ocea

nnet

wor

ks.c

a

Port Alberni / Sidney

March 27th - 28th 2014

A Tsunami Detection Initiative for British Columbia

A N I N I T I A T I V E O F

2

3

Copyright 2014 Ocean Networks Canada

Ocean Networks Canada

Technology Enterprise Facility (TEF)

University of Victoria

PO Box 1700 STN CSC

2300 McKenzie Avenue

Victoria, BC, Canada, V8W 2Y2

250.472.5400

info@oceannetworks.ca

www.oceannetworks.ca

4

Table of Contents

Introduction 6We are Tseshaht [Ts’ishaa7ath] 7A message from Port Alberni Mayor John Douglas 9A letter from Ocean Networks Canada President Kate Moran 10About Ocean Networks Canada 11Smart Ocean Systems™ 12Detecting Tsunamis on the NEPTUNE network and the ONC Tsunami project 13

Abstracts 15Hydroacoustic waves modeling of Tsunami Early Warning Systems (TEWS) 15Uncertainty reduction in near field tsunami early warning 17Modelling of tsunami waves at the Institute of Ocean Sciences 18New NOAA Tsunami Forecasting System 18Tsunami wave impact on walls and beaches 19GPU-accelerated hydrodynamics for wave impact problems 20Modeling hazards from seismic and SMF sources 21Assessing the tsunami hazard of the British Columbia coastline 22Tsunami Modeling and Inundation Mapping in Alaska: Development of maximum credible tsunami scenarios 22In the wake of the 2011 Tohoku Tsunami: Modeling of tsunami wave propagation and evacuation. 23Utilizing power of present and future open-ocean tsunami monitoring networks 24Indian Tsunami Early Warning System 25Cascadia Megathrust Rupture Models for Tsunami Modelling 26The Activities of the US Tsunami Warning Centers During the Haida Gwaii Tsunami 27

Summary of Discussions 29Instrumentation 29Earthquake Source Definition 32Tsunami modeling 36

Attendees Biographies 41

5

“The foggy dawn of Saturday, March 28, 1964, broke on a stunned and unbelieving populace in the Alberni Valley. Residents were stunned by the small glimpse they’d already had of the fantastic damage done to their communities… unbelieving at the reports that no one had been drowned or even seriously injured in the tidal wave which swept the low-lying areas in the midnight hours following the great Alaska earthquake.”

Civil Defence Circular.

Special report on Alberni tidal wave disaster.

Provincial emergency program report.

6

IntroductionFifty years ago the most powerful earthquake ever recorded in North America hit south

central Alaska, causing a tsunami that swept along the BC coast, heavily damaging

coastal communities such as Port Alberni. Fifty years later, much has been learned

about how the Earth’s crust works, but ocean scientists are still working to understand

the dynamics of the great waves.

On March 27th and 28th, 2014 Ocean Networks Canada hosted an international

workshop on tsunami modeling and instrumentation. Forty-four attendees from

Canada, Japan, India, USA, Germany, France and Italy congregated in Port Alberni,

British Columbia, on the 50-year anniversary of the tsunami that devastated this city.

The shore station that hosts the entry of the cable from the NEPTUNE observatory

was the venue for this event. The Tseshaht First Nation and the Mayor of Port Alberni

opened the day with a welcoming followed by a full day of presentations from scientists

from the major agencies responsible for tsunami mitigation. The most recent tsunami

research was presented with an emphasis on the coast of British Columbia and the

geohazards present along the Cascadia margin. The second day of the workshop,

hosted at the Institute of Ocean Sciences in Sidney, focused on near-field tsunamis

and the instrumentation currently available for their detection. The experts discussed

the design and technology that should be implemented for the detection and forecast

of near-field tsunamis on the British Columbia coast. Three groups were formed that

centered their discussions on (1) instrumentation, (2) earthquake source definition and

(3) modeling.

This report summarizes the information shared during the workshop. Videos and

presentations are available on our website: http://www.oceannetworks.ca/science/

getting-involved/workshops/tsunami-workshop-2014.

7

We are Tseshaht [Ts’ishaa7ath]

The Tseshaht First Nation is a vibrant community on the West Coast of Vancouver Island

in Port Alberni, British Columbia, with an active and progressive natural resources-

based economy. We are proud of our culture and work as a community to preserve our

traditional values and the teachings.

Tseshaht translates as “the people of Ts’ishaa,” a place

on what is known today as Benson Island, one of the

Broken Group Islands in Barkley Sound. We are one

of the 14 Nations that make up the Nuu-chah-nulth

[Nootka] people of western Vancouver Island.

At the core of Tseshaht is our chronicle of creation; our

spiritual origin. We were created at Ts’ishaa, and our

first ancestors (Tseshaht man and women) were given by Nas (creator) the highest

spiritual responsibility and stewardship of the Broken Group Islands.

Our ownership of land is based on the Nuu-chah-nulth laws of hahuulhi, which means

the territory of a nation under the stewardship of a King (Head of state). Being king

indicated the closest spiritual bloodline to our chronicle of creation.

