Mapping total suspended solids (TSS) and dissolved organic carbon (DOC) in complex coastal waters using deep learning enhanced remote sensing

Satellite remote sensing enables the mapping of large-scale spatiotemporal variations of Water Quality (WQ) parameters, including Total Suspended Solids (TSS) and Dissolved Organic Carbon (DOC). Ground-based bio-optical measurements are used to quantify and evaluate the relationship between the remo...

Full description

Saved in:
Bibliographic Details
Main Authors: S.L. Kesav Unnithan, Nagur Cherukuru, Timothy Ingleton, Eric Lehmann, Matt Paget, Yiqing Guo, Nathan Drayson, Gemma Kerrisk
Format: Article
Language:English
Published: Elsevier 2025-12-01
Series:Ecological Informatics
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1574954125002857
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849233609177694208
author S.L. Kesav Unnithan
Nagur Cherukuru
Timothy Ingleton
Eric Lehmann
Matt Paget
Yiqing Guo
Nathan Drayson
Gemma Kerrisk
author_facet S.L. Kesav Unnithan
Nagur Cherukuru
Timothy Ingleton
Eric Lehmann
Matt Paget
Yiqing Guo
Nathan Drayson
Gemma Kerrisk
author_sort S.L. Kesav Unnithan
collection DOAJ
description Satellite remote sensing enables the mapping of large-scale spatiotemporal variations of Water Quality (WQ) parameters, including Total Suspended Solids (TSS) and Dissolved Organic Carbon (DOC). Ground-based bio-optical measurements are used to quantify and evaluate the relationship between the remote sensing reflectance (Rrs) and the in-situ WQ parameters. However, the non-linear relationship is strongly influenced by optically complex coastal regions. We focus on quantifying the non-linear Rrs – WQ relationship using a dense Deep Learning (DL) model trained on an augmented spectral library dataset covering estuarine and off-shore regions along New South Wales coastline, Australia. The spectral library consists of WQ parameters and Inherent Optical Properties (IOP) of water samples, including absorption and backscattering coefficients that drive a radiative transfer model to derive resultant Rrs. We incorporate uncertainty into the WQ−IOP measurements and simulate an augmented spectral library for training a DL model for aquatic Remote Sensing (DL-RS) applications. The model error analysis revealed a) an inherent Mean Absolute Error (MAE) of 7 % for TSS and 22 % for DOC, b) a predictive MAE of 75 % (TSS) and 31 % (DOC) in temperate waters, and c) an in-situ optical closure predictive MAE of 18 % and a Mean Percentage Error of ±5 % upon comparing forward modelled Rrs outputs from DL-RS-derived IOP and WQ parameters with atmospherically corrected Landsat-8 imagery (2013−2022). The DL-RS model demonstrated improved performance and added computational efficiency over existing inversion models. The model thus provides valuable insights into the dynamics of WQ parameters in optically complex waters influenced by diverse land cover catchments.
format Article
id doaj-art-9f9c1e48b0754bd7843865587a0e5ecb
institution Kabale University
issn 1574-9541
language English
publishDate 2025-12-01
publisher Elsevier
record_format Article
series Ecological Informatics
spelling doaj-art-9f9c1e48b0754bd7843865587a0e5ecb2025-08-20T05:05:30ZengElsevierEcological Informatics1574-95412025-12-019010327610.1016/j.ecoinf.2025.103276Mapping total suspended solids (TSS) and dissolved organic carbon (DOC) in complex coastal waters using deep learning enhanced remote sensingS.L. Kesav Unnithan0Nagur Cherukuru1Timothy Ingleton2Eric Lehmann3Matt Paget4Yiqing Guo5Nathan Drayson6Gemma Kerrisk7Commonwealth Scientific and Industrial Research Organisation, Environment, Canberra 2601, Australia; Corresponding author.Commonwealth Scientific and Industrial Research Organisation, Environment, Canberra 2601, AustraliaNew South Wales Department of Climate Change, Energy, the Environment and Water, Sydney 2141, AustraliaCommonwealth Scientific and Industrial Research Organisation, Data61, Canberra 2601, AustraliaCommonwealth Scientific and Industrial Research Organisation, Environment, Canberra 2601, AustraliaCommonwealth Scientific and Industrial Research Organisation, Data61, Canberra 2601, AustraliaCommonwealth Scientific and Industrial Research Organisation, Environment, Canberra 2601, AustraliaCommonwealth Scientific and Industrial Research Organisation, Environment, Canberra 2601, AustraliaSatellite remote sensing enables the mapping