University of Twente Student Theses

Login

Integrating Probabilistic Weather Forecasting for Flood Early Warning in Dominica

Roon, Katherine van (2024) Integrating Probabilistic Weather Forecasting for Flood Early Warning in Dominica.

[img] PDF
65MB
Abstract:Climate change is expected to intensify extreme weather events, particularly tropical cyclones, leading to increased flooding risks in vulnerable regions like the Caribbean. This research addresses the need for accurate, rapid flood forecasting tools that provide clear visualizations and quantify forecast uncertainties. By integrating probabilistic (ensemble) forecasts into the FastFlood model, this study focuses on Dominica, recently affected by severe hurricanes, using historical data from Hurricane Maria (2017). Results show that ensemble forecasts effectively support early flood warnings in Dominica, demonstrating an inverse relationship between lead times and forecast uncertainty. FastFlood's outputs were accurate compared to historical records, and using quantiles for extreme precipitation prediction showed promise with lead times of up to eleven days. Local stakeholders in Dominica found several methodological elements beneficial, including rainfall thresholds and exposure prediction options. While promising, the implementation of ensembles in FastFlood faces challenges. Further applications in varied catchments and precipitation events are recommended to enhance method consistency and findings reliability. This research highlights the potential of probabilistic forecasting to improve flood early warning systems, laying a foundation for future studies.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Subject:38 earth sciences
Programme:Spatial Engineering MSc (60962)
Link to this item:https://purl.utwente.nl/essays/101105
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page