Our late King Adam Shewish, great-grandfather Chief Haayuupinuulh [born c.1830]

was the King of the Tseshaht when mum-ulth-ne settlement started in the Alberni

Valley. His name, meaning “getter of ten (whales)” signified his status as a prominent

chief, including his King’s right to hunt whales.

8

Tseshaht hahuulhi changed through marriage and alliances, a series of wars, and the

incorporation of affiliated groups. A prime example was a historic conflict with the

Tsumas7ath, already living on the Somass River, the Hikwuulh7ath and the Hach’aa7ath

established themselves at on the Somass River.

These hahuulhi enhancements continued as the Tseshaht absorbed the Hikwuulh7ath

and the Hach’aa7ath. As well as once exclusive properties of the Nash7as7ath,

Maktl7ii7ath, Ts’umaa7as7ath and the original Ts’ishaa7ath.

This meant that the hahuulhi of the these groups in the Broken Group Islands, central

Barkley Sound, a large extent of the Alberni Inlet and the Alberni Valley became

Tseshaht territories.

Tseshaht Tutuupata, the plural of tupaati, refers to the hereditary privileges or preroga-

tives that governed the ownership and use of practically everything of value in Tseshaht

society. These included resources like rivers, fish trap sites, and plant gathering sites, as

well as intellectual property resources like names, ceremonial songs, dances, and regalia.

Tutuupata determined rank in Tseshaht society, and were inherited within a family.

Tseshaht Seasonal Round

The traditional Tseshaht economy was determined by tupaati - the ownership of resources.

Tseshaht tupaati included both “outside” and “inside” resources throughout this territory.

This meant that in late winter and early spring the Tseshaht travelled to their “outside”

tupaati to utilize the resources of these traditional sites such as sea mammals, halibut,

rockfish and salmon and procurement areas in Barkley Sound. As the seasons changed,

the resources changed, and the Tseshaht moved back to their “inside” tupaati, following

the salmon up Alberni Inlet to the Somass River.

The Tseshaht now exert our jurisdiction over our hahuulhi based on our history.

9

A message from Port Alberni Mayor John Douglas

It gives me great pleasure to extend a warm

welcome to all the delegates and scientists

attending the international tsunami modeling

workshop hosted by Ocean Networks Canada

here in Port Alberni.

The waves that hit our community 50 years

ago left an indelible mark on the memories

of those who experienced it and forever

changed our understanding of the local

dangers posed by tsunamis.

The City of Port Alberni and the Alberni-Clayoquot Regional District are working closely

together to plan and prepare for such emergencies. Your knowledge and expertise

represents an important link to these efforts and our goal to be a world leader in the

field of emergency preparedness.

As an organization focused on our ocean planet and its complex systems, I congratu-

late Ocean Networks Canada for the important contributions they are making to the

advancement of science and technology. The City of Port Alberni is honoured to be part

of this important gathering.

Yours truly,

City of Port Alberni,

John Douglas - Mayor

10

A letter from Ocean Networks Canada President Kate Moran

On behalf of Ocean Networks Canada, an

initiative of the University of Victoria, I’m

delighted to welcome you, the world’s leading

tsunami researchers, to the shore station

of the 840-km long NEPTUNE cabled ocean

observatory.

Our sincere thanks go to Mayor Douglas and

the citizens of Port Alberni for their warm

welcome and for including us in community

events that mark this very historic time – the

fiftieth anniversary of a devastating earthquake

and tsunami. I also am grateful to the Tseshaht First Nation for sharing their ancestral

traditions with us.

This workshop location—at the terminus of the NEPTUNE cabled observatory in the

Northeast Pacific—is where, fifty years ago, a powerful tsunami came surging up the

Alberni Inlet and inundated the shoreline.

Because of this anniversary, it is particularly fitting to bring all of you together to Port

Alberni at this time.

I wish you all the best for a successful workshop and look forward to an exciting future

filled with rich collaborations.

Sincerely,

Kate Moran, PhD

President and CEO

11

About Ocean Networks Canada

Ocean Networks Canada operates world-class ocean observatories off the west coast

of British Columbia for the advancement of science and the benefit of Canada. The

NEPTUNE and VENUS cabled observatories collect data on physical, chemical, biological,

and geological aspects of the ocean over long time periods, supporting research on

complex ocean and Earth processes in ways not previously possible.

Figure 1: Ocean Networks Canada’s observatories in British Columbia.

The NEPTUNE regional observatory and VENUS coastal observatory, which have been

operating for four and seven years respectively, provide unique scientific and technical

capabilities that permit researchers to operate instruments remotely and receive data

at their home laboratories anywhere on the globe in real time. These facilities extend

and complement other research platforms and programs, whether currently operating

or planned for future deployment. The Ocean Networks Canada Innovation Centre

promotes the advanced technologies developed by NEPTUNE and VENUS.

Ocean Networks Canada contributes significantly to the international ocean research

and observing enterprise by providing infrastructure below the seafloor, on the

seafloor, in the water, and at the sea’s surface in a wide range of ocean environments.

