Abstract
Photogrammetry has become an effective tool for generating high-resolution digital terrain models (DTMs), particularly in mountainous and hard-to-access regions where conventional topographic methods face significant limitations. Its application is essential for geotechnical risk management and the accurate characterization of mass-movement processes in Andean environments. This study evaluates the accuracy and applicability of UAV-based photogrammetry in comparison with the official elevation models produced by the SIGTIERRASproject, a nationwide cadastral program in Ecuador that provides data with a spatial resolution of 4 m/pixel.. Although these institutional models offer extensive territorial coverage, their resolution constrains detailed volumetric analyses at the local scale. A DJI Matrice 30 unmanned aerial vehicle equipped with real-time kinematic (RTK) positioning was deployed, enabling centimeter-level accuracy without the need for extensive ground control points. The case studyfocuses on a landslide that occurred in the Yuquín sector (Pimampiro, Imbabura, Ecuador), an area characterized by steep slopes and limited accessibility. The research tested the hypothesis that low-resolution official models systematically overestimate displaced volumes compared to high-precision UAV-derived photogrammetric models. A post-event DTM with a resolution of 0.5 m/pixel was generated and compared with the pre-event SIGTIERRAS surface. Elevation differences yielded an estimated displaced volume of 232,735 ± 6 m³, supported by RTK positional corrections and internal accuracy assessments (RMSE ≤ 6 cm). The results indicate that institutional low resolution datasets do not introduce significant systematic errors in volumetric estimation and remain suitable for regional-scale assessments. However, UAV-based photogrammetry proves to be a precise, cost-effective, and highly compatible tool for local scale geotechnical studies and detailed risk analysis in highly vulnerable Andean regions.