Analysis of the Relationship Between Dissolved Organic Matter (DOM) and Watershed Land-use Based on Three-dimensional Fluorescence-Parallel Factor (EEM-PARAFAC) Analysis

Abstract

Dissolved organic matter (DOM) is a characteristic index of water quality, and reflects many factors, such as the economic development and protection policies of watershed. In this study, surface water samples were collected from four watersheds with different levels of urbanization in Ningbo. The DOM was analyzed using an excitation-emission matrix combined with parallel factor analysis (EEM-PARAFAC) to explore the relationship between land-use and DOM. The results show that the urbanization level affected both the amount and the composition of the DOM in the studied watersheds. The concentrations of DOM evaluated by dissolved organic carbon (DOC) in urban areas (DOC=3.18 mg·L-1) and an urban-rural combined area (DOC=7.45 mg·L-1) were much higher than those in rural areas with low urbanization (DOC between 2.16 and 2.62 mg·L-1, ANOVA, P<0.001). A total of seven PARAFAC components were identified in the studied watersheds, mainly including humic-like and protein-like substances. In the highly urbanized area, DOM was mainly composed of humic-like substances, with a proportion of 61.3%. However, the water samples from the urban-rural combined area exhibited a high proportion (59.4%) of protein-like substances, indicating a strong influence of sewage and industrial discharge. In contrast, although the DOM amounts in rural areas were relatively low, the proportions of humus-like substances were high, ranging from 63.6% to 65.7%. Agricultural non-point sources were the main contributor to DOM in these areas. Moreover, the results suggest that the urbanization process could intensify the damage to the surface waters. At the initial stage of urbanization (i.e., urban-rural combined area), contaminants are mainly discharged from sewage and industrial sources; when urbanization reaches a certain level, e.g., with a well-constructed sewage collection system, water contaminants originate more from surface runoff rather than sewage. The results of this study suggest that the EEM-PARAFAC technique can provide semi-quantitative source tracking of surface water, as well as an inexpensive and effective tool for policy makers to overcome the insensitivity of general water quality indices.