MODELING TSUNAMI INUNDATION AND ASSESSING TSUNAMI HAZARDS FOR

MODELING TSUNAMI INUNDATION AND ASSESSING TSUNAMI HAZARDS FOR THE U. S. EAST COAST Report to National Tsunami Hazard Mitigation Program February 8, 2012 Presented by: James T. Kirby Center for Applied Coastal Research University of Delaware Stephan T. Grilli Department of Ocean Engineering University of Rhode Island

Project Personnel: University of Delaware: James T. Kirby (PI, inundation and propagation modeling, project coordination) John Callahan (Scientist, Del. Geological Soc. , GIS/DEM database development, inundation map development) Fengyan Shi (Res. Asst. Prof, inundation and propagation modeling) Babak Tehranirad (Graduate Research Assistant, inundation modeling and mapping)) University of Rhode Island: Stephan Grilli (co-PI, coordination of URI tasks), Chris Baxter (Assoc. Prof. , MC simulations of SMF sources). Jeff Harris (postdoc, integration of source simulations and propagation modeling), Stephane Abadie (Visiting prof. , Fulbright Scholar, Cumbre Vieja, SMF source modeling), Teresa Kraus and Tamara Eggeling (Graduate Research Assistants, MC simulations of SMF sources)

University of Delaware and University of Rhode Island 2010 -2011 (FY 10 -11) Accomplishments $393, 855 (Total Funding Received, 8/1/10 -7/31/12) Subtasks 1. 1 -1. 3 Completion • Literature Review • MC modeling of East Coast SMF • Simulation of co-seismic sources and flank collapse of Cumbre Vieja Volcano using 3 D multi-fluid VOF model • 100% - Report published • 75% - Final evaluation of sources underway • 100%

University of Delaware and University of Rhode Island 2010 -2011 (FY 10 -11) Accomplishments $393, 855 (Total Funding Received, 8/1/10 -7/31/12) Milestones for Tasks 1. 4 - 1. 5 Completion • Inundation modeling for high risk East Coast communities • Construct inundation maps for modeled East Coast communities • 20% - Work limited to DEM and GIS development • 20% -

Issues Encountered: Task 1. 2 • Questions about appropriateness of SMF events from a geological standpoint • Estimated sources are being examined by USGS personnel working with the UD/URI team. • This effort is expected to be completed (summer) • Questions about whether it is appropriate to use probabilistic results for slides, or whether worst-case scenarios based on historical input should form the basis of results for mapping. • UD/URI feel that there should be input from the NTHMP program on this question. • Guidance on this question could be improved after the planned workshop on probabilistic hazard analysis. (enough guidance to proceed obtained during MMS meeting) Both of these situations lead to Task 1. 2 extending into FY 12, beyond the planned completion in the original proposal

Task 1. 1: Literature review established large selection of potential sources

Computational domains for FUNWAVE propagation modeling

• Task 1. 3: An Overview of Distant Sources I: Flank collapse of Cumbre Vieja Volcano using 3 D multi-fluid VOF model 2 D slope stability computations (with 2 different models) on various cross sections. Geotechnical parameters are progressively reduced (mimicking hydrothermal alteration) until unstable state. 3 D slide volumes are inferred using a 3 D ellipsoid shape, based on the 2 D failure contours and geological considerations for lateral extent. Slide volumes range from 40 to 70 km 3, depending on the scenario. [Vizualisation of the quasi elliptic failure contour (Drucker-Prager failure criterion) calculated with the finite element code ADELI]

Set-up of 3 D tsunami generation model (THETIS) Geometry of THETIS domain Nested within larger scale domain for propagation model 3 -D view of THETIS domain

CVV impact on East Coast: New England

CVV impact on East Coast: Mid-Atlantic

CVV impact on East Coast: South-East (Georgia – South Carolina)

Overview of Distant Sources II: Caribbean subduction zone tsunamigenic earthquakes Summary • Historical examples include 1867 (7. 7 near US Virgin Islands); 1918 (7. 3 from PR); 2010 (7. 0 from PR) • Tsunami risk studied by many (e. g. , Zahibo and Pelinovsky 2001) • NOAA Forecast Source Database (Gica et al. 2008) modeled series of potential sources • Potential high risk to particular communities (e. g. , S. Carolina) SIFT sources (Gica et al 2008) used in CSZ simulation

