Integrating fractional-order derivatives of soil and leaf hyperspectral reflectance for improved estimation of mangrove soil organic carbon

Mangrove ecosystems are vital for carbon sequestration and coastal protection, yet accurate estimation of soil organic carbon (SOC) using remote sensing remains challenging due to spectral interference caused by dynamic vegetation cover. This study presents a novel framework integrating fractional-o...

Full description

Saved in:
Bibliographic Details
Main Authors: Yibo Luo, Chunlin Li, Jinhong Huang, Chengcheng Dong, Junjie Wang
Format: Article
Language:English
Published: Elsevier 2025-06-01
Series:Geoderma
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S0016706125001624
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850150744775196672
author Yibo Luo
Chunlin Li
Jinhong Huang
Chengcheng Dong
Junjie Wang
author_facet Yibo Luo
Chunlin Li
Jinhong Huang
Chengcheng Dong
Junjie Wang
author_sort Yibo Luo
collection DOAJ
description Mangrove ecosystems are vital for carbon sequestration and coastal protection, yet accurate estimation of soil organic carbon (SOC) using remote sensing remains challenging due to spectral interference caused by dynamic vegetation cover. This study presents a novel framework integrating fractional-order derivative (FOD) techniques with machine learning algorithms for SOC estimation in mangrove wetlands. A total of 201 soil samples were collected from five mangrove wetlands in southern China. FOD was applied to both soil and leaf hyperspectral reflectance to amplify subtle spectral variations typically overlooked by conventional approaches. SOC-sensitive wavelengths were identified using the SHAP-XGBoost (Shapley Additive Explanations-Extreme Gradient Boosting) method. A total of 363 modeling strategies were constructed using Random Forest, XGBoost, and CatBoost (Categorical Boosting) algorithms across 11 vegetation cover levels (0–100 %) and 11 fractional orders (0–2 at 0.2 intervals). Results indicate that fractional orders between 0.8 and 1.4 consistently yielded superior performance. The CatBoost model under 10 % vegetation cover and a fractional order of 1.2 achieved the highest accuracy (R2 = 0.730, RMSE = 0.858 %). Incorporating key soil and terrain variables (e.g. soil iron, clay content, pH, salinity, redox potential, and elevation) into the spectra-based SOC estimation model significantly enhanced prediction accuracy, highlighting the complementary roles of spectral signals, soil characteristics, and topographic features in SOC modeling. This framework holds the potential for advancing blue carbon accounting and supporting sustainable mangrove conservation and management under changing environmental conditions.
format Article
id doaj-art-9ae9b2ca9eac46e6bc1f2d9b068c2a73
institution OA Journals
issn 1872-6259
language English
publishDate 2025-06-01
publisher Elsevier
record_format Article
series Geoderma
spelling doaj-art-9ae9b2ca9eac46e6bc1f2d9b068c2a732025-08-20T02:26:27ZengElsevierGeoderma1872-62592025-06-0145811732410.1016/j.geoderma.2025.117324Integrating fractional-order derivatives of soil and leaf hyperspectral reflectance for improved estimation of mangrove soil organic carbonYibo Luo0Chunlin Li1Jinhong Huang2Chengcheng Dong3Junjie Wang4College of Life Sciences and Oceanography, Shenzhen University, 518060 Shenzhen, ChinaCollege of Life Sciences and Oceanography, Shenzhen University, 518060 Shenzhen, ChinaCollege of Life Sciences and Oceanography, Shenzhen University, 518060 Shenzhen, ChinaSchool of Urban Planning and Design, Shenzhen Graduate School, Peking University, 518071 Shenzhen, ChinaCollege of Life Sciences and Oceanography, Shenzhen University, 518060 Shenzhen, China; MNR Key Laboratory for Geo-Environmental Monitoring of Great Bay Area & Guangdong Key Laboratory of Urban Informatics, Shenzhen University, 518060 Shenzhen, China; Corresponding author at: College of Life Sciences and Oceanography, Shenzhen University, 518060 Shenzhen, China.Mangrove ecosystems are