Our network is especially well suited to support research that requires continuous data

collection from a diverse array of ocean parameters over long time periods because

of their pivotal role in enabling a more complete understanding of ocean and Earth

12

processes. Real-time flow of data to on-shore laboratories and data centres permits

rapid analysis of information on natural hazards such as earthquakes, tsunamis, storm

surge, and underwater landslides.

Because the cabled infrastructure is fixed on the seafloor, observatory data collection

is being complemented with ever more sophisticated platforms such as gliders,

unmanned undersea and aerial vehicles, and remotely-located observatories.

NEPTUNE and VENUS share the same data management and archive system called

Oceans 2.0. It provides users with open access to real-time and archived data and

supports a collaborative work environment. The University of Saskatchewan hosts the

observatories’ data backup system, which collects over 80 terabytes per year.

Smart Ocean Systems™

The Innovation Centre created an integrated portfolio of ocean observing products and

services branded as “Smart Ocean Systems™”. The Centre’s portfolio of Smart Ocean

System™ products and services (Figure 1) fall naturally into three general categories:

•  Sensors and Instruments

•  Ocean Observation Systems

•  Ocean Analytics

Figure 2: Smart Ocean Systems™ portfolio of products and services

These products and services address growing global demand for informed and rapid

decision making in response to marine monitoring or geo-hazard events such as

earthquakes, tsunamis, oil spills, and vessel navigation.

Smart Ocean Systems™ are advanced ocean sensor and instrument technologies that

13

capture data on events such as changes in temperature, currents, dissolved gasses,

seismic, sound and video recordings triggered by natural or anthropogenic change.

Ocean Observation System technologies comprising:

• Ocean Observing Technology: hardware infrastructure including undersea

high-voltage and fibre-optic cable, connection nodes, highbandwidth com-

munications, robust servers and storage; and

• ONC’s digital infrastructure, Oceans 2.0, which is a data management system

for sensor interfacing, data capture, storage and archiving, manipulation and

annotation, and internet presentation; and

• Ocean Analytics, which are the discovery and communication of meaningful

patterns in ocean data resulting in information-based decision support

products and services that allow informed responses by researchers, industry

and a broad range of other users.

Detecting Tsunamis on the NEPTUNE network and the ONC Tsunami project

The NEPTUNE cabled observatory array off the west coast of Vancouver Island detected

the passage of the tsunami wave generated by the far field Tohoku earthquake and

tsunami event of 2011. The Bottom Pressure Recorders at three different nodes along

the length of the NEPTUNE cabled network measured the wave height at different times

as it approached the shores of Vancouver Island and as shown in the graphic below.

The NEPTUNE array can be used to detect near field events with existing instrumentation

including seismometers and hydrophones with the potential of incorporating High

Frequency coastal RADAR to directly measure and profile incoming tsunami waves

several minutes before they impact the shoreline.

The ONC Tsunami project is a shared initiative between ONC and the University of

Victoria (UVic). This program is designed to conduct high speed analysis of NEPTUNE

sensor data from a near field tsunami event and combine these data with tsunami

model algorithms in order to simulate the inundation impact onshore. A near-field

seismic event detected at the Cascadia subduction zone may yield only 15 minutes of

14

warning between the start of an event and the wave impact on the western shoreline

of Vancouver Island.

The University of Victoria Computer Science department program has been working to

calculate complex tsunami models and provide predicted inundation maps within only

a few minutes by optimizing the processing time of models using High Performance

Computing infrastructure with potentially hundreds of parallel CPU’s. By combining

with ONC’s infratsructure, existing international operational systems and research, the

ONC Tsunami system could prototype near-real-time forecasts that consider unique

BC geography and conditions. The simulation models will use one or more cities on

the west coast of B.C.’s Vancouver Island as case studies and will be designed in such a

way that it can be reconfigured for other locales that are exposed to tsunami risk. The

system will be designed to be open and flexible enough to support different applications

that require high speed simulation modelling and data fusion from multiple sources.

Figure 3. Tsunami detection in the Ocean Networks Canada marine observatories. Credit NOAA

15

AbstractsHydroacoustic waves modeling of Tsunami Early Warning Systems (TEWS)

Ali Abdolali1, Claudia Cecioni1, Giorgio Bellotti1, Paolo Sammarco2

1University of Roma TRE, 2University of Roma Tor Vergatat

The Tsunami Early Warning Systems (TEWS) are the most effective tool to reduce or avoid human victims of tsunamis.

State of the art tsunami warning procedures currently relies on seismic and sea level measurements. TEWS must show

some mandatory features in order to be effective, as: fast prediction, reliable response and efficient alarm. Tsunamis have

the energy to propagate across the ocean, therefore the longer are the distances covered, the longer is the time to spread

the alarm and more effective actions can be taken by local emergency authorities in order to save lives and properties.

However, in the areas where the source area is close to the populated coasts, the traveling time of devastating tsunamis

are short. Therefore, it is unfeasible to wait the measurement of the tsunami itself before spreading the alert.

