Grain size analysis in mountain rivers using UAV photogrammetry: proposal and validation of a D 50 homologation methodology

This study presents and validates a novel methodology for granulometric characterization in mountain rivers with gravel beds using unmanned aerial vehicles (UAVs). Traditional grain-size assessment methods, such as the Wolman count, offer high accuracy but are labor-intensive and spatially limited. In contrast, remote sensing techniques such as laser scanning cover larger areas but incur higher costs. This work proposes a cost-effective approach that combines UAV photogrammetry with Basegrain software for grain-size distribution analysis. A homologation equation was developed to correct the overestimation of median grain size (D50) when images are captured at suboptimal flight altitudes, resulting in coarser ground sampling distances (GSD). The method was applied in the Teapa River, Chiapas, Mexico, and compared against traditional measurements obtained through the Wolman method. Results show that the proposed equation reduces estimation errors and produces D50 values consistent with ground-based data, with a mean absolute error of 6.78 mm and a mean error of 0.46 mm. This confirms that UAV photogrammetry, when properly calibrated, can serve as a reliable and scalable alternative for granulometric mapping in complex fluvial environments.