Does the topographic data source truly influence the routing modelling of debris flows in a torrent catchment?
Mauro Boreggio, Martino Bernard, Carlo Gregoretti
Abstract
Abstract To cope with debris‐flow hazards, a common practice is the mapping of threatened areas through routing models. Considering the primary role of topography in affecting the mobility of gravity‐driven flows, its proper representation through digital elevation models (DEMs) is a requirement in routing modelling applications. The ‘quality’ of DEMs mainly depends on the quality, resolution and spatial arrangement of the topographic measurements (i.e. on the employed survey technology). Nevertheless, no attempt to systematically evaluate the influence of the topographic data source on the behaviour of routing models has been carried out. To address this, we initially assess the performances of both terrestrial‐ (i.e. global navigation satellite system, GNSS) and airborne‐based (i.e. full‐waveform LiDAR and structure‐from‐motion, SfM digital photogrammetry) survey technologies in characterizing the topography of a debris‐flow channel. Afterwards, we investigate whether the topographic data source can effectively influence the behaviour of a geographic information system (GIS)‐based cell routing model. Regarding the assessment of the survey technologies performances, the ‘standard’ statistic‐based approach indicated that GNSS and full‐waveform LiDAR can provide an accurate digital representation of the gully. However, the analysis of the shapes stressed that the most faithful and finer reproduction of the topographic singularities is yielded by the photogrammetrically reconstructed surface due to the extremely high data source resolution. Furthermore, the pairwise comparison of derived elevation models pointed out that meaningful discrepancies among tested survey technologies can be detected in morphologically complex areas because of the inherent limits of the terrestrial‐based method. Here, this research showed how these discrepancies have the potential to affect simulated flow dynamics, even if not in a meaningful way from a risk planning and management point of view. Overall, it appears that the topographic data source does not truly represent a determining factor in modelling applications of channelized debris‐flow routing.