Detection of precursor components of tsunami waves, i.e. pressure waves generated due to the water column compression

triggered by sudden seabed movement, could therefore significantly enhance the efficiency and promptness of TEWS.

This low frequency waves propagate with the speed of sound in water, which is significantly larger than the long wave

tsunami celerity. The records of the faster hydro-acoustic waves can cope with the shortening decision time of spreading

the alarm and enhance the accuracy and trustworthiness of TEWS. Here we present a numerical model based on a depth-

integrated equation for the complete reproduction of these waves.

HYDRO-ACOUSTIC WAVE MODEL.

The idea of using hydroacoustic wave detection for tsunami warning dates back to the work of Ewing et al. (1950). Several

analytical investigations have been carried out to study the physical characteristics of acoustic waves generated by sudden

displacement of sea bottom. It has been proved that there exists a relationship between the tsunamigenic source and

the hydroacoustic waves. Accurate computations of hydroacoustic waves can be carried out using potential flow three

dimensional models. However such category of models is computationally expensive and cannot be easily applied to large

domain. We have therefore developed a numerical model based on the solution of a hyperbolic mild slope equation, valid

in weakly compressible fluids (MSEWC), detailed in Sammarco et al. (2013):

Equation (1) is expressed in terms of fluid velocity potential at the free-surface ψ_n (x,y,t).

16

The model gave us the opportunity of investigating the propagation of such waves in the far field.

Recent experimental evidence of the existence of low-frequency elastic waves generated by the seabed motion has been

found during the Tokachi-Oki 2003 tsunami event. The observatory of the Independent Administrative Institution, Japan

Agency for Marine-Earth Science and Technology (JAMSTEC), detected the pressure signal induced by the earthquake

(Nosov & Kolesov, 2007). In addition, during occurrence of 2012 Haida Gwaii earthquake, Ocean Network Canada

measurement in the southern side of earthquake zone showed fast elastic oscillation collected by bottom pressure

gauges. These two events are the only available records of hydroacoustic waves with appropriate sampling frequency.

LARGE SCALE APPLICATION TO 2012 HAIDA GWAII TSUNAMI

The current study is the first large geo-

graphical scale application of the numerical

model based on eq. (1). The model is

applied to simulate the hydroacoustic

wave propagation, generated by 2012

Haida Gwaii earthquakes (Fig.1). The

model has been validated against the

full three-dimensional weakly compress-

ible model. The results comparison

with measurements collected by Ocean

Network Canada will be presented in the

conference. The general characterization

of these pressure waves depending on

the generation mechanism, the source

location, the bottom topography and the

depth of the pressure recording point

can be identified from collected and

simulated hydroacoustic signals. According to the numerically reproduced scenario results, we shall propose preliminary

indications for the implementation of innovative TEWS based on hydroacoustic wave measurements tsunami warning this

methodology can even be extended to an inversion procedure, in which the tsunami sources can be reconstructed from

sea level observations. This leads to accurate assessment of the approaching tsunami hazard.

References

Nosov, Kolesov, (2007). Elastic oscillations of water column in the 2003 Tokachi-Oki tsunami source: in-situ measurements and 3-D numerical

modelling. Nat. Hazards Earth Syst. Sci. 7 (2), 243–249.

Sammarco, Cecioni, Bellotti and Abdolali (2013). Depth-integrated equation for large scale modelling of low-frequency hydroacoustic waves.

Journal of Fluid Mechanics, 722(R6), doi:10.1017/jfm.2013.153.

Figure 1 – Snapshots of the free-surface hydro-acoustic perturbation given by the

2012 Haida Gwaii earthquake.

17

Uncertainty reduction in near field tsunami early warningHow to combine HF radar-based and conventional sea level monitoring, seismic and GNSS observations with advanced modeling

Jörn Behrens1, Thomas Helzel2, Anna Dzvonkovskaya2, Klaus-Werner Gurgel1

1University of Hamburg, Germany, 2Helzel Messtechnik, Kaltenkirchen, Germany

Traditional tsunami early warning systems (TEWS) rely mainly on seismic observations for early detection of potentially

tsunamogenic earthquakes. Additionally, sea level observations from deep-sea stations (buoys) and coastal tide gauges

are used to confirm or cancel the assessment based on earthquake characteristics. In far-field tsunami warning this

methodology can even be extended to an inversion procedure, in which the tsunami sources can be reconstructed from

sea level observations. This leads to accurate assessment of the approaching tsunami hazard.

In near-field tsunami warning operations this approach is not feasible, since the time needed to gather and invert from

sea level observations is too long. Therefore, hazard assessment in this situation is much more dominated by uncertainty.

Usually conservative assumptions are made on the connection between earthquake magnitude and potential hazard,

leading to large numbers of false warnings.

Recently, additional observation methods have greatly improved the ability to accurately assess potential tsunami hazard.

Global navigation satellite systems (GNSS) allow for accurate determination of crust deformation. The actual rupture

mechanism can be estimated from these measurements in near real-time. HF radar based observations of the tsunami

wave make it possible to continuously monitor a larger coastal area for incoming wave hazard prior to reaching coastal

sea level gauges, and more importantly settlements and installations.

