Explainable machine learning to quantify the value of proximal remote sensing in latent energy flux estimation

Proximal remote sensing has the potential to provide critical information on vegetation biophysical factors that can predict land-atmosphere exchange of water and energy. Latent energy (LE) flux is traditionally estimated using process-based models which rely on vegetation parameters that change dur...

Full description

Saved in:
Bibliographic Details
Main Authors: Srishti Gaur, Guler Aslan-Sungur (Rojda), Andy VanLoocke, Darren T. Drewry
Format: Article
Language:English
Published: Elsevier 2025-08-01
Series:Agricultural Water Management
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S0378377425003579
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850075893468233728
author Srishti Gaur
Guler Aslan-Sungur (Rojda)
Andy VanLoocke
Darren T. Drewry
author_facet Srishti Gaur
Guler Aslan-Sungur (Rojda)
Andy VanLoocke
Darren T. Drewry
author_sort Srishti Gaur
collection DOAJ
description Proximal remote sensing has the potential to provide critical information on vegetation biophysical factors that can predict land-atmosphere exchange of water and energy. Latent energy (LE) flux is traditionally estimated using process-based models which rely on vegetation parameters that change during the growing season. Data-driven models have the potential to address these issues by offering flexible predictor selection and more efficient utilization of the information in predictor sets. These models require careful choice of predictors to avoid redundancy and allow robust cross-validation. In this study we present a systematic and comprehensive evaluation of machine learning (ML) models to assess the capability of meteorological and proximal sensing data for predicting LE at a half-hourly temporal resolution across multiple growing seasons for an agricultural system. The results presented here demonstrate that a model using four environmental predictors in combination with two proximal sensing variables can capture 88 % of the variability in LE. ML models using only three predictors (one meteorological and two proximal remote sensing) captured 81 % of LE variability, offering the best trade-off between performance and complexity. An ML model utilizing only two predictors, one proximal remote sensing variable and downwelling radiation, captured 77 % of LE variability. These results demonstrate the power of proximal remote sensing and meteorological observations to estimate land-atmosphere water vapor exchange, providing a solution where more direct methods such as eddy covariance are not available and for evaluations of agronomic management and genotypic variations.
format Article
id doaj-art-252a1a1c7a694a32a6a9ca775a659059
institution DOAJ
issn 1873-2283
language English
publishDate 2025-08-01
publisher Elsevier
record_format Article
series Agricultural Water Management
spelling doaj-art-252a1a1c7a694a32a6a9ca775a6590592025-08-20T02:46:09ZengElsevierAgricultural Water Management1873-22832025-08-0131710964310.1016/j.agwat.2025.109643Explainable machine learning to quantify the value of proximal remote sensing in latent energy flux estimationSrishti Gaur0Guler Aslan-Sungur (Rojda)1Andy VanLoocke2Darren T. Drewry3Department of Food, Agricultural and Biological Engineering, Ohio State University, Columbus, OH, USADepartment of Agronomy, Iowa State University, Ames, IA, USADepartment of Agronomy, Iowa State University, Ames, IA, USADepartment of Food, Agricultural and Biological Engineering, Ohio State University, Columbus, OH, USA; Department of Horticulture and Crop Science, Ohio State University, Columbus, OH, USA; Translational Data Analytics Institute, Ohio State University, Columbus, OH, USA; Corresponding author at: Department of Food, Agricultural, and Biological Engineering, Ohio State University, Columbus, OH, USA.Proximal remote sensing has the potential to provide critical information on vegetation biophysical factors that can predict land-atmosphere exchange of water and energy. Latent energy (LE) flux is traditionally estimated using process-based models which rely on vegetation parameters that change during the growing season. Data-driven models have the potential to address these issues by offering flexible predictor selection and more efficient utilization of the information in predictor sets. These models require careful choice of predictors to avoid redundancy and allow robust cross-validation. In this study we present a systematic and comprehensive evaluation of machine learning (ML) models to assess the capability of meteorological and proximal sensing data for predicting LE at a half-hourly temporal resolution across multiple growing seasons for an agricultural system. The results presented here demonstrate that a model using four environmental predictors in combination with two proximal sensing variables can capture 88 % of the variability in LE. ML models using only three predictors (one meteorological and two proximal remote sensing) captured 81 % of LE variability, offering the best trade-off between performance and complexity. An ML model utilizing only two predictors, one proximal remote sensing variable and downwelling radiation, captured 77 % of LE variability. These results demonstrate the power of proximal remote sensing and meteorological observations to estimate land-atmosphere water vapor exchange, providing a solution where more direct methods such as eddy covariance are not available and for evaluations of agronomic management and genotypic variations.http://www.sciencedirect.com/science/article/pii/S0378377425003579Proximal remote sensingExplainable machine learningLatent energy fluxEvapotranspirationSurface energy balanceVegetation biophysics
spellingShingle Srishti Gaur
Guler Aslan-Sungur (Rojda)
Andy VanLoocke
Darren T. Drewry
Explainable machine learning to quantify the value of proximal remote sensing in latent energy flux estimation
Agricultural Water Management
Proximal remote sensing
Explainable machine learning
Latent energy flux
Evapotranspiration
Surface energy balance
Vegetation biophysics
title Explainable machine learning to quantify the value of proximal remote sensing in latent energy flux estimation
title_full Explainable machine learning to quantify the value of proximal remote sensing in latent energy flux estimation
title_fullStr Explainable machine learning to quantify the value of proximal remote sensing in latent energy flux estimation
title_full_unstemmed Explainable machine learning to quantify the value of proximal remote sensing in latent energy flux estimation
title_short Explainable machine learning to quantify the value of proximal remote sensing in latent energy flux estimation
title_sort explainable machine learning to quantify the value of proximal remote sensing in latent energy flux estimation
topic Proximal remote sensing
Explainable machine learning
Latent energy flux
Evapotranspiration
Surface energy balance
Vegetation biophysics
url http://www.sciencedirect.com/science/article/pii/S0378377425003579
work_keys_str_mv AT srishtigaur explainablemachinelearningtoquantifythevalueofproximalremotesensinginlatentenergyfluxestimation
AT guleraslansungurrojda explainablemachinelearningtoquantifythevalueofproximalremotesensinginlatentenergyfluxestimation
AT andyvanloocke explainablemachinelearningtoquantifythevalueofproximalremotesensinginlatentenergyfluxestimation
AT darrentdrewry explainablemachinelearningtoquantifythevalueofproximalremotesensinginlatentenergyfluxestimation