WORKSHOP

Coastal DisastersRecent Tsunamis & Cyclone Storm Surges

Tomoya ShibayamaProfessor, Waseda University

Tsunami, Storm Surge, High Wave (Coastal Erosion), Earthquake, Liquefaction, Volcanic Eruption, Fire, River Flood, Drought, Landslide,

Basic Approach① Field Survey＋Numerical Simulation＋Hydraulic Experiment

Creation of Real Image of Disaster Common Images with Local Residents

② Variety of different scenarios of disasters in local conditions It is necessary to decipher the social context of disasters,

to prepare disaster reduction scenarios, andto work with local government staffs and local

residents.

1

Frequent Attacks of Tsunamis and Storm Surges

Recent Field Surveys of my own Number of Losses and Unknowns

2004 Indian Ocean Tsunami Sri Lanka, Indonesia,Thailand 220,000

2005 Storm Surge by Hurricane Katrina, USA 1,200

2006 Java Tsunami, Indonesia 668

2007 Storm Surge by Cyclone Sidr, Bangladesh 5,100

1970: 400,000 1991: 140,000 (Construction of Cyclone Shelters)

2008 Storm Surge by Cyclone Nargis, Myanmar 138,000

2009 Tsunami in Samoa Islands, Samoa 183

2010 Chile Tsunami, Chile 500

2010 Tsunami in Mentawai islands, Indonesia 500

2011 Tohoku Tsunami, Japan Death15,782 Unknown 4,086

2012 Storm Surge by Hurricane Sandy, USA (New York City) 170 (USA80)

2013 Storm Surge by Typhoon Yolanda, Phillipines 4,011+1,602

2014 Storm Surge in Nemuro, Hokkaido Island, Japan, 0

2018 Storm Surge by Typhoon Jebi, Osaka, Japan, 13

2018 Tsunami in Palu, Sulawesi, Indonesia, 2,081+1,3092

（Program Officer）Tomoya Shibayama, ProfessorMarch, 1977 B. E. in Civil Engineering, University of TokyoMarch, 1985 Dr. of Engineering, University of TokyoMay, 1985 Lecturer, University of Tokyo

March, 1986 Associate Professor, University of TokyoApril, 1987 Associate Professor, Yokohama National University (YNU)August, 1997 Professor, Yokohama National UniversityApril, 2009 Professor, Waseda University Emeritus Professor, YNU

Professor Tomoya Shibayama is one of the top technical experts of tsunami and storm

surge disaster mitigation in Japan. He performs hydraulic laboratory experiments, field surveys and numerical simulations for his mitigation study. He served as a team leader of all major tsunami and storm surge events in Japan and overseas over the last fifteen years.

Background• Japanese islands are so-called “Disaster Islands”.• There is no concrete national strategy and procedure for protection of residents. Impact on Society• Change viewpoints from the government to individual resident. • Quantify complex risk at citizen’s level.• Disaster risk is added to individual's needs such as selection of residents, properties,

schools, super markets etc. • Change the future land-use based on the individual selection• Based on above implications, Japanese people gradually withdraw from high-risk areas

in next 100 years.• New social system is proposed for gradual change to conquer local vulnerability by

relocating citizens, based on Japanese social changes including declining birthrate and aging.

Challenges and Possible Impact to Society

Remarkable Innovation

Approach to Success• Based on the most advanced research results of disaster studies in Japan, we will establish a

concept of complex disaster. Individual disaster such as Earthquake, Tsunami, Storm Surge are re-organized to time history of a complex disaster such as, (1) Earthquake, Fire, Liquefaction, Tsunami, (2) Typhoon, Strong Wind, Heavy Rain, Storm Surge, Flood or (3) Earthquake, Volcanic Eruption, Volcanic Ash Distribution , Flood.

• We organize co-operative research system and promote joint research in project group. Management Strategy• A part of the project will be planned based on public offering.• Every year, we examine members of the project and will change the members. New

members will be invited if necessary. • The Program Officer participate weekly research group meeting and make all information

open to all members.• Project members have individual consultation with the Program Officer every 2 weeks and

deliver presentation of progress report to all members every one month. These accelerate development speed and harmonize the cooperation of members.

Final Goal• Final Judgement Scale: Completeness and Practicality of the Package of New Risk Evaluation

System of Complex Disaster and Social System.• Required Level：High Level Quantification to Change Individual Decision Making Process.

Risk of Project : It is necessary to quantify different risks in the same level of accuracy.

Breakthrough Limitation of Existing Technology:• Residents continue to live in the same area.• Risk is not reduced. How to exceed the limitation? • Individual risk is quantified. Final Goal:• No loss of lives due to disasters.

