Abstract Total Suspended Solids (TSS) significantly degrade water quality by reducing light penetration and oxygen availability, while facilitating the transport of toxic contaminants. Managing TSS in watersheds requires an understanding of both hydrological connectivity and pollutant dynamics; however, these efforts are significantly constrained by data scarcity, particularly in extensive or remote watersheds in developing countries. This study develops a network‐based advection‐reaction model to simulate TSS transport across the Canal del Dique watershed. The watershed is represented as a directed graph, where rivers and streams form the edges of the network, and confluence points serve as nodes. To address the challenge of data scarcity, machine learning techniques are employed to estimate missing TSS values at unmonitored locations, and an optimization framework is implemented to determine the most effective TSS mitigation strategies. Results highlight the role of hydrological connectivity in TSS transport, with the model revealing that at low mitigation levels, interventions should prioritize high‐TSS nodes. As mitigation resources increase, interventions shift toward pollutant source nodes and less connected areas, preventing downstream pollutant accumulation. This study demonstrates that highly connected nodes, although crucial for flow, are less effective targets for pollution control. The proposed methodology offers a novel, data‐driven approach for optimizing TSS mitigation strategies, providing a scientifically grounded tool for improving water quality management. By prioritizing resource allocation in critical areas, this work enhances the efficiency of watershed management and supports sustainable water resource policies, especially in data‐limited regions.
Pérez et al. (Sun,) studied this question.