Uncertainty in hurricane surge simulation due to land cover specification

Ferreira, Celso M.; Irish, Jennifer L.; Olivera, Francisco

doi:10.1002/2013jc009604

Cited by 32 publications

(27 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Manning's n coefficient of 0.03 and 0.17 might be associated with pasture or grassland and mixed forest [31], therefore, the model is very sensitive to the land cover classification. It might confirm what Ferreira et al (2014) [43] demonstrated, and an incorrect land cover choice may lead to error in surge of 7.0%. Resio and Westerink (2008) [9] discussed that an increased frictional resistance slows the rate the flood wave moves inland.…”

Section: Storm Surge In Overland Areassupporting

confidence: 56%

Storm Surge Modeling in Large Estuaries: Sensitivity Analyses to Parameters and Physical Processes in the Chesapeake Bay

Garzon

Ferreira

2016

JMSE

View full text Add to dashboard Cite

Large estuaries are especially vulnerable to coastal flooding due to the potential of combined storm surges and riverine flows. Numerical models can support flood prevention and planning for coastal communities. However, while recent advancements in the development of numerical models for storm surge prediction have led to robust and accurate models; an increasing number of parameters and physical processes' representations are available to modelers and engineers. This study investigates uncertainties associated with the selection of physical parameters or processes involved in storm surge modeling in large estuaries. Specifically, we explored the sensitivity of a hydrodynamic model (ADCIRC) and a coupled wind-wave and circulation model system (ADCIRC + SWAN) to Manning's n coefficient, wind waves and circulation interaction (wave setup), minimum depth (H 0 ) in the wetting and drying algorithm, and spatially constant horizontal eddy viscosity (ESLM) forced by tides and hurricane winds. Furthermore, sensitivity analysis to Manning's n coefficient and the interaction of waves and circulation were analyzed by using three different numerical meshes. Manning's coefficient analysis was divided into waterway (rivers, bay and shore, and open ocean) and overland. Overall, the rivers exhibited a larger sensitivity, and M 2 amplitude and maximum water elevations were reduced by 0.20 m and 0.56 m, respectively, by using a high friction value; similarly, high friction reduced maximum water levels up to 0.30 m in overland areas; the wave setup depended on the offshore wave height, angle of breaking, the profile morphology, and the mesh resolution, accounting for up to 0.19 m setup inside the bay; minimum depth analysis showed that H 0 = 0.01 added an artificial mass of water in marshes and channels, meanwhile H = 0.1 partially solved this problem; and the eddy viscosity study demonstrated that the ESLM = 40 values reduced up to 0.40 m the peak of the maximum water levels in the upper side of narrow rivers.

show abstract

Section: Storm Surge In Overland Areassupporting

confidence: 56%

Storm Surge Modeling in Large Estuaries: Sensitivity Analyses to Parameters and Physical Processes in the Chesapeake Bay

Garzon

Ferreira

2016

JMSE

View full text Add to dashboard Cite

show abstract

“…ADCIRC is a finite‐element hydrodynamic model that generates water levels and current velocities, and is widely used for storm surge modelling in the east coast of United States and Gulf of Mexico (e.g. Ferreira et al , ). We used the 2D, depth‐integrated version of ADCIRC (Luettich and Westerink, ) that solves the vertically integrated generalised wave continuity equation (GWCE) and the momentum equations (Eqns and , respectively):

\frac{\partial h}{\partial t} + \nabla h ((), \overset{U}{true} h) = 0

\frac{\partial \overset{U}{true}}{\partial t} + ((), U . \nabla h) \overset{U}{true} = prefix- g \nabla h ((), ζ + \frac{p ((),, x, y)}{g ρ}) + f \overset{k}{true} \times \overset{U}{true} + \frac{\overset{τ}{true} normals}{h ρ} - \frac{\overset{τ}{true} normalb}{h ρ}

where h is bathymetric depth, t is the time, ζ is surge elevation above MSL,

\overset{U}{true}

is depth‐averaged horizontal velocity vector, p is barometric pressure, f is the Coriolis force coefficient,

\overset{k}{true}

is a vertical unit vector, τ s is the free surface shear stress, τ b is the bottom shear stress, ρ is the water density and g is gravitational acceleration.…”