Scenario for Success and Goal

Institute of Future Sustainable Society, Waseda University

Support Acceleration

Social MovementIndividual Decision

• Image of Complex Disaster

• Quantified Risk

• Education of Disaster Consultants

• New Social System

From Government to Individual Citizen

Reduction of Total Risk of

Complex Disaster

Relocationof Land Use

Disengagement from Disaster Archipelago

Individual Behavior of Selection Reduces Change of Land-use

Timely Social System Accelerates Social Changes

High Disaster Risk Difficulty in Complex Disaster

Citizens Know and Select Their Safety

What should you do if a Tsunami was coming to your area in 40 minutes?

Take our first edX course

Study the mechanisms of coastal disasters

Plan for disaster evacuation

Tsunamis and Storm Surges: Introduction to Coastal Disasters

Prof. Tomoya ShibayamaA leading researcher of coastal disaster prevention and coastal engineering at Waseda University

Waseda edX

Started January 18th, 2016 Take the course for freeNow self-paced

Enroll Now!

2500 learners 120 countries

USA（19％）Japan（15％） India（６％） U.K.

（４％） Chile（３％） Canada（３％） Spain（３％） Indonesia（３％） Netherland（２％）Philippines（２％）

MOOC( Massive Open Online Course)

4

5

Recent Progress of Numerical Simulation and Hydraulic

Laboratory Experiment for Tsunami and Storm Surge Study

Tomoya ShibayamaProfessor, Department of Civil and Environmental Engineering

Takahito Mikami and Ryota Nakamura

6

Frequent attacks by tsunamis and storm surges

Field Surveys New Findings New Analysis

1) Mechanism of Disasters ---- Hydraulic Laboratory Experiment, Turbulence Numerical Model

2) Change of Storm Trend due to Global Warming----- Improvement of Numerical Prediction Methods and Future Predictions

3) (Social Study of Coastal Region) Social Stratification, Preparedness

7

Important findings through field surveys to be examined by laboratory tests.

1. Different Wave Behaviors Solitary Wave (Java Tsunami)Dam Break WaveContinuous Overflow (Tohoku Tsunami)Bore Wave: both for tsunami and storm surge events

Cyclone Sidr (Shibayama et al., 2009)Typhoon Yolanda (Roeber and Bricker, 2015, Nakamura et al., 2016)

2. Coastal dykes failed due to a continuous and prolonged water overflow for fifteen minutes or more, such as during the 2011 Tohoku Tsunami or the storm surge during hurricane Katrina (see Shibayama, 2015). 3. Coastal forests played an important role to increase water depth behind coastal dykes and prevent the appearance of supercritical flow over dykes (Tohoku Tsunami, Matsuba et al., 2014).4. Inundation patterns were strongly influenced by the local topography in Tohoku Tsunami, Indian Ocean Tsunami, storm surges due to hurricane Katrina (Shibayama et al., 2005) and the storm surge in Nemuro.5.Detached breakwaters are a common way to protect against wind waves along the shorelines in Japan, though these can offer some degree of protection against tsunamis. 6.The transportation of debris by flooding water can exacerbate the destruction caused to structures and houses (storm surge due to hurricane Katrina and typhoon Yolanda, Chilean tsunami, Tohoku Tsunami).

Tohoku Tsunami: Japanese Debates on the Reconstruction Process

The idea that hard measures can protect against the loss of life has been discarded.

Level I Tsunami Protection Height

1. The function of coastal structures would be to attempt to protect property or to help

evacuation process against the more frequent but low-level events (typically with a return

period of several decades to 150 years). “Once for 100 years” Coastal protection

structures are designed for this tsunami height.

Level II Tsunami Protection Height

2. Soft measures (Evacuation), on the other hand, would be used to protect lives, and be

designed with more infrequent higher level events (with much longer return periods, for

example 1,000 years). “once for 1000 years”

For Level II tsunami, structures are overflowed but are required not to be destructed. They are

expected to reflect tsunami partially and will assist evacuation process by reducing tsunami

height and delaying tsunami flood time.

4 m

9 m

0.5 m1 m

0.5 m

chambers

air valves

4 chambers

4 air valves for each chamber

connected to vacuum pump

slope(1/20)

slope(1/10)

2 m 2 m0.5 m

: Current Meter

sho

relin

e

detached breakwaters

: Wave Gauge

2 m

ld

lo

9

Survey Report of

the 2018 Sulawesi Tsunami(Septermber 28, 2018)

Waseda Tsunami Survey Team

Team Leader: Prof. Tomoya Shibayama

10

27-31 October 2018

Research Institute of Sustainable Future Society

Waseda University

Members are from Indonesia, Japan, USA, Canada, and

Germany

11

1）Survey of tsunami inundation heights distribution

2) Interviews and questionnaire survey to residents on Tsunami behavior and their evacuation

3）Three dimensional image processing by using a drone

4）Sonner sounding of shallow water bathymetry in Palu Bay

5）Survey of broken structures and buildings

Field Interviews

• All the residents mentioned that three wave hit the coast throughout the bay (going from smallest to largest), which does not necessarily fit the multiple landslides narrative.

