<p>This study systematically investigates the spatiotemporal variation patterns and pollution driving mechanisms of water quality in transboundary rivers, utilizing monthly monitoring data (2020–2023) from 63 cross-boundary sections in the Haihe River Basin. The results demonstrate: (1) Significant spatiotemporal heterogeneity in water quality, with the annual proportion of "good/excellent" water bodies increasing by 17%, yet decreasing by16.8–18.4% during flood seasons (June–September) due to combined urban/agricultural non-point source pollution. (2) Cluster analysis categorized sections into three types, and established flood/non-flood season-specific WQImin models (R² ≥ 0.789), reducing monitoring indicators to 6–9 parameters while maintaining accuracy. Furthermore, to address practical governance needs, we developed a unified model and a unified parameters model. (3) The APCS-MLR(Absolute Principal Component Score-Multiple Linear Regression) was employed to identify the dominant influencing factors and quantify their impact magnitude on the key parameters of each model. (4) The proposed dual-mode monitoring network ("classified models–unified parameters") and dynamic traceability management strategy provide a scientific foundation for transboundary river governance. Building upon a comprehensive understanding of transboundary river water quality dynamics, this study developed WQImin for distinct cross-boundary sections to reduce monitoring and management costs. By prioritizing key parameters, we systematically identified dominant influencing factors across models and hydrological periods using APCS-MLR. This dual approach—combining model simplification with source apportionment—provides actionable insights for precision management of transboundary water environments.</p>