Section: Methodsmentioning

confidence: 99%

Simulation of cyclone‐induced storm surges in the low‐lying delta of Bangladesh using coupled hydrodynamic and wave model (SWAN + ADCIRC)

Deb

Ferreira

2016

J Flood Risk Management

View full text Add to dashboard Cite

Bangladesh is vulnerable to several natural disasters and cyclone‐generated storm surges have resulted in the deaths of over 700 000 people since 1960. Advancing our capability to model and simulate storm surges using numerical models is utmost important to support early warning and emergency response efforts in the region. This study primarily explored the effectiveness of a hydrodynamic model (ADvanced CIRCulation, ADCIRC) coupled with wave model (Simulating WAves Nearshore, SWAN) under a high‐performance computing environment to simulate storm surge and inundation in coastal regions of Bangladesh. The modelling framework was validated using data from freely available historical reports and buoy data. The model‐generated storm surge water level shows good agreement with the observations with maximum R 2 value of 0.98 and root mean square error of 0.30 m. Ultimately, research findings have highlighted the importance of the coupled wave and hydrodynamic modelling to calculate storm surges in a region with poor observational coverage.

show abstract

“…Bottom friction is parameterized in ADCIRC using a quadratic bottom stress, where the bottom drag coefficient is dependent on the Manning's n coefficient and depth at each mesh node (Kerr, Martyr, et al). Correct nearshore and overland frictional representation is critical to accurately simulating peak coastal water levels and inundation (Ferreira et al, ). A Manning's n of 0.025 and 0.028 was used for the Atlantic Ocean and the Caribbean Sea, respectively.…”

Section: Model Data and Experimental Descriptionmentioning

confidence: 99%

U.S. IOOS Coastal and Ocean Modeling Testbed: Hurricane‐Induced Winds, Waves, and Surge for Deep Ocean, Reef‐Fringed Islands in the Caribbean

Joyce

Gonzalez-Lopez²,

Westhuysen³

et al. 2019

JGR Oceans

View full text Add to dashboard Cite

As part of a U.S. Integrated Ocean Observing System (IOOS) funded Coastal and Ocean Modeling Testbed (COMT), hindcasts of waves and storm surge for 2017 Hurricanes Irma and Maria are examined and compared to wave and water level gauge data in the vicinity of Puerto Rico and the U.S. Virgin Islands. The region is characterized by adjacent deep ocean water, narrow shelves, and coral reef systems providing coastal protection. The storm physics are analyzed using an unstructured grid third‐generation wave circulation coupled modeling system (ADCIRC+SWAN) with respect to tides, winds, atmospheric pressure, waves, and wave radiation stress‐induced setup. The water level response is generally dominated by the pressure deficit of the hurricanes. Wind‐driven surge is important over the shallow shelf to the east of Puerto Rico and wave‐induced setup becomes significant at locations in close proximity to the coastline. Contrary to conditions along the Gulf of Mexico shelf, geostrophically induced setup is negligible. Characteristics from a range of meteorological forcing models are assessed, and the associated errors in the hydrodynamic response are quantified. A data‐assimilated tropical planetary boundary model leads to the smallest atmospheric pressure, water level and wave property errors across both storms. Through comparisons between ADCIRC+SWAN and SLOSH‐FW (a structured grid first‐generation wave circulation coupled model), it is shown that the response to atmospheric forcing is similar; however, nearshore wave setup is smaller in SLOSH‐FW due to its coarser resolution here. Further, in addition to erroneous wind‐driven surge through depth limiting over the open ocean, numerical oscillations in the water level time series develop in SLOSH‐FW likely due to its small domain size.

show abstract

Uncertainty in hurricane surge simulation due to land cover specification

Cited by 32 publications

References 36 publications

Storm Surge Modeling in Large Estuaries: Sensitivity Analyses to Parameters and Physical Processes in the Chesapeake Bay

Storm Surge Modeling in Large Estuaries: Sensitivity Analyses to Parameters and Physical Processes in the Chesapeake Bay

Simulation of cyclone‐induced storm surges in the low‐lying delta of Bangladesh using coupled hydrodynamic and wave model (SWAN + ADCIRC)

U.S. IOOS Coastal and Ocean Modeling Testbed: Hurricane‐Induced Winds, Waves, and Surge for Deep Ocean, Reef‐Fringed Islands in the Caribbean

Contact Info

Product

Resources

About