These additional observational data can help to reduce the uncertainty in near-field tsunami warning operations. However,

the quest is to find a suitable way of combining such observational data with a modeling framework. Modeling assisted

tsunami warning is required, since the number and spatial/time distribution of measurements does not allow for covering

the whole forecast region.

In this presentation, we will introduce the characteristics of different observational methods for tsunami detection and

monitoring, emphasizing the application of HF radar. We will describe the uncertainty involved in near-field tsunami early

warning and derive a rigorous mathematical formulation for capturing this uncertainty. From this we derive a method

to assimilate the available data into a modeling framework for tsunami early warning. Examples from an operational

tsunami early warning system in Indonesia will demonstrate the applicability and usefulness of the approach.

18

Modelling of tsunami waves at the Institute of Ocean Sciences

Josef Cherniawsky and Isaac Fine, Fisheries & Oceans Canada, Sidney, BC

We plan to present an outline of recent tsunami modelling activity at the Institute of Ocean Sciences, showing several

examples of tsunami model results for the British Columbia coast. These include tsunami models for historical and more

recent subduction type earthquakes and for submarine landslides. Such models can be used to generate community-based

tsunami inundation maps for plausible worst-case scenarios, for tsunami risk assessment and for tsunami warnings, thus

replacing the current coarse system of tsunami notification zones. The largest inundations risk on the southern BC coast

is from a future Cascadia subduction zone (CSZ) earthquake and tsunami. Therefore most of the scenarios considered are

for this potentially very destructive event.

New NOAA Tsunami Forecasting System

Marie Eble and Vasily Titov, NOAA/Pacific Marine Environmental Laboratory, USA

The sequence of devastating tsunamis across the globe over the past 10 years has significantly heightened awareness and

preparation activities associated with these high-impact events. Since the catastrophic 2004 Sumatra tsunami, NOAA has

invested significant efforts in modernizing the U.S. tsunami warning system. Recent developments in tsunami modeling

capability, inundation forecasting, sensing networks, dissemination capability and local preparation and mitigation

activities have gone a long way toward enhancing tsunami resilience within the United States. It is a matter of when, not if,

the mainland US will be impacted by a major tsunami possibly to the level of the devastating 11 March 2011 Tohoku event.

Forecast capability is an essential part of the tsunami resilient coastal communities development for the US coastlines.

Short-Term Inundation Forecasting for Tsunamis (SIFT) system developed by the Pacific Marine Environmental Laboratory

was accepted in NOAA operations in 2013. SIFT is the first forecast system that predicts potential tsunami flooding in real

time, while tsunamis are propagating across the ocean. Direct tsunami observations from NOAA’s network of deep-ocean

tsunami detectors, DARTs, provide real-time data for SIFT to ensure accuracy of the forecast. Forecast products include

estimates of tsunami amplitudes, flow velocities, and arrival times and coastal inundation areas. Forecast inundation

models have been developed to provide high-resolution real-time tsunami predictions for selected coastal locations while

the tsunami is propagating through the open ocean, before the waves have reached many coastlines. Forecast inundation

models are incorporated into the SIFT system for use at NOAA’s Pacific and National Tsunami Warning Centers

19

Tsunami wave impact on walls and beaches

Jannette B. Frandsen, Institut National de la Recherce Scientifique, Quebec, Canada

This talk presents a mix of small and large scale experiments as well as numerical model predictions.

Developments of tsunami warning systems have progressed rapidly since the Indian Ocean tsunami in 2004. In the

receiving end, at the location of harbors, beaches and coastal cities, the process on solving and improving the coastal

protection structures is an involved and overwhelming task.

Run-up/run-down processes consist of violent free-surface flows with a mix of sediment and/or debris. The impact itself

on structures may contain high peak pressures occurring over relatively short time. The long waves of tsunamis are

dominated by the horizontal velocity flow component when approaching the shore. Therefore, it is justifiable to some

degree to apply conventional depth integrated models. However, the role of wave breaking and viscous effects during

run-up/run-down are unclear and cannot be answered accurately with those inviscid model approaches. Development

of numerical models to accurately capture nonlinearities at the free surface and the wave-sediment interaction may

be important features to capture in advancing research in tsunami prediction. This talk introduces and focuses on an

alternative method to traditional numerical modeling, rooted in the Lattice Boltzmann Equations (LBE), to examine the

underlying physics of breaking wave and boundary layer physics. The model approach is intended to be used locally with

coupling to regional models. Simplified LBE models can, however, be used for large domain modeling but the advantages

of the approach diminishes with coarsening resolution. Herein, is introduced a simplified LB model which discretizes the

nonlinear shallow water equations in rotational flows. The free surface model is fully non-linear. Mathematically, this is

taking into account through the collision integral. The collision between particles assumes single time relaxation. Some

validation studies of nonlinear LBE wave loads in tanks will be shown. At shallow depth, bore physics are generated

resembling physics at the tongue of tsunami run-up.

