Evaluation of machine learning methods for forecasting turbidity in river networks using Sentinel-2 remote sensing data

Turbidity is an important indicator of river water quality and of great interest to improve the data acquisition methods and efficiency of decision support systems for sustainable ecosystem management. However, river water quality monitoring stations are very expensive to operate and maintain and la...

Full description

Saved in:
Bibliographic Details
Main Authors: Victor Oliveira Santos, Paulo Alexandre Costa Rocha, Jesse Van Griensven Thé, Bahram Gharabaghi
Format: Article
Language:English
Published: Elsevier 2025-12-01
Series:Ecological Informatics
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S157495412500322X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849233555631112192
author Victor Oliveira Santos
Paulo Alexandre Costa Rocha
Jesse Van Griensven Thé
Bahram Gharabaghi
author_facet Victor Oliveira Santos
Paulo Alexandre Costa Rocha
Jesse Van Griensven Thé
Bahram Gharabaghi
author_sort Victor Oliveira Santos
collection DOAJ
description Turbidity is an important indicator of river water quality and of great interest to improve the data acquisition methods and efficiency of decision support systems for sustainable ecosystem management. However, river water quality monitoring stations are very expensive to operate and maintain and lack spatial coverage. Therefore, this study takes advantage of the vast spatial coverage of remote sensing datasets from satellites to provide a more efficient hybrid system with comprehensive coverage of both spatial and temporal changes in water quality across a vast river network. Spectral bands from Sentinel-2 were analyzed using machine learning algorithms, namely XGBoost, Random Forests, GMDH, Support Vector Regression, k-Nearest Neighbors and Least Absolute Shrinkage and Selection Operator to model turbidity, using data from twelve monitoring stations across the Mississippi River, USA. Results show that considering the individual monitoring stations, the ML algorithms for turbidity modeling were satisfactory at locations with a larger range and standard deviation of turbidity values, achieving a mean R2 value of 59.5 %. Tree-based models were the best overall approach, often ranking as the best or second-best performing model. Using all the samples from the monitoring stations, the XGBoost provided a superior output for turbidity modeling, reaching R2 equal to 75.7 %. This represents an improvement of over 16 % compared to the average metric value for the individual stations. A comprehensive comparison with the literature found that the models implemented using this study's methodology could provide competitive results, deeming it as an alternative for turbidity modeling from remote sensing data.
format Article
id doaj-art-9018d15261cc40a29b16d4d9b2a8764e
institution Kabale University
issn 1574-9541
language English
publishDate 2025-12-01
publisher Elsevier
record_format Article
series Ecological Informatics
spelling doaj-art-9018d15261cc40a29b16d4d9b2a8764e2025-08-20T05:05:43ZengElsevierEcological Informatics1574-95412025-12-019010331310.1016/j.ecoinf.2025.103313Evaluation of machine learning methods for forecasting turbidity in river networks using Sentinel-2 remote sensing dataVictor Oliveira Santos0Paulo Alexandre Costa Rocha1Jesse Van Griensven Thé2Bahram Gharabaghi3School of Engineering, University of Guelph, 50 Stone Rd E, Guelph, ON N1G 2W1, CanadaSchool of Engineering, University of Guelph, 50 Stone Rd E, Guelph, ON N1G 2W1, Canada; Mechanical Engineering Department, Technology Center, Federal University of Ceará, Fortaleza, Ceará 60020-181, BrazilSchool of Engineering, University of Guelph, 50 Stone Rd E, Guelph, ON N1G 2W1, Canada; Lakes Environmental, 170 Columbia St. W, Waterloo, ON N2L 3L3, CanadaSchool of Engineering, University of Guelph, 50 Stone Rd E, Guelph, ON N1G 2W1, Canada; Corresponding author.Turbidity is an important indicator of river water quality and of great interest to improve the data acquisition methods and efficiency of decision support systems for sustainable ecosystem management. However, river water quality monitoring stations are very expensive to operate and maintain and lack spatial coverage. Therefore, this study takes advantage of the vast spatial coverage of remote sensing datasets from satellites to provide a more efficient hybrid system with comprehensive coverage of both spatial and temporal changes in water quality across a vast river network. Spectral bands from Sentinel-2 were analyzed using machine learning algorithms, namely XGBoost, Random Forests, GMDH, Support Vector Regression, k-Nearest Neighbors and Least Absolute Shrinkage and Selection Operator to model turbidity, using data from twelve monitoring stations across the Mississippi River, USA. Results show that considering the individual monitoring stations, the ML algorithms for turbidity modeling were satisfactory at locations with a larger range and standard deviation of turbidity values, achieving a mean R2 value of 59.5 %. Tree-based models were the best overall approach, often ranking as the best or second-best performing model. Using all the samples from the monitoring stations, the XGBoost provided a superior output for turbidity modeling, reaching R2 equal to 75.7 %. This represents an improvement of over 16 % compared to the average metric value for the individual stations. A comprehensive comparison with the literature found that the models implemented using this study's methodology could provide competitive results, deeming it as an alternative for turbidity modeling from remote sensing data.http://www.sciencedirect.com/science/article/pii/S157495412500322XSentinel satellite constellationMachine learningTurbiditySpectral indicesMississippi riverMissouri river
spellingShingle Victor Oliveira Santos
Paulo Alexandre Costa Rocha
Jesse Van Griensven Thé
Bahram Gharabaghi
Evaluation of machine learning methods for forecasting turbidity in river networks using Sentinel-2 remote sensing data
Ecological Informatics
Sentinel satellite constellation
Machine learning
Turbidity
Spectral indices
Mississippi river
Missouri river
title Evaluation of machine learning methods for forecasting turbidity in river networks using Sentinel-2 remote sensing data
title_full Evaluation of machine learning methods for forecasting turbidity in river networks using Sentinel-2 remote sensing data
title_fullStr Evaluation of machine learning methods for forecasting turbidity in river networks using Sentinel-2 remote sensing data
title_full_unstemmed Evaluation of machine learning methods for forecasting turbidity in river networks using Sentinel-2 remote sensing data
title_short Evaluation of machine learning methods for forecasting turbidity in river networks using Sentinel-2 remote sensing data
title_sort evaluation of machine learning methods for forecasting turbidity in river networks using sentinel 2 remote sensing data
topic Sentinel satellite constellation
Machine learning
Turbidity
Spectral indices
Mississippi river
Missouri river
url http://www.sciencedirect.com/science/article/pii/S157495412500322X
work_keys_str_mv AT victoroliveirasantos evaluationofmachinelearningmethodsforforecastingturbidityinrivernetworksusingsentinel2remotesensingdata
AT pauloalexandrecostarocha evaluationofmachinelearningmethodsforforecastingturbidityinrivernetworksusingsentinel2remotesensingdata
AT jessevangriensventhe evaluationofmachinelearningmethodsforforecastingturbidityinrivernetworksusingsentinel2remotesensingdata
AT bahramgharabaghi evaluationofmachinelearningmethodsforforecastingturbidityinrivernetworksusingsentinel2remotesensingdata