Search

Flood events, as well as the channel and floodplain landforms are crucial for the transport and spatial distribution of heavy metals and metalloids in soils in the river valleys. The paper outlines the results of obtained preliminary indications about the flood proneness of two study areas located in the valleys of Ogosta River near the village of Gorna Kovachitsa and Lom River near the village of Vasilovtsi (NW Bulgaria) based on hydro-geomorphological indices. The detection of flood-prone areas has been done using two key data products of UAV photogrammetry data ? the digital terrain model (DTM) and the orthophotos with grid size 1 m. Flood-prone areas have been identified using GIS techniques based on several hydro-geomorphological indexes extracted from digital terrain models (DTMs). Firstly, they were evaluated individually by their hydro-morphographic and topographic meaning according to the literature and analysis in geomorphological context, as well as through defining thresholds and analysis in topographical context to select the best performing classifiers. Secondly, the selected hydro-geomorphic descriptors have been binary classified to distinguish between flood-prone and non-flooded areas and used for creation of binary flood-prone areas maps, which were merged together and evaluated according their altitude above the river. A second methodological approach applied in this work is based on calculation of potentially flood areas from the raster “Vertical Distance to Channel Network” using the GRASS GIS module r.lake implemented in QGIS and binary classification. The orthoimages were used for extraction of the rivers geometry using object-based segmentation and classification. Additionally, the obtained results have been compared with the previous flooding data in the area of Ogosta Valley near Gorna Kovachitsa from flood in 2014. The obtained results highlight the potential for the GIS-based detection of flood-prone areas over ungauged areas and will be used for modelling the spatial distribution of heavy metals in soils in the study areas. [ABSTRACT FROM AUTHOR]

Floodplains are complex systems which include a wide number of socio-economic activities like agriculture, freshwater fisheries, electricity from power plants and others. They have often highly dynamic property, due to the alternation of fluvial processes, floods, and ongoing sediment transport, as well as of environmental specifics and natural or anthropogenic processes. In the context of the investigation of soil pollution with heavy metals in the riverine floodplains, the availability of geomorphological maps at large scale is of great importance. The rapid development of new remote sensing and computer technologies offers nowadays the possibility for acquisition of high-resolution topographic and land surface data which can serve as basis for digital terrain analysis and geomorphologic mapping. This paper outlines the preliminary results of applying unmanned aerial vehicle (UAV) data for geomorphological mapping of the study area. The methodological approach is based on the two key data products of UAV photogrammetry − the digital terrain model (DTM) and the orthophoto mosaic. The digital terrain model is used for calculation of terrain derivatives such as altitude above the channel network and topographic ruggedness index (TRI), using Geographic Information Systems (GIS). The orthophoto mosaic, on the other hand, is used to calculate local statistical measures, enabling detection of textural and structural properties, as well as for differentiation of features with similar spectral responses but different surface structures. The obtained results show that UAV photogrammetry is a powerful and inexpensive tool for fluvial remote sensing analysis and has the potential for high-resolution geomorphological mapping. [ABSTRACT FROM AUTHOR]