NH042-01
The Varying Drivers and Impacts of Extreme Total Water Levels in the Pacific Basin

Thursday, 17 December 2020: 07:05
Virtual
Meredith Leung1, Peter Ruggiero2, Fernando J. Mendez3, Dylan Anderson4, Ana Rueda3, Laura Cagigal5, Alba Cid Carrera3, Nicolas Ripoll Cabarga3 and John J Marra6, (1)Oregon State University, College of Earth, Ocean, and Atmospheric Sciences, Corvallis, OR, United States, (2)Oregon State University, Corvallis, OR, United States, (3)University of Cantabria, Ciencias y Tecnicas del Agua y del Medio Ambiente, Santander, Spain, (4)Oak Ridge Institute for Science and Education, United States Army Corps of Engineers, Kitty Hawk, NC, United States, (5)University of Auckland, School of Environment, Auckland, New Zealand, (6)NOAA Honolulu, Honolulu, HI, United States
Recorded Presentation
Abstract:
Total water levels (TWLs) are a critical metric in determining the consequences of a changing climate on chronic coastal hazards such as flooding and erosion. The statistical modeling framework TESLA (Time varying Emulator for Short and Long-term Analysis of coastal flooding and erosion), produces spatially-varying, probabilistic forecasts of TWLs that enable quantitative assessment of the drivers and impacts of extreme TWLs supporting informed management of the coast. Here, the TESLA framework is used to generate future TWLs at three sites in the Pacific Basin allowing for a comparison of the different effects of evolving chronic hazards on each site and analyses of critical drivers forcing these differences.

TESLA is a stochastic climate emulator that uses statistical analysis and machine learning techniques on sea surface temperature, sea level pressure, and sea level pressure gradient data to produce patterns representing large scale climate, seasonality, intra-seasonal variability, and daily weather (Anderson et al., 2019). These climate and weather patterns are linked using conditional probabilities and use auto-logistic regression models to produce hypothetical timeseries of TWL drivers (wave conditions, storm surge, and monthly mean sea level anomalies). Multiple iterations of these hypothetical timeseries in combination with forecasted tides and regional sea level rise projections allows for the calculation of probabilistic future TWLs at each of our sites.

TESLA-derived TWLs were generated for military bases in Kāne‘ohe Bay, HI; San Diego, CA; and for community management in Tillamook County, OR. These sites each experience unique processes driving extreme TWLs (e.g., extra-tropical vs tropical cyclones) and have unique concerns regarding the effective management of their coasts. TESLA allows for exploration of which processes play the most significant role in forcing extreme TWLs and assessment of how climate change induced shifts in the frequency of large-scale climate patterns (i.e., El Nino Southern Oscillation) may impact the severity of future chronic coastal hazards. Using TESLA TWLs, we demonstrate that different processes drive extreme TWLs at each site; therefore, awareness of the distinct processes acting on each site is needed for informed approaches to management.

Previous Abstract | Next Abstract >>