Detection of dissolved organic matter in saline–alkali soils using synchronous fluorescence spectroscopy and principal component analysis

PCA factor scores vs. spectral wavelength. These plots indicate which spectral shapes are associated with the variance in the overall signal, which is used to identify the fluorophores that contribute to the variance in the soil DOM. PCA has been shown to be an effective technique for identifying th...

Full description

Saved in:
Bibliographic Details
Published inSpectrochimica acta. Part A, Molecular and biomolecular spectroscopy Vol. 104; pp. 280 - 286
Main Authors Guo, Xu-jing, Yuan, Dong-hai, Jiang, Jin-yuan, Zhang, Hui, Deng, Yu
Format Journal Article
LanguageEnglish
Published England Elsevier B.V 01.03.2013
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:PCA factor scores vs. spectral wavelength. These plots indicate which spectral shapes are associated with the variance in the overall signal, which is used to identify the fluorophores that contribute to the variance in the soil DOM. PCA has been shown to be an effective technique for identifying the constituent components of synchronous fluorescence spectra of DOM. Six fluorescence peaks, T1 (275nm), T2 (283nm), T3 (291nm), H1 (318nm), H2 (345nm) and H3 (362nm), can be identified by PCA. [Display omitted] ► The analysis of PCA yielded 2 PCs that account for 97% of the variance. ► Two dominant fluorophores can be identified by PC1 and PC2 respectively. ► Five main clusters can be identified by HCA in 32 soils. ► Land use and agricultural irrigation are important factors of humification process. Dissolved organic matter (DOM) is attributed a key role in soil major biogeochemical processes. Its molecular characteristics can reflect both its source and its biogeochemical history. Saline–alkali soil samples were collected from a field in desert and semi-desert region. DOM was analyzed by synchronous fluorescence spectroscopy. Principal Components Analysis (PCA) of synchronous fluorescence spectra was used to detect and describe changes in the DOM in different soil profiles. The analysis of PCA yielded 2 PCs that account for 97% of the variance. DOM is characterized by a dominant tyrosine-like fluorophore in the Region I (PC1>0.9, PC2<0.3), including soils Ke02, Ke03, Ke04, Rh02, Rh03, Rh04, Sa04 and Sc04. The region II including Kf02, Sa01, Yb01, Yzd01 and Yzd02 obtain higher PC2 (>0.9) and lower PC1 (<0.3) loading, which is dominated humic-like fluorophore. Hierarchical cluster analysis (HCA) identified five main clusters. The results of hierarchical cluster analysis (HCA) were consistent with the PCA. The ratio of fluorescence intensity of humic substances at long wavelengths and short wavelengths was used to indicate the humification of organic matter. Two ratios (I345/318 and I362/318.) can be used to indicate the humification of DOM. The humification degree of the remaining soils has no systematic trend because of land use, agricultural irrigation, soil erosion and salinization. Soils obtained a relatively high humification degree in the region II and a relatively low humification degree in the region I.
ISSN:1386-1425
1873-3557
DOI:10.1016/j.saa.2012.11.081