Finally, the new large scale wave canal in Quebec will be introduced. The flume has a depth and a width of 5 m and is

120 m long. The flume is designed for modeling the interactions of waves, tides, currents and sediment transport. The

wavemaker is a piston type with a maximum stroke length of 4 m. Various initial conditions can be set-up including

classical regular/irregular waves and a host of user-defined functions, e.g. landslide and earthquake generated tsunami.

Large amplitude waves can be generated reaching the top of the flume walls with water depths ranging from 2.5 - 3.5 m

with wave periods of 3 - 10 s. Recent example studies will be shown.

20

GPU-accelerated hydrodynamics for wave impact problems

Christian F. Janssen, Institute for Fluid Dynamics and Ship Theory, Hamburg University of Technology

In this talk, we briefly introduce the Lattice-Boltzmann method (LBM), a very efficient numerical method to solve for

strongly nonlinear and turbulent free surface flows. The LBM has recently matured as a viable alternative to classical CFD

approaches, i.e. Finite-Volume or particle-based methods. The method solves a discretized Boltzmann equation that

describes the evolution of particle distribution functions on Cartesian grids. Whilst modeling essentially similar physics

as Navier-Stokes procedures, LBM features a number of performance-related advantages, particularly concerning data

locality and parallel computing. The presented LBM solver elbe [1,2] uses graphics processing units (GPUs) to accelerate

the hydrodynamic computations and allows for simulations near real-time.

After a quick presentation of the basics of

the LBM, the talk will focus on the concepts

for an application of elbe to the simulation

of violent free surface flows and wave-struc-

ture interactions. Floating body motions are

commonly performed by using deforming

meshes or rigid moving grids or a

combination of the two. In case of the LBM,

which operates on fixed, equidistant and

Cartesian grids, the grid update (i.e.

remeshing) reduces to the calculation of subgrid distances of the Eulerian lattice nodes to the surface of the structure [3].

Similarly, complex bathymetry and topology information can be considered in the simulations. The floating body motions

themselves are described by a quaternion motion modeler [4] or a PE physics engine [5], that are coupled to the LBM in a

bidirectional, explicit manner.

Several state-of-the-art validation cases will be presented at the workshop to show that the proposed numerical

methodology is able to reproduce accurate results for wave impact problems and wavestructure interactions, in a very

competitive computational time. Hence, the elbe solver can be a viable tool for fast and efficient simulations of complex,

potentially violent flows, such as Tsunami waves and debris flow.

References

[1] C. Janssen et al., The efficient lattice boltzmann environment elbe. http://www.tuhh.de/elbe.

[2] C. Janssen and M. Krafcyzk, Free surface flow simulations on GPUs using the LBM. Computers & Mathematics with Applications, 61 (12), pp. 3549-3563, 2011.

[3] N. Koliha, C. Janssen and T. Rung, A fast and rigorously parallel surface voxelization technique for GPGPU-accelerated CFD simulations. Submitted to Communications in Computational Physics.

[4] N. Koliha and T. Rung, Development of a quaternion-based 6-DOF model for the simulation of floating bodies. Bachelor thesis, Schriftenreihe Schiffbau, Hamburg University of Technology, 2011.

[5] K. Iglberger and U. Rüde, Large-scale Rigid Body Simulations. Multibody System Dynamics, 25 (1), p. 81-95,

Figure 1. A Tsunami wave hits New York (computer simulation with elbe, 2E7 grid nodes, FDS, TUHH)

21

Modeling hazards from seismic and SMF sources

James Kirby, Center for Applied Coastal Research, Department of Civil and Environmental Engineering, University of

Delaware

This talk addresses three aspects of ongoing work with colleagues at the University of Delaware and collaborating institu-

tions. The first part of the talk discusses ongoing hazard analysis and inundation mapping being carried out across the US

by the NTHMP program, with special attention to our own work on the US East Coast. Work in this area involves analysis of

distant sources, for which detection by seismic inversion and conventional DART buoy technology is the assumed warning

approach. However, a significant proportion of our work is aimed at analyzing large scale SMF events, which are known

to have a weak link to seismic signal characteristics and hence present unique problems for detection and warning. The

talk will describe several east Coast SMF scenarios in order to establish notions about arrival times and impacts of wide

continental shelf geometries, which are likely to be distinct from western Canadian settings.

The second part of the talk will focus on the modeling technology employed to describe tsunami generation, propagation

and inundation. Application of the Boussinesq model FUNWAVE to propagation and inundation modeling will be described.

The use of a recently-developed, 3D nonhydrostatic model NHWAVE to describe initial tsunami generation during both

seismic and SMF events will be described, and recent work on extending the model to describe deformable slides using a

number of approaches will be discussed.

Finally, to highlight the potentially important need for detection of SMF events occurring in conjunction with seismic

events, we present a hypothesis for an SMF-related mechanism for the generation of high local runup values along the

Sanriku coastline during the 2011 Tohoku-oki tsunami. Our scenario involves the utilization of conventional seismic

and GPS inversions to describe the main event and it’s long distance tsunami signature, together with an SMF event

geo-located and scaled by travel time arrivals and amplitudes at GPS buoys that produces the observed coastal runup and

eliminates the need for invoking a poorly understood slow-slip mechanism.

