Back to search

University of British Columbia

Advancing probable maximum precipitation and probable maximum flood estimation : a case study of the Alouette watershed

Abstract

dc:description

Probable maximum precipitation (PMP) and probable maximum flood (PMF) estimates are critical inputs for dam safety and hydrologic risk assessment. Substantial uncertainties exist, including the meteorological drivers of extreme events, the influence of streamflow model selection, and the impact of climate change. This dissertation advances PMP and PMF science through three different studies focused on the Alouette watershed in southwestern British Columbia, Canada. First, the characteristics of atmospheric rivers (ARs) producing extreme precipitation and streamflow events were examined using a range of statistical tests and reanalysis data. The majority of extreme-event-producing ARs approach the BC coast from compass directions between 210° and 220°. More westerly ARs can generate extreme events with integrated vapour transport values as low as 400 kg m⁻¹ s⁻¹, compared with over 800 kg m⁻¹ s⁻¹ for southerly ARs. Rain-on-snow occurrence, antecedent soil moisture, and AR approach angle strongly modulate streamflow response. We propose how the American (CW3E) AR severity scale should be enhanced for BC watersheds. Second, the sensitivity of PMF estimates to streamflow model selection and parametrisation was quantified using three structurally diverse models (Weather Research and Forecasting Model Hydrological Modelling System (WRF-Hydro), University of British Columbia Watershed Model (UBCWM) in Raven, and a neural network), all driven by an identical PMP. Model choice produced PMF peak-flow differences of up to 440 m³ s⁻¹ (28%), while parameter estimation uncertainty contributed an additional 12–16%. The machine learning model successfully extrapolated streamflow more than 40% beyond its training range, producing PMF estimates comparable to the process-based models. Third, a model-based PMP framework was developed combining convection-permitting WRF simulations of historical AR events, k-medoids storm selection, initial and boundary condition shifting, physically constrained moisture amplification, and pseudo-global warming (PGW) perturbations. Current-climate PMP estimates were consistent with site-specific operational values, while BC government-provided MetPortal values were significantly lower. Under future 2071–2100 PGW, the 24-hour PMP and PMF increased by 39% and 41% respectively compared to present, while the 96-hour PMP increased only 12%, with future precipitation efficiency declining in most storms. These results demonstrate that precipitation-only climate scaling factors cannot capture the impacts of climate change.

Degree

thesis:*
Name thesis:degree_name
Doctor of Philosophy - PhD
Level thesis:degree_level
doctoral
Discipline thesis:degree_discipline
Atmospheric Science
Grantor dc:publisher
University of British Columbia
Year dc:date
2026

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Legarth, Elise
Contributors dc:contributor
  • Stull, Roland B., 1950-

Rights

dc:rights
Statement dc:rights
  • Attribution-NonCommercial-NoDerivatives 4.0 International
Language dc:language
eng

Identifiers

dc:identifier.*
Handle dc:identifier
http://hdl.handle.net/2429/95646

Chain of custody

source
Harvested from
University of British Columbia
Base URL
circle.library.ubc.ca/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Legarth, Elise. Advancing probable maximum precipitation and probable maximum flood estimation : a case study of the Alouette watershed. doctoral thesis, University of British Columbia, 2026. http://hdl.handle.net/2429/95646