Various sources associated with mining activities adversely affect water quality in aquatic ecosystems. This study aimed to estimate the sources of dissolved trace elements (DTEs: Al, As, B, Ba, Cd, Co, Cr, Cs, Cu, Fe, Li, Mn, Mo, Ni, Pb, Rb, Sb, U, and Zn) affecting water chemistry in the upper Nakdong River (UNR) catchment and determine the hydrological processes associated with the inflow from anthropogenic sources. Hierarchical cluster analysis (HCA) and principal component analysis (PCA) were employed to estimate main sources affecting water chemistry and its spatiotemporal variability in the four tributaries and main channel of the UNR. Water samples were classified into four groups based on their chemistry. The spatiotemporal variations of DTEs were driven by the presence of anthropogenic sources (including smelters, AMD, mining-related sources), while running toward downstream. The correlations among PCA (and HCA), Zn as smelter impact, and deuterium excess (d-excess = δ2H – 8 × δ18O) indicated that smelters had a strong influence on water chemistry in the main channel after passing through the smelter. In the dry season, d-excess < ~12 ‰ was observed with increasing Zn levels and factor score for PC2, especially samples influenced by DTEs from smelter, indicating that groundwater affected by the smelters recharged mainly by summer rainwater impacted the spatiotemporal variability of DTEs. This multivariate study suggests that the spatiotemporal variation in DTEs is associated with the complex contributions of natural and anthropogenic sources to water chemistry, which are strongly influenced by hydrological processes and seasonality.