22

Assessing the tsunami hazard of the British Columbia coastline

Lucinda Leonard1, Garry Rogers2

1School of Earth and Ocean Sciences, Univ. Victoria, 2Natural Resources Canada

I will present the results of our recent tsunami hazard assessment for the British Columbia coastline. This assessment is

useful in terms of illustrating the relative importance of the various potential tsunami sources (including local and distant

earthquakes and landslides) and it provides estimates of the recurrence of damaging tsunami waves for different parts

of the coastline. The validity of our mainly empirical approach has been confirmed by run-up measurements of the 2012

Haida Gwaii tsunami.

I will also discuss the limitations of this preliminary assessment, and the studies that are needed in order to provide

communities with appropriate information for planning and risk-assessment purposes. It is important to determine not

just how often damaging tsunami waves may occur, but how large these waves may be at specific locations. The amplifi-

cation of waves from the 1964 Alaskan tsunami in Alberni Inlet (and other long narrow inlets) is a good example of how

certain locations are more at risk than others. Of critical importance is the determination of the maximum wave height

that could be expected.

The main factors affecting local tsunami wave height include variations in (1) the source (e.g., along-strike differences

in seafloor rupture), and (2) the tsunami’s travel path to specific locations (affected by the shape of the seafloor and the

coastline). Modelling studies are needed that address both types of variations, particularly for the source most hazardous

to BC: Cascadia subduction zone earthquakes. Modelling will require use of detailed bathymetry and onland topography

data to sufficiently estimate the extent of inundation from a range of tsunami scenarios.

Tsunami Modeling and Inundation Mapping in Alaska: Development of maximum credible tsunami scenarios

Dmitry Nicolsky1, Elena Suleimani1, Richard Koehler2, Jeffrey T. Freymueller1

1Geophysical Institute, University of Alaska Fairbanks, USA

2Alaska Division of Geological & Geophysical Surveys, Fairbanks, USA

The Alaska Earthquake Center (AEC) conducts tsunami inundation mapping for coastal communities distributed along

several segments of the Aleutian Megathrust and the Queen Charlotte-Fairweather fault system. Each fault segment is

characterized by a unique seismic history and tsunami generation potential. A critical component in the tsunami modeling

23

is accurate identification and characterization of potential tsunami sources.

We present a technique to develop maximum credible tsunami scenarios for locations that have short or nonexistent

paleoseismic/paleotsunami records, and lack modern seismic and GPS data. We employ the USGS plate interface model

and its discretization into a set of rectangle subfaults, empirical magnitude-slip relationships, and a numerical code that

distributes slip among various subfault elements. We also show how to take into account paleoseismic information on

uplift and subsidence as well as slip deficit models in areas where extensive GPS campaign surveys were conducted.

We perform simulations for each source scenario using AEC’s numerical model of tsunami propagation and runup, which

is validated through a set of analytical benchmarks and tested against laboratory and field data. Results of numerical

modeling combined with historical observations are compiled on inundation maps and used for site-specific tsunami

hazard assessment by local emergency planners.

In the wake of the 2011 Tohoku Tsunami: Modeling of tsunami wave propagation and evacuation. Research efforts at Tohoku University and the University of Hawaii.

Erick Mas1, Yoshiki Yamazaki2, Yefei Bai2, Kwok Fai Cheung2, Volker Roeber1

1International Research Institute of Disaster Science (IRIDeS), Sendai, JAPAN

2Department of Ocean & Resources Engineering, University of Hawaii and Manoa, Honolulu, USA

The 2011 Tohoku tsunami devastated the northeastern Japan coasts and caused localized damage to coastal infrastruc-

ture across the Pacific. We investigated the Tohoku tsunami with respect to its source mechanism and wave propagation

in the Pacific. A closer look at specific locations such as Hawaii showed that the tsunami resulted in strong currents that

led to closure of harbors and marinas for up to 38 h after its arrival. We utilize a non-hydrostatic model to reconstruct the

tsunami event from the seismic source for elucidation of the physical processes and inference of the coastal hazards. A

number of tide gauges, bottom pressure sensors, and ADCPs provided point measurements for validation and assessment

of the model results in Hawaii. Spectral analysis of the computed surface elevation and current reveals complex flow

patterns due to multi-scale resonance. Standing waves with 33–75 min period develop along the island chains, while oscil-

lations of 27 min or shorter are primarily confined to an island or an island group with interconnected shelves. Standing

edge waves with periods of 16 min or shorter, which are able to form nodes on the reefs and inside harbors, are the

main driving force of the observed coastal currents. Resonance and constructive interference of the oscillation modes

provide an explanation of the impacts observed in Hawaii with implications for emergency management in Pacific island

communities.

24

Numerical modeling gives researchers better understanding of the physics of tsunami events. In the recent years, a new

field of tsunami science has evolved to directly address hazard mitigation through modeling of complex social behavior. In

practice, organizing recurrent evacuation drills in a community demands budget, resources and community participation.

