Abstract: Internal and external seepage in long-distance water conveyance projects can lead to potential engineering risks and water quality hazards. Therefore, rapid and accurate monitoring and warning of water mass mixing events have become an emerging challenge. Conventional detection algorithms are often criticized for high false positive rates and low true positive rates. This paper presents a new water mass mixing detection method based on a multi-parameter matching algorithm, which integrates the Pearson correlation coefficient and Mahalanobis distance for water identification. The performance of the proposed method was evaluated using data from a water mixed experiment, and the Pearson correlation coefficient Euclidean distance-based method (PE) was compared with the Pearson correlation coefficient Mahalanobis distance-based method (PM). The results show that for mixing events characterized by significant differences in water quality indicators, both the PE and PM methods correctly detect 100% of all water mixing events, yielding a 0 false alarm rate. For mixing events with slight differences in water quality indicators, both methods correctly detect 95% of the events, and the corresponding false alarm rates are 1.92% and 0, respectively. Therefore, compared with the PE method, the PM method is less affected by the difference between external water and main canal water, offering a wider range for optimal threshold selection. The PM algorithm offers higher detection performance than the PE method and a lower false alarm rate, along with better stability and an greater ability to distinguish between water quality fluctuations and mixing-induced fluctuations.