of large-scale spatiotemporal variations of Water Quality (WQ) parameters, including Total Suspended Solids (TSS) and Dissolved Organic Carbon (DOC). Ground-based bio-optical measurements are used to quantify and evaluate the relationship between the remote sensing reflectance (Rrs) and the in-situ WQ parameters. However, the non-linear relationship is strongly influenced by optically complex coastal regions. We focus on quantifying the non-linear Rrs – WQ relationship using a dense Deep Learning (DL) model trained on an augmented spectral library dataset covering estuarine and off-shore regions along New South Wales coastline, Australia. The spectral library consists of WQ parameters and Inherent Optical Properties (IOP) of water samples, including absorption and backscattering coefficients that drive a radiative transfer model to derive resultant Rrs. We incorporate uncertainty into the WQ−IOP measurements and simulate an augmented spectral library for training a DL model for aquatic Remote Sensing (DL-RS) applications. The model error analysis revealed a) an inherent Mean Absolute Error (MAE) of 7 % for TSS and 22 % for DOC, b) a predictive MAE of 75 % (TSS) and 31 % (DOC) in temperate waters, and c) an in-situ optical closure predictive MAE of 18 % and a Mean Percentage Error of ±5 % upon comparing forward modelled Rrs outputs from DL-RS-derived IOP and WQ parameters with atmospherically corrected Landsat-8 imagery (2013−2022). The DL-RS model demonstrated improved performance and added computational efficiency over existing inversion models. The model thus provides valuable insights into the dynamics of WQ parameters in optically complex waters influenced by diverse land cover catchments.http://www.sciencedirect.com/science/article/pii/S1574954125002857Large-scale water quality mappingHigh-resolution optical remote sensingDeep learning
spellingShingle S.L. Kesav Unnithan
Nagur Cherukuru
Timothy Ingleton
Eric Lehmann
Matt Paget
Yiqing Guo
Nathan Drayson
Gemma Kerrisk
Mapping total suspended solids (TSS) and dissolved organic carbon (DOC) in complex coastal waters using deep learning enhanced remote sensing
Ecological Informatics
Large-scale water quality mapping
High-resolution optical remote sensing
Deep learning
title Mapping total suspended solids (TSS) and dissolved organic carbon (DOC) in complex coastal waters using deep learning enhanced remote sensing
title_full Mapping total suspended solids (TSS) and dissolved organic carbon (DOC) in complex coastal waters using deep learning enhanced remote sensing
title_fullStr Mapping total suspended solids (TSS) and dissolved organic carbon (DOC) in complex coastal waters using deep learning enhanced remote sensing
title_full_unstemmed Mapping total suspended solids (TSS) and dissolved organic carbon (DOC) in complex coastal waters using deep learning enhanced remote sensing
title_short Mapping total suspended solids (TSS) and dissolved organic carbon (DOC) in complex coastal waters using deep learning enhanced remote sensing
title_sort mapping total suspended solids tss and dissolved organic carbon doc in complex coastal waters using deep learning enhanced remote sensing
topic Large-scale water quality mapping
High-resolution optical remote sensing
Deep learning
url http://www.sciencedirect.com/science/article/pii/S1574954125002857
work_keys_str_mv AT slkesavunnithan mappingtotalsuspendedsolidstssanddissolvedorganiccarbondocincomplexcoastalwatersusingdeeplearningenhancedremotesensing
AT nagurcherukuru mappingtotalsuspendedsolidstssanddissolvedorganiccarbondocincomplexcoastalwatersusingdeeplearningenhancedremotesensing
AT timothyingleton mappingtotalsuspendedsolidstssanddissolvedorganiccarbondocincomplexcoastalwatersusingdeeplearningenhancedremotesensing
AT ericlehmann mappingtotalsuspendedsolidstssanddissolvedorganiccarbondocincomplexcoastalwatersusingdeeplearningenhancedremotesensing
AT mattpaget mappingtotalsuspendedsolidstssanddissolvedorganiccarbondocincomplexcoastalwatersusingdeeplearningenhancedremotesensing
AT yiqingguo mappingtotalsuspendedsolidstssanddissolvedorganiccarbondocincomplexcoastalwatersusingdeeplearningenhancedremotesensing
AT nathandrayson mappingtotalsuspendedsolidstssanddissolvedorganiccarbondocincomplexcoastalwatersusingdeeplearningenhancedremotesensing
AT gemmakerrisk mappingtotalsuspendedsolidstssanddissolvedorganiccarbondocincomplexcoastalwatersusingdeeplearningenhancedremotesensing