Light Detection and Ranging (LiDAR) technology has revolutionized the field of hydrological modeling by providing high-resolution and accurate Digital Elevation Models (DEMs). These LiDAR-based DEMs are widely used to extract catchment characteristics, such as drainage area, slope, and stream network, which are essential for various hydrological applications and flood risk assessments. However, the accuracy of these characteristics is subject to uncertainty due to the choice of ground filtering algorithms used to remove non-terrain points from the raw LiDAR data. Ground filtering is a crucial pre-processing step that directly impacts the quality of the resulting DEM and the derived catchment characteristics. This study, conducted in collaboration with the Norwegian Geotechnical Institute (NGI), aims to investigate the uncertainties associated with the estimation of catchment characteristics using LiDAR-based DEMs created by five different ground filtering algorithms: ATIN, CSF, MCC, MSBF, and SMRF. These algorithms, which can be applied using open-source software, are employed to create LiDAR-based DEMs for six urban areas in Norway. A total of thirty urban basins are delineated from these DEMs, and their morphometric characteristics, including catchment area, perimeter, mainstream length, slope, and hypsometric curve, are extracted. To quantify the uncertainty in the derived catchment characteristics, key statistical measures such as Relative Difference, Standard Deviation, and Coefficient of Variation are calculated. According to this study's assessment, although four basins showed consistent results across algorithms, Basins A and D exhibited higher uncertainty, particularly in complex urban landscapes. The MSBF algorithm demonstrated better performance in heavily urbanized basins, while other algorithms had limitations.
La tecnologia Light Detection and Ranging (LiDAR) ha rivoluzionato il campo della modellazione idrologica fornendo modelli digitali di elevazione (DEM) accurati e ad alta risoluzione. Questi DEM basati su LiDAR sono ampiamente utilizzati per estrarre le caratteristiche del bacino idrografico, come l'area di drenaggio, la pendenza e la rete di corsi d'acqua, che sono essenziali per varie applicazioni idrologiche e valutazioni del rischio di alluvioni. Tuttavia, l’accuratezza di queste caratteristiche è soggetta a incertezza a causa della scelta degli algoritmi di filtraggio del terreno utilizzati per rimuovere punti non legati al terreno dai dati LiDAR grezzi. Il filtraggio del terreno è una fase cruciale di pre-elaborazione che influisce direttamente sulla qualità del DEM risultante e sulle caratteristiche del bacino derivato. Questo studio, condotto in collaborazione con il Norwegian Geotechnical Institute (NGI), mira a indagare le incertezze associate alla stima delle caratteristiche del bacino utilizzando DEM basati su LiDAR creati da cinque diversi algoritmi di filtraggio del terreno: ATIN, CSF, MCC, MSBF e SMRF . Questi algoritmi, che possono essere applicati utilizzando software open source, vengono impiegati per creare DEM basati su LiDAR per sei aree urbane in Norvegia. Da questi DEM vengono delineati un totale di trenta bacini urbani e vengono estratte le loro caratteristiche morfometriche, tra cui bacino idrografico, perimetro, lunghezza della corrente principale, pendenza e curva ipsometrica. Per quantificare l'incertezza nelle caratteristiche derivate del bacino, vengono calcolate misure statistiche chiave come la differenza relativa, la deviazione standard e il coefficiente di variazione. Secondo la valutazione di questo studio, sebbene quattro bacini abbiano mostrato risultati coerenti tra gli algoritmi, i bacini A e D hanno mostrato una maggiore incertezza, in particolare nei paesaggi urbani complessi. L’algoritmo MSBF ha dimostrato prestazioni migliori nei bacini fortemente urbanizzati, mentre altri algoritmi presentavano dei limiti.
Uncertainty analysis of the catchment characteristics obtained from different LiDAR-Based DEMs
Vakily, Sayedehtahereh
2022/2023
Abstract
Light Detection and Ranging (LiDAR) technology has revolutionized the field of hydrological modeling by providing high-resolution and accurate Digital Elevation Models (DEMs). These LiDAR-based DEMs are widely used to extract catchment characteristics, such as drainage area, slope, and stream network, which are essential for various hydrological applications and flood risk assessments. However, the accuracy of these characteristics is subject to uncertainty due to the choice of ground filtering algorithms used to remove non-terrain points from the raw LiDAR data. Ground filtering is a crucial pre-processing step that directly impacts the quality of the resulting DEM and the derived catchment characteristics. This study, conducted in collaboration with the Norwegian Geotechnical Institute (NGI), aims to investigate the uncertainties associated with the estimation of catchment characteristics using LiDAR-based DEMs created by five different ground filtering algorithms: ATIN, CSF, MCC, MSBF, and SMRF. These algorithms, which can be applied using open-source software, are employed to create LiDAR-based DEMs for six urban areas in Norway. A total of thirty urban basins are delineated from these DEMs, and their morphometric characteristics, including catchment area, perimeter, mainstream length, slope, and hypsometric curve, are extracted. To quantify the uncertainty in the derived catchment characteristics, key statistical measures such as Relative Difference, Standard Deviation, and Coefficient of Variation are calculated. According to this study's assessment, although four basins showed consistent results across algorithms, Basins A and D exhibited higher uncertainty, particularly in complex urban landscapes. The MSBF algorithm demonstrated better performance in heavily urbanized basins, while other algorithms had limitations.File | Dimensione | Formato | |
---|---|---|---|
2024_03_Vakily_Tesi.pdf
accessibile in internet solo dagli utenti autorizzati
Dimensione
6.51 MB
Formato
Adobe PDF
|
6.51 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/219868