vital for carbon sequestration and coastal protection, yet accurate estimation of soil organic carbon (SOC) using remote sensing remains challenging due to spectral interference caused by dynamic vegetation cover. This study presents a novel framework integrating fractional-order derivative (FOD) techniques with machine learning algorithms for SOC estimation in mangrove wetlands. A total of 201 soil samples were collected from five mangrove wetlands in southern China. FOD was applied to both soil and leaf hyperspectral reflectance to amplify subtle spectral variations typically overlooked by conventional approaches. SOC-sensitive wavelengths were identified using the SHAP-XGBoost (Shapley Additive Explanations-Extreme Gradient Boosting) method. A total of 363 modeling strategies were constructed using Random Forest, XGBoost, and CatBoost (Categorical Boosting) algorithms across 11 vegetation cover levels (0–100 %) and 11 fractional orders (0–2 at 0.2 intervals). Results indicate that fractional orders between 0.8 and 1.4 consistently yielded superior performance. The CatBoost model under 10 % vegetation cover and a fractional order of 1.2 achieved the highest accuracy (R2 = 0.730, RMSE = 0.858 %). Incorporating key soil and terrain variables (e.g. soil iron, clay content, pH, salinity, redox potential, and elevation) into the spectra-based SOC estimation model significantly enhanced prediction accuracy, highlighting the complementary roles of spectral signals, soil characteristics, and topographic features in SOC modeling. This framework holds the potential for advancing blue carbon accounting and supporting sustainable mangrove conservation and management under changing environmental conditions.http://www.sciencedirect.com/science/article/pii/S0016706125001624Hyperspectral reflectanceFractional-order derivativeMangrove ecosystemSoil organic carbonMachine learning
spellingShingle Yibo Luo
Chunlin Li
Jinhong Huang
Chengcheng Dong
Junjie Wang
Integrating fractional-order derivatives of soil and leaf hyperspectral reflectance for improved estimation of mangrove soil organic carbon
Geoderma
Hyperspectral reflectance
Fractional-order derivative
Mangrove ecosystem
Soil organic carbon
Machine learning
title Integrating fractional-order derivatives of soil and leaf hyperspectral reflectance for improved estimation of mangrove soil organic carbon
title_full Integrating fractional-order derivatives of soil and leaf hyperspectral reflectance for improved estimation of mangrove soil organic carbon
title_fullStr Integrating fractional-order derivatives of soil and leaf hyperspectral reflectance for improved estimation of mangrove soil organic carbon
title_full_unstemmed Integrating fractional-order derivatives of soil and leaf hyperspectral reflectance for improved estimation of mangrove soil organic carbon
title_short Integrating fractional-order derivatives of soil and leaf hyperspectral reflectance for improved estimation of mangrove soil organic carbon
title_sort integrating fractional order derivatives of soil and leaf hyperspectral reflectance for improved estimation of mangrove soil organic carbon
topic Hyperspectral reflectance
Fractional-order derivative
Mangrove ecosystem
Soil organic carbon
Machine learning
url http://www.sciencedirect.com/science/article/pii/S0016706125001624
work_keys_str_mv AT yiboluo integratingfractionalorderderivativesofsoilandleafhyperspectralreflectanceforimprovedestimationofmangrovesoilorganiccarbon
AT chunlinli integratingfractionalorderderivativesofsoilandleafhyperspectralreflectanceforimprovedestimationofmangrovesoilorganiccarbon
AT jinhonghuang integratingfractionalorderderivativesofsoilandleafhyperspectralreflectanceforimprovedestimationofmangrovesoilorganiccarbon
AT chengchengdong integratingfractionalorderderivativesofsoilandleafhyperspectralreflectanceforimprovedestimationofmangrovesoilorganiccarbon
AT junjiewang integratingfractionalorderderivativesofsoilandleafhyperspectralreflectanceforimprovedestimationofmangrovesoilorganiccarbon