Error propagation of digital elevation model obtained by low cost UAVs for flood models

Abstract

River floods represent one of the most important natural hazards in the Andean cities due to their recurrence and economic impacts. Adequate flood risk management requires information on the probability of occurrence and depth of flood for events in different periods of recurrence. The objective of the article is to evaluate the potential use of Digital Elevation Models (DEM) produced with low cost aerial photogrammetry with unmanned aerial vehicles (UAV), using a section of the Santa Bárbara River. For the analysis, a flood scenario with a 50-year return period and two Hec-ras flood models and an agent-based model (ABM) will be used. The main phases are: (i) Geostatistical analysis of the spatial structure of errors. (ii) Generation of two sets of synthetic DEMs using geostatistical simulation. (iii) Simulation with the two flood models using the sets of synthetic DEMs, and (iv) analysis of the propagation of errors. The results indicate that a DEM generated with UAV is useful and a feasible alternative to carry out the simulation of floods due to overflow of rivers due to their detail over time in changes in the area. The ABMs are very useful for the generation of flood scenarios thanks to their possibilities of automation and execution in a short time. In addition, the values of sensitivity and specificity with ABM of the DEM with UAV indicate a high capacity to detect flooded areas and medium - high capacity for safe areas.