Automated Identification and Characterization of Compound Meander Loops

Accurate identification and characterization of meander loops are essential for understanding river evolution, managing water resources, planning hydraulic projects, and preventing geological disasters. Traditional methods for identifying river meanders rely on detecting inflection points, where channel curvature reverses, or measuring directional changes at fixed intervals along the channel. The former approach lacks reproducibility, while the latter requires careful interval selection and intervals complicated procedures. Consequently, these methods face limitation when applied to more complex channel geometries, such as asymmetrical or compound meanders. This study aimed to develop a novel method for automatically identifying and characterizing compound meander loops using channel centerline data, and primarily through measurements of three channel planform parameters: local maximum sinuosity (LMS), maximum rotation angle (MRA), and simple subloop numbers. The key technical steps in this method include (1) detecting bends with LMS value exceeding a defined threshold using a top-down iterative search algorithm, (2) identifying meander loops based on MRA, (3) identifying compound meander loops using subloop counts and neck length, and (4) measuring the geometric parameters of compound meander loops. The proposed method was tested on the Yavari, Tarauaca, Purus, and Jurua rivers in the Amazon Basin, which are characterized by high water and sediment discharge and are among the world’s fastest-migrating meandering rivers. Results indicate that this method provides a simple yet efficient approach for identifying and characterizing compound meander loops in complex river channels. Additionally, it offers a potential solution for detecting loops in other types of linear features, such as roads, contour lines, and coastlines.