIC), USA.
• Centre Borelli, Université Paris Saclay, ENS Paris-Saclay, CNRS, France.
• Department of Earth and Atmospheric Sciences, University of Nebraska-Lincoln, USA.

Short Summary

This study presents a data-driven framework using thermal satellite observations and change point detection to estimate irrigation timing and individual events. By comparing cropland land surface temperature (LST) with nearby natural vegetation, the method accurately identifies irrigation schedules in diverse agricultural regions.

Objective

• To develop a generalized, data-driven approach for characterizing irrigation timing attributes (start, end, and individual events) using thermal remote sensing to improve the representation of human-driven water use in land surface models.

Study Configuration

• Spatial Scale: Regional and site-specific analysis conducted over Nebraska (USA) and Mahabad (Iran).
• Temporal Scale: Interannual analysis focusing on the irrigation season; performance evaluated against in situ records.

Methodology and Data

• Models used: Change point detection algorithm (AgroTrack); hyperparameter tuning for optimization across different irrigation practices.
• Data sources: MODIS Land Surface Temperature (LST) satellite observations; in situ irrigation data for validation (Nebraska and Mahabad sites).
• Analytical Approach: Comparison of LST at cropland pixels against hydrologically similar natural land cover pixels nearby to isolate the cooling effect of irrigation from natural climatic variability.

Main Results

• Season Characterization: The method detected the start and end of the irrigation season with duration errors of 18% in Nebraska and 15% in Mahabad.
• Event Detection: Individual irrigation event detection yielded F1-scores (combined precision and recall) ranging from 0.59 to 0.74 across 10 sites in Nebraska.
• Error Analysis: Cloud cover during the transition periods (start or end of the season) was identified as the primary source of error in timing estimation.
• Parameter Optimization: Extensive hyperparameter tuning resulted in specific algorithmic configurations tailored to different regional irrigation practices.

Contributions

• Provides a generalized framework for irrigation timing that does not rely on the simplified assumptions or complex parameterizations typical of current land surface models.
• Demonstrates the utility of thermal remote sensing (LST) as a robust proxy for human-induced changes in the water cycle.
• Offers a validated method for improving irrigation water use estimation by accurately capturing the "when" of water application.

Funding

• NASA Goddard Space Flight Center.
• Science Applications International Corporation (SAIC).
• [Specific project codes/reference numbers were not explicitly listed in the provided text].

Citation

@article{Jalilvand2025Characterization,
  author = {Jalilvand, Ehsan and Kumar, Sujay V. and Truong, Charles and Haacker, Erin and Mahanama, Sarith},
  title = {Characterization of irrigation timing using thermal satellite observations, a data-driven approach},
  journal = {Remote Sensing of Environment},
  year = {2025},
  doi = {10.1016/j.rse.2025.115153},
  url = {https://doi.org/10.1016/j.rse.2025.115153}
}

Generated by BiblioAssistant using gemini-3-flash-preview (Google API)