Sean Vitousek

Research Oceanographer, Pacific Coastal & Marine Science Center, Santa Cruz, CA

Coupled coastal change & flood modeling

(combining a data-assimilated shoreline model & flood models)

Motivation:

Coastal flooding will double within decades due to SLR:

Around the U.S., coastal flooding will double every 5-10 years:

Vitousek et al., 2017

Taherkhani et al., 2017

Coastal erosion will also accelerate with SLR

(Fortunately, coastal monitoring and prediction is getting very good)

Satellite-derived shoreline data from the CoastSat toolbox:

CoSMoS-COAST shoreline model:

(Vitousek et al., 2017, 2021, 2023)

Linking predicted shoreline change with coastal flooding

(Evolving elevation profile data with projected shoreline change … can be a bit tricky)

Erikson et al., 2017

“We show that for California, USA, the world’s 5th largest economy, over $150 billion of property equating to more than 6% of the state’s GDP and 600,000 people could be impacted by dynamic flooding by 2100; a three-fold increase in exposed population than if only SLR and a static coastline are considered.”

- Barnard et al., 2019

Future work: coupled shoreline/dune/flood modeling

(Include the evolution of coastal dunes in CoSMoS-COAST in collaboration with GFA, UC-SC,SB,SD)

The End

… Thanks!

EXTRAS

1,350 km(from satellites)

​

Forcing conditions:

Model Inputs:

Wave conditions

Wave hindcast

(1995-2020)

Model transects:

Governing equation:

Outputs:

• Transect locations

(100-200 m, shore-normal)

• Transect designations:

full model, cross-shore only,

rate only, or no prediction

• Non-erodible shoreline
• Beach slope
• No beach profile

elevation data needed

​

Localized

Ensemble

Kalman Filter (EnKF)

data assimilation

based on

‘littoral cells’

+

• Future shoreline positions
• Model parameter estimates
• Uncertainty estimates

via ensemble simulations &

comparisons w/ observations

CoastSat (Vos et al., 2019)

Wave forecast / projection

(2020-2100)

+ future sea-level conditions

Historical shoreline data:

• Lidar
• GPS
• Satellites

• For model calibration & validation:
• Sources:

​

CoSMoS-COAST: Coastal One-line Assimilated Simulation Tool

GFDL-ESM2M +

WaveWatch III +

SWAN

(look-up table)

Vitousek et al., 2017

Developed in:

Vitousek et al., 2021

California (state-wide)

projections

1,760 km

Model state variables and parameters:

Assimilated parameters:

transect #511

Vitousek et al., 2023

Transects

(i.e., model grid)

Wave Forcing

(single-realization or ensemble)

Historical shoreline data

(from satellites)

Shoreline prediction + uncertainty

(conservation of

long-shore & cross-shore sediment transport)

(ensemble Kalman filter based on littoral cells)

Governing equation + data-assimilation method

CoSMoS-COAST includes:

​

• wave-driven cross-shore (equilibrium) transport

(Wright et al., 1985, Yates et al., 2009)

​

• wave-driven longshore transport

(Pelnard-Considere 1956, Ashton & Murray 2006)

​

• SLR-driven cross-shore transport

(Bruun 1962, Anderson et al., 2015)

​

• residual (linear) shoreline trends (estimated via data assimilation)

(Hapke et al. 2006, Long & Plant 2012)

​

• random noise

(Vitousek et al., 2021)

​

​

​

​

1-D conservation of volume in the alongshore direction

Satellite vs. GPS (mean sea level - MSL) shoreline position at Ocean Beach, CA

Willmott, C. J. (1981). On the validation of models. Physical geography, 2(2), 184-194.

d ≈ 0.5-0.7

d > 0.5 for 58% of California

For the current application:

For “shoreshop”:

Montaño et al., (2020) Scientific Reports

Satellites enable large-scale model validation