| In this paper, DOM during the high flow and normal flow period of Daliao River and its estuary were studied. The excitation/emission matrix spectroscopy combined with PARAFAC, calculation of optical indices and principal component analysis were applied to discuss the effects of seasons, hydrologic conditions, environmental processes on dynamics of DOM and analyze the characteristics of DOM temporally and spatially. Stability of PARAFAC model was verified under our water storage condition. Field measurements combined with simulation of UVA illumination experiments in the laboratory provided the insights into impacts of terrestrial input, microbial processes, photo degradation and human activities on fluorescence intensity. The interactions between the common FDOM component (tryptophan) and NOM under UVA illumination was also discussed by developing the model compound experiments. The research achievements of this paper include the following three aspects:(1) The surface water samples in the two sampling seasons in Daliao River and its estuary were analyzed. The comparisons between two PARAFAC models of the FDOM components indicated that the dynamics of some terrestrial HS-like components (C3-C5 or P2-P4) were controlled by the same environmental processes. In addition, the results of UVA illumination demonstrated that solar ultraviolet illumination maybe the second important process influencing the FDOM components except for the main environmental processes. This finding emphasized that PARAFAC models combined with UVA illumination experiments were useful to investigate the dynamics of terrestrial HS in river-estuary systems. In addition, photoammonification of DOM under UVA illumination was preliminarily verified in our experiments.(2) The comparison of PARAFAC models showed that spectral shape and components were the same and produced no new types of FDOM components after immediate and 9-month storage measurements during high flow period and normal river flow period in Daliao River and its estuary, while the fluorescence intensities were different.There were no obvious changes of DOC, which illustrated that aerobic respiration was very weak. For the high flow period, HIX and BIX did not change (p>0.05, t-test) after the storage. Therefore, these two indices were stable over the storage, and the post-storage fingerprints could still be applicable to trace back the initial humification status (based on HIX) or the initial contribution of biological processes (based on BIX) of the initial samples. Moreover, absorbance decreased on average after the storage.According to the optical indices, S275-295 and SUVA254 related to molecular weight and mean aromaticity of DOM were calculated, respectively. The results showed that the molecular weight would decrease after 9-month storage. However, normal river flow period samples after 9-month storage showed an increase in HIX and the BIX remained unchanged; absorbance of samples and SUVA254 didn’t change but S275-295 decreased, which indicated the increase in molecular weight. Our experiments confirmed that the 9-month storage samples kept in dark condition at 4℃ had a stable PARAFAC model after filtration through 0.45 μm film.(3) Simulation experiments of UVA illumination were developed in the laboratory for model compound of Try-NOM system. As control experiments, tryptophan and NOM were illuminated respectively. EEMs combined with PARAFAC were applied to identify FDOM components, analyze the changes of fluorescence signal and investigate the interactions between tryptophan and NOM under UVA illumination. Two FDOM components were identified by PARAFAC for the set of tryptophan solutions:tryptophan C1 and HS-like C2; Two HS-like components HS-1 and HS-2 were referred to for NOM solutions. Two PARAFAC modeling approaches were employed to the mixture of tryptophan and NOM.On the one hand,PARAFAC without a prior assumption on the components number and spectral features was referred to as un-supervised PARAFAC.This approach identified three FDOM components:C1, C2 and HS-3 (a new HS-like component different from HS-1 and HS-2 mentioned above).On the other hand,the spectral features of pre-identified FDOM components in the Try solutions and in the NOM solutions were combined to form a composite model named supervised PARAFAC modeling.The components identified contained C1,C2,HS-1,HS-2 and no new FDOM components were produced in the mixture.The relationships between fluorescence intensity and illumination time indicated that NOM facilitated the photo-bleaching and photo-production of the new FDOM components and that tryptophan promoted the generation of HS-like fluorescence signal. Therefore, tracing the sources of DOM and the dynamics of HS-like material should take tryptophan-like fluorescence into account as well. |
PDF Full Text Request