文献类型：期刊文献

英文题名：Vis-NIR Spectroscopy Characteristics of Wetland Soils with Different Water Contents and Machine Learning Models for Carbon and Nitrogen Content

作者：Qu, Keying[1,2,3,4] Nie, Leichao[1,2,3,4,5] Cui, Lijuan[1,2,3,4] Li, Huazhe[1,2,3,4] Xiong, Mingshuo[1,2,3,4] Zhai, Xiajie[1,2,3,4] Zhao, Xinsheng[1,2,3,4] Wang, Jinzhi[1,2,3,4] Lei, Yinru[1,2,3,4] Li, Wei[1,2,3,4]

第一作者：Qu, Keying

通信作者：Li, W[1];Li, W[2];Li, W[3];Li, W[4]

机构：[1]State Key Lab Wetland Conservat & Restorat, Beijing 100091, Peoples R China;[2]Chinese Acad Forestry, Inst Wetland Res, Beijing 100091, Peoples R China;[3]Key Lab Wetland Serv & Restorat, Beijing 100091, Peoples R China;[4]Chahannaoer Wetland Ecosyst Res Stn, Ulanqab 013400, Peoples R China;[5]Wuhan Forestry Workstn, Wuhan 430023, Peoples R China

年份：2025

卷号：6

期号：4

外文期刊名：ECOLOGIES

收录：Scopus(收录号：2-s2.0-105026774537);WOS:【ESCI(收录号:WOS:001646380100001)】；

基金：This work was supported by China's Special Fund for Basic Scientific Research Business of Central Public Research Institutes (CAFYBB2021ZB003) and the National Key R&D Program of China (2017YFC0506200).

语种：英文

外文关键词：Vis-NIR spectroscopy reflectance; soil moisture; machine learning models; soil nutrients; random forest

摘要：Soil nutrient detection in wetlands is critical for rapidly and effectively managing these ecosystems. Our objective was to provide a methodological framework for identifying optimal data processing methods and machine learning model for predicting soil organic carbon (SOC) and total nitrogen (TN) content using Vis-NIR spectroscopy, under the confounding influence of varying soil moisture. Soil samples (474) were collected from the Shaanxi Yellow River Wetland Provincial Nature Reserve with five moisture levels (0, 5, 10, 20, and 30%). Using a Vis-NIR spectroscopy system (ASD FS4 spectrometer), soil organic carbon (SOC) and total nitrogen (TN) were detected within the 350-2500 nm spectral range. Machine learning models were established using the Random Forest model (RF), eXtreme Gradient Boosting (XGBoost), and Partial Least Squares Regression (PLSR). The results indicated: (1) spectral reflectance values increased as soil moisture content decreased, with the 0% moisture model being consistently more accurate; (2) models for SOC and TN on first-derivative spectra had higher accuracy; and (3) the RF exhibited higher inversion accuracy and stability (R2 = 0.30-0.69). (4) The SHAP analysis confirmed 1865 nm and 1419 nm as the most contributory bands for SOC and TN prediction respectively, validating the RF model's spectral interpretation capability.