Search

En este trabajo se evalúan los métodos de foto-reconstrucción automatizada basados en el uso conjunto de las técnicas Structure from Motion (SfM) y Multi-View Stereo (MVS) para medir, monitorizar y cuantificar la dinámica de tres formas geomorfológicas: i) el glaciar rocoso del Corral del Veleta (Granada, España), ii) un paisaje de cárcavas de tipo calanchi (Sicilia, Italia) y ii) cinco pequeñas cabeceras de cárcava (Cáceres, España). Se incluyen en este trabajo los resultados sobre la precisión, utilidad y aplicabilidad de estas técnicas. Para la cuantificación de la precisión se utilizan el error cuadrático medio (RMSE) de los puntos de control que se emplean en la georreferenciación y las distancias medias absolutas entre cada nube de puntos y una nube de puntos de referencia calculadas con los métodos cloud-to-cloud y multiscale model-to-model cloud comparison. Para los casos i) y iii) las nubes de referencia se capturaron utilizando un escáner láser terrestre, mientras que para el caso ii) se empleó un modelo digital de elevaciones obtenido mediante fotogrametría clásica aplicada a fotogramas aéreos capturados con un vehículo aéreo no tripulado. Los resultados de estos análisis mostraron precisiones de rango centimétrico con distancias medias a los modelos de referencia que oscilaron entre 0,01 m y 0,42 m. Las precisiones estimadas indican que los modelos obtenidos podrían utilizarse para calcular el volumen de cambio geomorfológico, tan sólo en el caso de las cabeceras. En el caso del glaciar y del calanchi los modelos podrían emplearse para elaborar cartografía precisa o estimar parámetros morfológicos, pero no para registrar cambios a escala anual.

Calanchi are a frequent type of badland landscape of Sicily (Italy), usually located in the middle of crops or forests. They are characterized by heavily dissected terrains with unvegetated slopes, knife ridge edges, V-shaped valleys and channels with a dendritic pattern, which incise and extend headwards. Calanchi exhibit, in smaller temporal and spatial scales, many of the geomorphic processes and landforms that may by observed in a fluvial landscape, hence, this type of badland may be considered as micro-watersheds where geomorphic dynamics can be related to their geometric features. The aim of this investigation was to evaluate the morphometric characteristics of calanchi landforms in Sicily (Italy). For this aim, DEMs obtained by LiDAR technology, with a horizontal and vertical resolutions of 2 m and 0.1-0.2 m, respectively, were used. In addition, orthophotographs with a pixel resolution of 0.25 m were exploited as a visual reference of the areas, as well as the images available on Google Earth. The morphometric characteristics were used to test two functional relationships: 1) an empirical relationship linking the volume of sediments eroded on a calanchi hydrographic unit to the total length of its main channel, and 2) a power relationship, established between two dimensionless groups of morphometric variables, that attests for a unique geometrical similarity condition between calanchi and smaller linear erosion landforms. Finally, the Hack’s law was tested demonstrating that the shape of calanchi units becomes wider with the increasing their drainage area. This result was also proved by the analysis of the maximum length and width of each landform, which showed that width increments slightly faster than length when increasing their size.

Three-dimensional photo-reconstruction methods based on Structure from Motion (SfM) and MultiView-Stereo techniques (MVS) are tested for measuring, monitoring and quantifying three different geomorphological features: i) the Corral del Veleta rock glacier and ii) five small gully headcuts in Spain and iii) Calanchi type badland in Italy. Results about the accuracy, usefulness and applicability of these techniques are presented here. The results of these analyses showed centimeter-level accuracies with average distances to the benchmark models ranging from 0.009 m to 0.42 m.

In this study, statistical models of earth-flow susceptibility were prepared using logistic regression. The analyses were carried out in a small (51 km2) basin of western Sicily, where 1,376 earth-flows were identified. To predict the spatial distribution of the mapped landslides, outcropping lithology and seven topographic attributes were exploited as explanatory variables. Before calculating these variables, a reconstruction of the pre-failure topography was performed. To evaluate the predictive skill and the robustness of the models, two groups made of five random subsets of earth-flows and stable cells were prepared. Absences of the first group were selected as individual cells whereas those of the second group where picked up using circular areas randomly distributed within the stable portions of the basin. The largely satisfactory accuracy and stability of the models demonstrated the reliability of the reconstructed pre-landslide topography. Furthermore, models prepared with the circular groups of absences showed better predictive skills.

In the Mediterranean areas, specifically in Sicily (Italy), irregular rainfalls, strong seasonal changes, scarce vegetation cover and, frequently, outcropping of clayey deposits favor water erosion phenomena. Badland landscapes are the result of severe erosion processes, characterized by steep slopes, sparse vegetation, high drainage density, rapid erosion rates and a shallow or non existing regolith profile. In this investigation we focused on the calanchi badland type, consisting of heavily dissected terrain with steep, unvegetated slopes and channels that rapidly incise and extend headwards. This research was carried out in two calanchi sites located in Sicily. The geometry of 25 badland channels was characterized in order to verify if relationships, already tested for minor linear erosion landforms (i.e. rills, ephemeral and permanent gullies), could also be verified for these bigger erosion channels. To this aim, two Digital Elevation Models (DEM) were processed in a Geographic Information System (GIS) environment: the first DEM obtained by a LIDAR survey, with 2 m and 0.1-0.2 m of horizontal and vertical resolution; the second DEM obtained by photogrammetry of 840 images captured by a drone, with 0.3 m and 0.05 m of horizontal and vertical resolution. Each channel was divided into segments delimited by transverse sections. Cumulative length and volume of all channels segments were plotted on scatter diagrams showing highly significant power relationships. Additionally, some morphometric attributes of channels segments (length, volume) and sections (depth, width), were combined into two dimensionless groups, already tested for minor erosion landforms, providing measured pairs also highly correlated by power relationships. The results of this experiment confirm that length of erosion channels is sufficient to predict volume of eroded material and evidence a morphological similarity between rill, ephemeral and permanent gullies and calanchi landforms.

Sicily is affected by severe erosion processes, locally causing large volumes of soil loss and high denudation rates. The effects of intense water erosion phenomena are testified by the occurrence of badlands that constitute a common landscape particularly in the central and southern areas of the Island, where slopes are frequently underlain by clay-rich deposits. This paper presents the first results of an investigation carried out in a badlands site located in the head sector of the Imera Meridionale river basin, where geometry of several erosion channels has been characterized. Since length of linear erosion landforms such as rills, ephemeral gullies and permanent gullies, has been proved to be a good estimator of the eroded volume in different environments, this experiment aims at verifying if similar relationships can be identified also for badlands channels. In order to quantitatively describe the geometry of the erosion landforms, a 2 m cell digital elevation model and GIS-spatial analysis tools were exploited. Width, depth and length of the channels have been measured on cross-sections spaced at 5 m or 3 m. Cumulative length and volume of the different channels have been plotted on scatter diagrams showing highly significant power relationships; moreover, length, depth, width and volumes of all measured channel segments were combined into two dimensionless groups, already tested for other channelized erosion landforms, that provided measured pairs also highly correlated by power relationships. The result of this experiment agrees with those of other investigations carried out for rills, ephemeral and permanent gullies, confirming that: i) strong functional relationships between length of linear erosion landforms and eroded volumes can be identified regardless of the channels size; ii) badland channels show a morphological similarity with the other investigated types of linear erosion landforms.