Puerto Rico trench example Distribution of maximum surface displacements in Mid-Atlantic due to CSZ simulations • • • • Location: Direction: Dip: Rake: Magnitude: Slip: Depth: Length: Width: Shear modulus: Radius: Depth of slip: Water depth: 19. 674° N 65. 806° W 92° 165° 50° 9. 0 1. 0 m 40 km 600 km 150 km 4. 2 x 1010 kg/m s 2 200 km 400 m 7000 m

Overview of distant sources III: Azores-Gibraltar convergence zone Summary • • Region best known for 1755 Lisbon earthquake (8. 5 -9. 0 magnitude; 100, 000 deaths) Earthquake extensively studied (e. g. , Johston 1996; Baptista et al. 1998; Gutscher et al. 2006; Grandin et al. 2007) Source of 1755 quake not known; Barkan et al. 2009 simulated possibilities based on far-field effect Not included in NOAA Forecast Source Database Sources considered for East Coast Inundation Studies

Azores-Gibraltar convergence zone example Lisbon 1755 -like source • • • • Location: 36. 015° N 11. 467° W Direction: 345° Dip: 40° Rake: 90° Magnitude: 9. 0 Slip: 13. 1 m Depth: 30 km Length: 200 km Width: 80 km Shear modulus: 6. 5 x 1010 kg/m s 2 Radius: 300 km Depth of slip: 40 km Water depth: 4709 m

Status of Task 1. 2: Probabilistic analysis of coastal hazards associated with submarine mass failures (SMF) Summary • Past tsunamis caused by Grand Banks (1929) and Currituck (24 -50, 000 years ago) • 33% of US East Coast continental slope covered by landslide scars and deposits (Twichell et al. 2009) • Large number of submarine mass failures have been analyzed using Monte Carlo analysis in addition to historical examples Grand Banks source

Basis of MC analysis • Compiling surficial sediment data for the study area and enter into the GIS [Continental Margin Mapping (CONMAP) sediments grain size distribution for the United States East Continental Margin]

• Generation of cross-shelf/slope transects for slope stability analyses [Idealized coastline and transects for the upper east coast of the U. S. , Grilli et al 2009]

• Simplified tsunami propagation, runup, inundation in MC model => source selection for actual propagation with FUNWAVE [Tsunami propagation picked within an angular spread from the failed transect direction, with insert of Gaussian distribution of coastal runup (Grilli et al. , 2009)].

• Simplified coastline with names of corresponding coastal states, ranges of indices of studied coastal points, numbered N-S (Baxter et al. , 2011; Krauss, 2011). Points 1 -899 correspond to the upper East Coast already studied in Grilli et al. (2009).

Right panel: MC estimates of runups for Mid-Atlantic. Left panel: SMF transects (blue lines) used in MC analysis and location and size of underwater landslides causing 500 year runup (red ellipses).

Events identified in MC analysis and chosen for use in inundation mapping: Ongoing Task 1. 2 evaluation

Tsunami elevation computed with NHWAVE (up to 15 mins. ) and FUNWAVETVD, in a 500 m regional grid, for the first SMF source. (a) instantaneous elevation after 75 mins of propagation; (b) maximum envelope of elevation

University of Delaware and University of Rhode Island FY 12 Funding Plan Task (Activity) Name NTHMP Performance Metric(s) (from Strategic Plan) Supported Amount of Requested Funds Task 1. 2: Probabilistic analysis Successful execution of NTHMP tsunami of coastal hazards associated mapping, modeling, mitigation, planning and with submarine mass failures education efforts; Prioritize inundation map development $0 $25, 000 Task 1. 3: Simulation of transoceanic tsunamis from co-seismic and other relevant sources Successful execution of NTHMP tsunami mapping, modeling, mitigation, planning and education efforts $60, 000 $35, 000 Task 1. 4: Inundation modeling Successful execution of NTHMP tsunami for high risk East Coast mapping, modeling, mitigation, planning and communities education efforts $79, 854 Task 1. 5: Construct inundation maps for modeled east Coast communities $60, 000 Tsunami inundation maps that support informd decision making