LiDAR and Geomorphic Parameters for Anthropogenic Feature Extraction in Floodplains

Floodplains are among the world’s most modified landscapes, and in such environments, human alteration is reflected by anthropogenic features such as levees and scarps that directly influence hydrological processes, with consequences on flow stage and flood risk. Analyzing the distribution of these features at multiple scales, therefore, is critical for an effective floodplain management. However detecting such structures is technically challenging, and up-to-date, accurate and sufficiently attributed digital data are usually lacking, especially when dealing with large-scale applications. As a consequence, there is a clear need for developing high quality, cost-effective techniques to generate accurate, inexpensive spatial datasets. LiDAR Digital Terrain Models (DTMs) are readily available for many public authorities, and there is a greater and more widespread interest in the application of such information to solve geomorphological and hydrological problems. Anthropogenic feature extraction from DTMs in floodplain is a relatively new field of study, that can offer for large-scale applications a quick and accurate method to improve topographic databases, and that can overcome some of the problems associated with traditional field mapping. In natural contexts, morphological indicators derived from LiDAR DTMs have been proven to be reliable for feature extractions, and the use of statistical operators as thresholds showed a high effectiveness in identifying specific elements. The goal of this research is to test if these morphological indicators and objective thresholds can be feasible also in floodplains. In this work, different geomorphic parameters are tested and applied at different scales on a LiDAR DTM of typical floodplain. The box-plot is applied to identify the threshold for feature extraction, and a filtering procedure is proposed to improve the quality of the final results. The results highlight the capability of high resolution topography, geomorphic indicators and statistical thresholds for anthropogenic features extraction and characterization in a floodplains context.