Feasibility of pollution source identification by hydrochemical characteristics method in urban water bodies

• Ion-based source tracing was most accurate in dry and dry-to-wet seasons with low flow. • Fluorescence index enhanced source identification accuracy in urban watershed analysis. • Agricultural non-point sources dominated TN pollution in the Longjiang River’s upstream. • Seasonal dynamics were observed: PS dominated in dry, NPS in wet season. The hydrochemical characteristic ion-based quantitative analysis method is an emerging approach for identifying pollution sources, but its applicability in urban rivers subject to strong external disturbances remains uncertain. In this study, hydrochemical characteristic ion analysis was combined with fluorescence indices from three-dimensional fluorescence spectroscopy over a full hydrological year to trace point source (PS) and non-point source (NPS) pollution in the Longjiang River, an independent urban river discharging into the sea. The performance of the ion-based method under different hydrological conditions was also evaluated. Results showed that upstream pollution was dominated by NPS-derived total nitrogen (TN) from agriculture, while the relative contributions of PS and NPS to the mainstream varied seasonally. Source apportionment accuracy based on hydrochemical ions and fluorescence indices was high in both the dry and dry-to-wet seasons (agreement or equivalence, accuracy = 1 or 0.5), and poor in the wet season (disagreement, accuracy = −1). These findings suggest that the ion-based method is more suitable for source tracing under stable, low-flow conditions, whereas its reliability decreases in the wet season due to enhanced hydrological and anthropogenic disturbances. This study provides theoretical guidance for rapid water quality monitoring and pollution source identification in urban water bodies.