Abstract: With the development of urbanization and industrialization, more and more chemicals were produced in large quantities and widely used. As a result, they entered the water environment through sewage discharge and agricultural drainage. Currently, the analytical methods of contaminants in water include three main strategies, including target screening, suspect analysis and nontarget screening. The number of compounds identified using target screening methods has increased to hundreds, which is far from meeting the current demand for emerging contaminants (ECs) detection in water environment. To comprehensively screen potential emerging contaminants in the Yangtze River in Jiangsu Province, a nontarget screening strategy combined spectral library search with a molecular network was proposed in Jiangsu-Anhui provincial boundary cross section and 22 drinking water sources in Jiangsu section of Yangtze River. Under this strategy, through spectral library screening using a in-house library, a total of 118 emerging contaminants were identified by accurate mass, isotope distribution and MS/MS fragment ions. These contaminants involved pesticides, herbicides, fungicides, drugs, antibiotics, persistent organic pollutants, flame retardants, plasticizers, indicated active agents, chemical intermediates, and other categories. Since traditional targeted screening relies on commercial standards and screening lists, nontarget screening is a more favorable approach to study ECs and their transformation products in water. Therefore, unannotated chemical substances with similar structures to known contaminants were identified through molecular network technology based on the premise that structurally similar compounds have similar fragment ions. Through using the Global Natural Products Society molecular network (GNPS) to extract molecular families, another contaminant with similar toxicity to amantadine, N-methyl-1-adamantaneamine was found. Furthermore, twenty-three contaminants were confirmed through commercial standard, and 14 of them were detected at a rate of 100%. The quantitative analysis revealed the prevalence of perfluorinated compounds, pharmaceuticals, and personal care products in water bodies, and the total concentrations of the 23 contaminants in each sample ranged from 6.46×10² ng/L to 1.72×10³ ng/L. Seven drugs and personal care products were detected. Amantadine was detected at the highest concentration among all locations, and perfluorooctanoic acid (PFOA), amantadine and 2,4-dinitrophenol were detected with the highest mean concentration levels in each location. The nontarget screening strategy using spectral library search and molecular network enabled high-throughput identification of contaminants in water without prior information and standards, in which the structures of unannotated characteristic peaks were inferred from annotated substances with spectral libraries. This strategy will accelerate the systematic discovery of emerging and unknown contaminants in the aqueous environment.