_______________________________________

The GeoRisk project, reference PID2019-103974RB-I00 is funded by MCIN/ AEI/10.13039/501100011033.

Improving 3D models for rockfall risk analysis

Detailed terrain models with known accuracy are important for some of the monitoring, analysis and simulation tasks that are carried out in the initial stages of a risk quantification study. These models must be of the generating wall and of the terrain through which blocks would move. Products derived from digital terrain models (DTMs), such as slope or roughness models, will also influence the definition of the trajectories.

The high resolution of the models obtained by photogrammetry or TLS (Terrestrial Laser Scanner) is an advantage in geology. It allows accurate and continuous modelling of the reality, and even identification of microfractures. Microfractures do not mark lines of discontinuity, but they give us clues about areas where the rocky massif is weaker and less cohesive, and therefore has a higher probability of breakage and instability. However, drawbacks exist, e.g. in the cases of complex geometries, such as overhangs, the use of aerial techniques is not possible, since occlusions formed by rocky projections prevent the capture of information. Moreover, on the terrain there are elements, such as vegetation, changing with the time. This information must be removed to conduct a multitemporal analysis of the rocky massif or to integrate techniques in its modelling over time. Vegetation is a variable factor as it grows and undergoes seasonal variations that affect its morphology.

Taking into account all this issues the objectives to achieve are:

Objective 1a: To establish processes and methods for combining geomatic techniques, considering the characteristics of precision and resolution that allow the discontinuity detection.

Objective 1b: To evaluate the process to eliminate the vegetation in the point clouds for geological applications. To establish the optimal methodology to filter with morphological, colour and reflectivity characteristics achieved by photogrammetry and TLS.

Objective 1c: To calculate or indirectly to measure the spatial distribution of the roughness in the deposit from point clouds. Thus, we expect to obtain a more realistic model of the propagation of rockfalls in which the reaction of the block that impacts will depend on the point of the deposit where that impact occurs.