University of Twente Student Theses

Login
As of Friday, 8 August 2025, the current Student Theses repository is no longer available for thesis uploads. A new Student Theses repository will be available starting Friday, 15 August 2025.

A Dynamic Model for Landslide Early Warnings for the Road Network in Colombia

Urueña Ramirez, D.A. (2025) A Dynamic Model for Landslide Early Warnings for the Road Network in Colombia.

[img] PDF
22MB
Abstract:Landslides routinely disrupt Colombia’s Andean road network, yet existing early-warning practice relies on fragmented event records and static rainfall thresholds. This thesis develops a dynamic, data-driven framework that links a unified landslide inventory with real-time precipitation forecasts to issue actionable road alerts. Eight national and regional catalogues were harmonised into a 17,824-event database (2000-2024). Spatial coverage was evaluated with kernel-density mapping; temporal completeness was quantified with a new Adjusted Temporal Units Completeness Score, revealing both well-documented hotspots and data gaps. A space-time Generalised Additive Model combines static predictors (slope, lithology, land cover, road class) with dynamic rainfall indices, antecedent totals, plus triggers from CHIRPS-GEFS. Cross-validation respecting spatial and temporal dependencies yields AUROC ≈ 0.75 and Brier ≈ 0.15, confirming operational suitability. A Youden-optimised probability threshold (0.30) feeds a traffic-light scheme that aggregates slope-unit probabilities into road-segment alerts. Back-testing on the January 2023 storm cluster captured 18 of 21 reported landslides and correctly highlighted affected road segments, while narrowing field action to manageable warning volumes. The study delivers (i) a quality-scored national inventory, (ii) a validated real-time landslide probability model, and (iii) an alert prototype ready for a road authority pilot deployment. This work is advancing Colombia toward risk-informed, climate-resilient road management.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Subject:38 earth sciences, 43 environmental science
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:https://purl.utwente.nl/essays/107616
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page