Evaluating essential processes and forecast requirements for meteotsunami-induced coastal flooding

Abstract

Meteotsunamis pose a unique threat to coastal communities and often lead to damage of coastal infrastructure, deluge of nearby property, and loss of life and injury. The Great Lakes are a known hot-spot of meteotsunami activity and serve as an important region for investigation of essential hydrodynamic processes and model forecast requirements in meteotsunami-induced coastal flooding. For this work, we developed an advanced hydrodynamic model and evaluate key model attributes and dynamic processes, including: (1) coastal model grid resolution and wetting and drying process in low-lying zones, (2) coastal infrastructure, including breakwaters and associated submerging and overtopping processes, (3) annual/seasonal (ambient) water level change, and (4) wind wave-current coupling. Numerical experiments are designed to evaluate the importance of these attributes to meteotsunami modeling, including a “representative storm” scenario in the context of regional climate change in which a meteotsunami wave is generated under high ambient lake-level conditions with a preferable wind direction and speed for wind-wave growth. Results demonstrate that accurate representation of coastal topography and fully resolving associated hydrodynamic processes are critical to forecasting the realistic hazards associated with meteotsunami events. As most of existing coastal forecast systems generally do not resolve many of these features due to insufficient model grid resolution or lack of essential model attributes, this work shows that calibrating or assessing existing forecast models against coastal water level gauges alone may result in underestimating the meteotsunami hazard, particularly when gauging stations are sparse and located behind harbor breakwaters or inside estuaries, which represent dampened or otherwise unrepresentative pictures of meteotsunami intensity. This work is the first hydrodynamic modeling of meteotsunami-induced coastal flooding for the Great Lakes, and serves as a template to guide where resources may be most beneficial in forecast system development and implementation.

The Hydrologic Response to a Meteotsunami in an Isolated Wetland: Beaver Island in Lake Michigan, USA

“…This provides challenges for analysis of travel times, directionality, reflection/refraction, and estimates of tsunami runup as we are comparing datasets with different distances from meteotsunami origin, coastline shapes, and bathymetries, as well as limitations in the accuracy of numerical weather forcing, hydrodynamic model physics, and topographic and bathymetric representation. A recent meteotsunami study has illustrated the sensitivity of meteotsunami simulations to shoreline resolution, which can impact wave amplitude and period (Huang et al., 2021). Additionally, we note there is insufficient information about how wetland systems beyond our studied location may respond to differing event magnitude, coastline shape, nearshore bathymetry, and lake levels.…”

“…To supplement the analysis of observed water level and atmospheric data during the 20 July 2019, event, we used a hydrodynamic model to simulate the water surface response and help in identification of meteotsunami generation. The model is based on the Finite Volume Community Ocean Model (FVCOM; Chen et al., 2006), which has been adapted for freshwater and successfully implemented for Great Lakes meteotsunami simulation (Anderson et al., 2015; Anderson & Mann, 2021; Huang et al., 2021) and other physical processes (Anderson et al., 2018; Anderson & Schwab, 2013, 2017). The model uses an unstructured grid with horizontal resolution that ranges from 100 m in the nearshore to 2,500 m in offshore regions.…”

“…The event was simulated using 15‐min meteorological forcing (2‐m air pressure, 10‐m meridional and zonal wind, 2‐m air temperature, 2‐m humidity, and downward solar radiation) from the 00:00 UTC forecast on 20 July 2019, of the NOAA High‐Resolution Rapid Refresh (HRRR; Benjamin et al., 2016) version 3, which is part of NOAA's short‐range weather forecast system. The HRRR has been employed to successfully simulate meteotsunami generation for past events on Lake Michigan (Anderson & Mann, 2021; Huang et al., 2021). Pressure was adjusted to MSLP as required by FVCOM using the hypsometric relationship (Smithsonian, 1951) and assuming a standard atmosphere (NOAA, 1976), while other variables were used in their native formats.…”

Climate projections over the Great Lakes Region: using two-way coupling of a regional climate model with a 3-D lake model

“…The version of RegCM used in this study is a 3-D, hydrostatic, compressible, primitive equation, σ -coordinate version (RegCM4.3.4) synchronously coupled with the Community Land Model version 4.5 (CLM4.5, Oleson et al, 2010Oleson et al, , 2013, which has the option to include prognostic treatment of the carbon and nitrogen cycles and dynamic vegetation (RegCM-CLM-CNDV, Wang et al, 2016). The FV-COM is an unstructured grid, finite-volume, 3-D, primitive equation, hydrodynamic model with a generalized terrainfollowing coordinate system in the vertical and triangular meshes in the horizontal, and is widely applied to coastal oceans and the Great Lakes (Xue et al, 2014(Xue et al, , 2015Anderson et al, 2018;Huang et al, 2019Huang et al, , 2021aYe et al, 2019Ye et al, , 2020Ibrahim et al, 2020). The version of FVCOM used in this study is derived from FVCOM 4.3.1 without applying nudging or other similar nonphysical constraints.…”

Data_Sheet_2.zip

“…bottom friction, and depth-induced wave breaking, as well as energy redistribution through nonlinear wave-wave interactions. Recognized as a reliable coastal community wave model, SWAN has been widely applied for wave hindcasting and forecasting (Rogers et al, 2003(Rogers et al, , 2007 for the Great Lakes (Anderson et al, 2015;Mao et al, 2016;Huang et al, 2021) and coastal oceans (Dietrich et al, 2011;Huang et al, 2013;Cheng et al, 2015;Xie et al, 2016;Niroomandi et al, 2018;Xie et al, 2019).…”