Purpose: Passive particle traps (PTs) are cost-efficient and easy to operate tools for the time integrative monitoring of particulate contaminants in surface fresh-waters. Box-type PTs are designed as boxes through which the river flow is slowed down in order to collect Suspended Particulate Matter (SPM). However, a particle-size distribution bias in collection suggests that the finest particles pass through PTs without settling. The aim of this study is to quantify SPM trapping efficiency as a function of particle size, and to develop a predictive model. Methods: We designed laboratory experiments using box-type PTs. The trap was either submerged in a SPM-laden flow (in-flow experiment); or directly supplied with water and SPM with a pipe connected to its inlet (in-line experiments). SPM concentration, particle-size distributions (PSD), flow velocity were controlled and recorded. Results: In-line experiment results confirmed that coarser particles are more efficiently trapped than smaller ones due to their higher settling velocity. The particle-size trapping efficiency was regressed against PT dimensions and incoming flow velocity. The application of these results to in-flow and field experiments confirmed the validity of our predictive model. Conclusion: This study highlighted how crucial it is to deploy particle traps in an area with low current velocities, in order to limit the granulometric bias. We developed a predictive model that will be an asset to quantify particle-size distribution bias and to better understand potential particulate contaminants concentrations shifts. This work represents a significant step forward to a thorough use of PTs.