University of Twente Student Theses


Analyzing the suitability of OpenLISEM hazard debris flow modelling for land use planning in Colombia

Arango Carmona, Maria Isabel (2021) Analyzing the suitability of OpenLISEM hazard debris flow modelling for land use planning in Colombia.

Link to full-text:
(only accessible for UT students and staff)
Abstract:The Andean region in Colombia is a tropical and mountainous environment where high-intensity rainfall often triggers concatenated phenomena at the catchment scale that includes the coupled or cascading effects of landslides, flash floods, and hyperconcentrated, and debris flows. In Colombia, such events have caused more than 3000 deaths and affected more than 1 million people between 1920 and today. Recent changes in disaster risk management policies have made it mandatory to include debris flow hazard assessment in the land use planning process. Nevertheless, this analysis has always been evaluated from different approaches, where none of them accounts for its multi-hazard nature and includes all the processes involved in the phenomena. OpenLISEM Hazard is an integrated multi-hazard model that considers and models different hydro-meteorological phenomena and their interactions, like hydrology, runout, slope stability, slope failure, runout, and entrainment. This thesis project aims to analyse the suitability of applying the OpenLISEM Hazard model for developing debris flow hazard maps as a basis for land use planning in Colombia, focusing on evaluating the available datasets influence in terms of resolution and quality. The model was calibrated and validated using two past debris flow events in Colombia with different sources and data quality. The accuracy assessment of the model was evaluated for each subprocess, including slope failure, runout of hillslope debris flows, and flooding. The process with the higher accuracy is flooding, and the lowest one is the slope failure locations. In general terms, we could not achieve a simulation that represents all the subprocesses with acceptable accuracies in a single run. The calibration of the model included many different variables that were difficult to tune since they influence each sub-process in a different and not traceable way. This research shows that the main challenges of using multi-hazard modelling with limited information are the propagated uncertainties from unknown or estimated parameters that come into play in the final modelling output. While multi-hazard models show usefulness in research, their application to actual hazard and risk assessment in areas with limited data availability like Colombia should not be undertaken without an improved strategy for dealing with such uncertainties.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page