• The Tsunami Inundation Survey team also talked to residents closer to the epicenter who observed a wave travelling south (towards Palu Bay) so the wave also likely contributed to the tsunami in Palu.

Tsunami Source SurveyField Investigation of 2018 Palu City Tsunami

13

Tyhoon Yolanda

Track data: The Regional Specialized Meteorological Center (RSMC) Tokyo Best Track Datahttp://www.jma.go.jp/jma/jma-eng/jma-center/rsmc-hp-pub-eg/trackarchives.html

Time of Generation: 2013-11-04 00:00 UTCTime of Disappearance: 2013-11-11 06:00 UTCDuration: 174 hoursMinimum Pressure: 895 hPaMaximum Wind Speed: 64.3 m/sNumber of Victims: 6201 (Jan. 14 in the Philippines)Number of Unknowns: 1785(Jan. 14 in the Philippines)

14

Coast levee or Dike

Typhoon, Cyclone, Hurricane

(1) Wave (Run-up)

Wind

(2) Wind Driven Surge

(3) Pressure Surge

(4) Tide

Future Predictions-------Numerical SimulationNakamura, R., Shibayama, T., Ohira, K., Tasnim, M.K. (2014)

*(2) + (3)= Storm SurgeFig. An image of phenomena when typhoon

Components of Storm Surge

Second Findings: Change of Storm Behaviors due to Climate Changes

Rapid Development: Yolanda (2013), Nemuro (2014)Route Change : Nargis (2008)

WRF

XTide

Result

Wind speed, Pressure, Heat flux, Vapor flux, Rain fall

Storm Surge (Wind + Pressure)

Wind Waves, Wave setup←S-WAVE(SWAN by Delft Univ. of Tech, Booji et al., 1999)

Weather Research and Forecasting model

Mesoscale Model, Skamarock et al, 2008)

Tidal Prediction

GFS Data (NOAA)Meteorological data of

whole globe

Choosing Area

FVCOMThe Unstructured Grid Finite Volume Coastal Ocean Model

(Chen et al., 2003)

Tide

TC-Bogus Scheme

Improve vortex condition in the center

15

MIROC5

Future Prediction of Meteorological Condition

Surfece Temperature of

Ocean and Ground

Storm Surge Prediction Model (Nakamura and Shibayama, 2014)

Coupled Weather-Storm surge-wave-tide model

WRF-FVCOM-Xtide-MIROC5

(Watanabe et al., 2008)

(Flater, 1998)

(Kurihara et al., 1993)

NUMERICAL SIMULATION OF TROPICAL CYCLONES ANDSTORMSURGES INTHEARABIANSEA

Mohsen Soltanpour, K. N. Toosi University of TechnologyZahra Ranji, K. N. Toosi University of Technology

Tomoya Shibayama, Waseda UniversitySarmad Ghader, University of Tehran

Shinsaku Nishizaki, Waseda University

K. N. ToosiUniversity of Technology

Waseda University

Tehran University

Study Area

Motivation

Numerical

Simulation

Conclusion

3/19

Gonu, 2007 Ashobaa, 2015

Target

Cyclones

FUTURE WAVE PROJECTION DURING THE TYPHOON AND WINTER STORM SEASON

1Shinsaku Nishizaki, 2Ryota Nakamura, 3Tomoya Shibayama, 4Jacob Stolle

1. Shikoku Electric Power Co., Inc.

2. Toyohashi University of Technology

3. Waseda University

4. University of Ottawa

Figure 1. Severe climate events around Japanese islands

Figure 2. Historical events in Oct. and Dec., 2014(image after; Japan Meteorological Agency)

12/17

9AM

10/14

9AM

• 2 Typhoons, Phanfone and Vongfong, landed at Japanese islands in October 2014 and maximum recorded significant wave heights were observed

• One extratropical cyclone developed at northern Japan and caused a storm surge in Nemuro, Hokkaido in December 2014

• The year 2014 is considered to be one of the most active years in terms of wave heights

2. Target area and period 19

Predict future waves

Calculate significant wave heights and periods

Wave Model SWAN

（Booij et al., 1999）

Input wind results extracted from WRF simulation

Meteorological Model WRF（Skamarock et al., 2008）

WPS WRF＋

Give initial and boundary conditions

Meteorological data (NCEP FNL)

Sea surfacetemperature

Temperature

Atmospheric pressure

Relative humidity

Wind speed etc…

Ensemble results of 26 GCMs are used for the difference of SST between present and future cases.

