Advancing Understanding of Atmospheric River Flood Hazards: From Antecedent Conditions to Social Vulnerability
Loading...
Authors
Webb, Mariana
Issue Date
2025
Type
Dissertation
Language
en_US
Keywords
antecedent soil moisture , AR scale , ARkStorm , atmospheric river , flood , social vulnerability
Alternative Title
Abstract
Atmospheric rivers (ARs) play a critical role in the global water cycle, producing extreme precipitation and nearly a quarter of global runoff. In many regions, such as the Western United States (U.S.), ARs are a dominant driver of flood hazards and damages. With climate change, the frequency and intensity of AR-driven floods are projected to increase. However, anticipating the magnitude and impacts of AR-driven flooding remains a major challenge. Standard metrics of AR strength, such as the intensity and duration of atmospheric moisture transport, do not always correspond to the hydrologic response. A critical gap remains in connecting ARs to the underlying land surface and social conditions that determine flood severity, timing, and impact. This dissertation addresses this gap by investigating the physical and social factors that shape AR flood impacts across a range of hydrologic and climate settings. In Chapter 2, I evaluate the role of antecedent soil moisture (ASM) in shaping peak streamflow response during ARs across 122 catchments on the U.S. West Coast. I find that ASM exerts a strong non-linear control on event-scale streamflow, with flood magnitudes more than doubling when ASM exceeds a critical, site-specific threshold. With four out of five AR flood events occurring under wet antecedent conditions, I show that ASM offers the best value for AR flood prediction in catchments with lower hydrologic storage capacity driven by shallower soils, less snowpack, and more clay-rich soils with limited infiltration rates. In Chapter 3, I assess the ability of the AR scale to characterize flood-generating ARs using streamflow observations from 145 catchments in California and central Chile. I demonstrate that the current scale, based solely on atmospheric vapor transport, can underestimate flood risk because it does not account for runoff processes controlled by ASM conditions. I propose a modified AR scale for flood hazards that incorporates antecedent precipitation, which significantly enhances the identification of flood-generating ARs across California and central Chile. In Chapter 4, I evaluate the catchment-scale impacts of a plausible, long-duration flooding event simulated from a sequence of 30 days of back-to-back ARs using future climate projections. Using a series of hydrologic and hydraulic models, I present a framework for evaluating how AR flood hazard and exposure evolve. Furthermore, I show that flood duration, not just peak magnitude, drives impacts and exacerbates exposure inequality in areas often overlooked by traditional flood planning. Together, these chapters contribute a new understanding of how integrating land surface and social vulnerability conditions can enhance our knowledge of AR flood hazards.