What's streamflow got to do with it? A probabilistic simulation of the competing oceanographic and fluvial processes driving extreme along-river water levels

Katherine A Serafin,Kai Parker,David F Hill,Peter Ruggiero

doi:10.5194/nhess-19-1415-2019

Katherine A Serafin, Kai Parker + Show 2 more

Open Access

https://doi.org/10.5194/nhess-19-1415-2019

Copy DOI

Abstract

Abstract. Extreme water levels generating flooding in estuarine and coastal environments are often driven by compound events, where many individual processes such as waves, storm surge, streamflow, and tides coincide. Despite this, extreme water levels are typically modeled in isolated open-coast or estuarine environments, potentially mischaracterizing the true risk of flooding facing coastal communities. This paper explores the variability of extreme water levels near the tribal community of La Push, within the Quileute Indian Reservation on the Washington state coast, where a river signal is apparent in tide gauge measurements during high-discharge events. To estimate the influence of multiple forcings on high water levels a hybrid modeling framework is developed, where probabilistic simulations of joint still water level and river discharge occurrences are merged with a hydraulic model that simulates along-river water levels. This methodology produces along-river water levels from thousands of combinations of events not necessarily captured in the observational records. We show that the 100-year still water level event and the 100-year discharge event do not always produce the 100-year along-river water level. Furthermore, along specific sections of river, both still water level and discharge are necessary for producing the 100-year along-river water level. Understanding the relative forcing driving extreme water levels along an ocean-to-river gradient will help communities within inlets better understand their risk to the compounding impacts of various environmental forcing, which is important for increasing their resilience to future flooding events.

Highlights

Coincident or compound events are a combination of physical processes in which the individual variables may or may not be extreme; the result is an extreme event with a significant impact (Zscheischler et al, 2018; Bevacqua et al, 2017; Wahl et al, 2015; Leonard et al, 2014)
The following section first validates the presence of a riverinduced water level within the tide gauge signal and demonstrates the effectiveness of the surrogate models in representing along-river water levels for unmodeled HECRAS boundary conditions
While the ADCSWAN runs only explore one instance of this phenomenon, it provides physics-based evidence that anomalously high ηSS at the La Push tide gauge is likely being driven by large discharge events

Summary

Introduction

Coincident or compound events are a combination of physical processes in which the individual variables may or may not be extreme; the result is an extreme event with a significant impact (Zscheischler et al, 2018; Bevacqua et al, 2017; Wahl et al, 2015; Leonard et al, 2014). Studies at the global (e.g., Ward et al, 2018), national (e.g., Wahl et al, 2015; Svensson and Jones, 2002; Zheng et al, 2013) and regional scale (e.g., Odigie and Warrick, 2017; Moftakhari et al, 2017) have evaluated the likelihood for variables such as high river flow and precipitation to occur during high coastal water levels, demonstrating that dependencies often exist between these individual processes. Estimates of nontidal residuals are determined by subtracting predicted tides from the measured wa-

Methods

Results

Discussion

Conclusion