Sea surfacetemperature

Meteorological data (NCEP FNL)

Sea surfacetemperature

Temperature

Atmospheric pressure

Relative humidity

Wind speed etc…

3.2 Methodology (Future Case) 20

domain 1 122.1°E ~ 151.9°E 19.6°N ~ 49.4°N

domain 2 125.6°E ~ 147.8°E 23.9°N ~ 46.8°N

domain 1 0.15° 201 × 201

domain 2 0.05° 445 × 460

Area

resolution

Input data intervals

Calculation intervals

Number of vertical layers

WSM 3-class simple ice scheme

36 layers

60 s

6 hours

Micro physics

rrtmg schemeShortwave radiation

Meteorological data FNL ( 1°× 1°)

10/01/2014 00:00 ~ 11/01/2014 00:00

Domain

settings

12/01/2014 00:00 ~ 01/01/2015 00:00Calculation Period (UTC)

Geography data USGS

Longwave radiation

Planetary Boundary Layer

rrtmg scheme

YSU scheme

Table 1. Calculation conditions for WRF

Table 2. Calculation conditions for SWAN

Figure 4. Calculation domains for WRF and SWAN

Kanazaw

a

Shizuoka

Omaezaki

domain 1 122.5°E ~ 151.0°E 20.0°N ~ 48.5°N

domain 2 127.0°E ~ 147.5°E 25.0°N ~ 46.0°N

domain 1 0.15° 190 × 190

domain 2 0.05° 410 × 420

10/01/2014 00:00 ~ 11/01/2014 00:00

12/01/2014 00:00 ~ 01/01/2015 00:00Calculation Period (UTC)

Domain

Settings

Area

resolution

Calculation mode Non-stationary / 2 dimension

Whitecapping Komen

Direction division number 36 (θ=10°)

Frequency division number 39 (0.025 ~ 1.0 Hz)

maximum number of iterations 5

time step 5 min

Bathymetry data GEBCO2014

3.3 Calculation conditions 21

• Although the SST increase resulted in more intense typhoons and

it led to higher Hs on the Pacific Ocean in October, the effects of

SST increase were less important for the extratropical cyclones in

December.

• It is clearer that Hs < 2.0 m decreased and 2.0 m ≤ Hs < 5.0 m

increased on the Pacific Ocean in October when more intense

typhoons were formed.

• It is probable that the frequency distribution of significant waves

will change around the area where future SST increment is

dominant.

6. Conclusion 22

36th International Conference on Coastal Engineering August 1, 2018

23

TSUNAMI CASUALTY ESTIMATION CONSIDERING

INTENDED EVACUATION BEHAVIOR

OF LOCAL RESIDENTS AND VISITORS

Tomoyuki TAKABATAKE, Tomoya Shibayama, Esteban Miguel

Waseda University, Tokyo, Japan

36th International Conference on Coastal Engineering August 1, 2018

1. Introduction

24

Most studies have expressed evacuation start time for evacuees in a very simple

way (e.g. start at the same time / follow a simple distribution function)

A variety of evacuation triggers that can prompt evacuees to start their

evacuation should be considered (Mikami & Shibayama 2016)

Most studies have assumed that evacuees proceed to the closest evacuation place

via the shortest route (the detailed intended evacuation behavior of visitors has

been neglected)

Visitors’ evacuation route choice should be considered (Takabatake et al.

2017)

This study proposes an agent-based tsunami evacuation simulation model that

considers a variety of types of intended evacuation behavior of local residents,

tourists and beach users

The developed model was applied to Yuigahama Beach in Kamakura City, Japan

to estimate to estimate the potential casualties that can result from a tsunami

event

36th International Conference on Coastal Engineering August 1, 2018

4. Results - Casualty estimation -

25

Keicho earthquake & tsunami

The number of evacuees and the age distribution ratio (0-14, 15-64, 65>) was set

based on the study area’s census (2015).

Local residents: 5,778 / Tourists: 5,877 / Beach users: 9,840

100m 200m0

The Tohoku Re-Construction Process after the 2011 Tsunami

Tohoku Tsunami: Natori City Coastal Erosion

Teizan Canal

Original Shoreline

Three different approaches

1. Construction of high coastal dike again and protection of low land area against the next tsunami attack. (Tarou)

2. Making artificial elevated area over old downtown in low land area by land fill. Use the elevated surface for residential area. (Rikuzentakata)

3. Movement to higher hill side and developing new residential area. (Onagawa)

Those methods are commonly used in part in these cities.

Tarou

Height of Tsunami Barrier wall is changed

from 10m to 14.7m.

New Wall

Old Wall

Rikuzentakata

Raise the Ground Level Elevation for 10-12m

There are selections for residents.1) Natural Hill (45 ha)2) New Artificial Hill (91 ha)

New Artificial Hill

Onagawa

Change of Land-use:

Downtown in Low Land

No Residence, Open space

Construction of New Residence Area in Natural Hill

New Residential Area

Old Downtown

Arahama

Movement

All residents moved to a new area in inland.