Study region: The Amu Darya River (ADR) basin in Central Asia.Study focus: To understand the spatiotemporal patterns and underlying driving mechanisms of river salinization in arid environments, this study gathered 50 years (1970-2019) of water chemistry data from 12 locations along the ADR. The variations in discharge and salinity were assessed by a linear regression model and violin plot. The salinity-discharge relationships were evaluated by a general hyperbolic model and Spearman's rank correlation coefficient. Random forest models were also constructed to identify the predominant drivers of river water saliniza-tion. Finally, a conceptual model of river water salinization was constructed.New hydrological insights for the region: The water salinity (S) in the upper stream of the ADR was 541-635 mg/L. Salinity showed an increasing trend along the river course, reaching 751-1560 mg/L downstream. In the downstream, the river salinity before the 1990 s (751-1128 mg/L) was slightly lower than that after the 1990 s (983-1560 mg/L). Generally, water salinity was notably correlated with river discharge (Q) in upstream, exhibiting a relationship of S= 17,497Q 0.62, p < 0.05, before the 1990 s. Interannual variation in river salinity is mainly controlled by secondary salinization, and intra-annual variation is controlled by river flow. From upstream to down-stream, the controlling salinization process changes from primary salinization to secondary salinization. Specifically, secondary salinization has accelerated due to intensified agricultural activities in recent years.