Remote sensing and uavs for the geomorphological and habitat analyses in ephemeral and permanent mediterranean streams

Resumen

Riparian ecosystems show great variability, from a geomorphological, hydrological and ecological point of view, including the complex interactions that riparian morphology and vegetation can present. Vegetation appears as a very influential physical factor in river systems, with a direct relationship in the geomorphological processes that take place in river corridors. The detection, monitoring and evaluation of the processes that take place in the riparian space are key when it comes to understanding the ecological functions and development of these habitats, and therefore for making decisions for their conservation and restoration. According to the distribution of species and plant traits, plant communities and their dynamics present different characteristics in the riparian ecosystem, to which detection and monitoring methods must be adapted. The constant changes that these spaces undergo over time are largely due to physical processes related to the dynamics of erosion and sedimentation, variations in the path of the channel, variations in the distribution of species and vegetation in the riparian forest, etc. These processes also are due to the anthropogenic impact, which can generate major imbalances in the ecological dynamics of the ecosystems in question. Due to the interactions of various anthropogenic processes and alterations, and the complex spatio-temporal dynamics, it is necessary to continue developing theoretical and practical methodologies for the monitoring and characterization of these ecosystems. Remote sensing, including the use of drones, is presented as a very interesting and optimal tool for mapping and collecting information in these natural spaces. The benefits demonstrated by unmanned aircraft -UAV- include improvements in the spatial and temporal resolution of the captured data, as well as the mapping of large areas in a short time, which makes them key instruments in the management and conservation tasks of riparian spaces. The need to study the geomorphological dynamics that occur in river channels has been the main motivation in the studies presented in this doctoral thesis. Chapters 2 and 3 are based on ground-based laser scanner (TLS) data capture techniques and modelling of UAV photogrammetric flight data. They have characterized the processes that take place in a certain study area, an ephemeral riverbed in the southeast of the Iberian Peninsula, the Rambla de la Azohía (Murcia). These studies have also made it possible to compare the fit and precision of the data captured from two different techniques. In addition, the interest in characterizing the fluvial channels with a permanent flow has motivated the study of the submerged topography in a stretch of river, segmented by types of mesohabitat. Thus, chapter 4 presents an algorithm and a correction tool for the effect of refraction in a stretch of the Palancia river (Castellón), to carry out the correct representation of the submerged bed morphology. From the proposed methodology and the developed algorithm, it is possible to minimize the distortion effects due to the presence of water, to obtain the three-dimensional reconstruction of the bed from images taken with UAVs. The construction of the 3D model was carried out using the Structure from Motion technique. Finally, and as a key element in the dynamics of riparian ecosystems, chapter 5 develops a methodology to classify the phases of succession of riparian forest vegetation. These succession phases are based on the RIPFLOW project methodology, which is also implemented in the dynamic CASiMiR-vegetation model.