Instead, numerical and social models can be used as a tool to explore mitigation policies, residents’ behavior, disaster

impact, etc.

We will demonstrate the strengths and limitations of tsunami evacuation simulation by introducing an agent-based model

as an example of a disaster mitigation tool. Tsunami evacuation simulation contributes to the analysis of future implemen-

tations on tsunami mitigation and countermeasures (i.e. new infrastructure for sheltering, social education, emergency

support, etc.). Case studies such as bottleneck and congestion analysis in evacuation routes, demand and spatial distribu-

tion of shelters, human behavior in case of emergencies and casualty estimation due to tsunamis, will be presented.

Utilizing power of present and future open-ocean tsunami monitoring networks

Elena Tolkova, NorthWest Research Associates, USA

A tsunami forecasting technique is presented which allows to use tsunami observations in the region for providing fast

local forecasts. The suggested technique:

•  provides a prediction independent of the tsunami source estimate;

•  increases accuracy of predicting the first arriving as well as later waves using off-shore tsunami measurements

close to the target area;

•  allows for checking the accuracy of the conventional inversion-based forecast before the wave hits the coast.

This technique relies on a formalism of direction-specific response functions describing tsunami wave transformation at

a regional scale introduced in (Power and Tolkova, 2013). It was demonstrated that the measurements of an approaching

trans-oceanic tsunami taken in the open ocean as far as 1000 km off-shore can be directly converted into the wave time

history at the coast with a straightforward computation which can be performed almost instantaneously. This technique

has been successfully dem-

onstrated for the coast of

New Zealand and California.

For instance, numerical

simulations showed that a

single station northeast of the

Chatham Islands might be Left: record of the 2012/10/28 Haida Gwaii tsunami at DART 46411; right: the tsunami record

at Monterey tide gage (red) and its forecast (blue). The forecast is been made as the wave is been

registered at the DART one hour before arriving at the gage.

25

sufficient to accurately predict the tsunami wave at the east shore of North Island, New Zealand about 2 h in advance, for

a tsunami originating anywhere on Central/ South America subduction zone.

Today, existing instrumentation permits the use of the suggested technique in Alaska and the West Coast of the Unites

States, where an approaching far-field tsunami is first registered at DART stations situated along the coast at an average

distance of 370 km from land. It can also be used with tsunami detectors in cabled underwater networks where available.

Future instrumentation on submarine communication cables will supply larger selection of open-ocean measurements

and many more opportunities for this method.

References

W. Power and E. Tolkova. Forecasting tsunamis in Poverty Bay, New Zealand, with deep-ocean gauges. Ocean Dynamics, Vol. 63, Issue 11 (2013),

pp. 1213-1232; doi: 10.1007/ s10236-013-0665-6

Indian Tsunami Early Warning System

Dr. T. Srinivasa Kumar

Indian National Tsunami Early Warning Centre, Indian National Centre for Ocean Information Services, Earth System

Sciences Organisation, Ministry of Earth Sciences, Government of India

Until December 2004, not many people in India were aware of tsunami and its devastating capacity to washout coastal

areas. The great Sumatra earthquake (Mw 9.3) of 26th December, 2004 generated a tsunami which exposed the vulner-

ability of the Indian coastline and caused unprecedented loss of life and damage to property in the Indian Ocean rim

countries. In response to this, the Government of India established the Indian Tsunami Early Warning System (ITEWS), with

the warning centre operating from the Indian National Center for Ocean Information Services (INCOIS), Hyderabad, India.

The Indian Tsunami Early Warning System comprises a real-time network of seismic stations, Bottom Pressure Recorders

(BPR), tide gauges and 24 X 7 operational Warning Centre to detect tsunamigenic earthquakes, to monitor tsunamis and

to provide timely advisories to vulnerable communities. The Warning Centre is capable of issuing Tsunami bulletins in less

than 10 minutes after any major earthquake in the Indian Ocean thus providing a response/lead time of about 10 – 20

minutes for near source regions and a few hours in the case of far source regions. Timely tsunami advisories (Warning /

Alert / Watch) are disseminated to the vulnerable community using Decision Support System developed based on one of

its kind Standard Operating Procedure (SOP).

The criteria for generation of tsunami advisories (Warning / Alert / Watch) for a particular region of the coast are based

on the available warning time (i.e. time taken by the tsunami wave to reach the particular coast). The Indian Warning

26

Criteria are based on the premise that coastal areas falling within 60 minutes travel time from a tsunamigenic earthquake

source need to be warned based solely on earthquake information, since enough time will not be available for confirma-

tion of water levels from BPRs and Tide Gauges. Those coastal areas falling outside the 60 minutes travel time from a

tsunamigenic earthquake source could be put under a watch status and upgraded to a warning only upon confirmation

of water-level data. This implies that while the possibility of false alarms is comparatively more for areas close to the

earthquake source, for far source regions the rate of false alarms is less, since the warnings are issued only after con-

firmation of tsunami from water-level data. To reduce the rate of false alarms even in the near source regions bulletins

are generated by analysing pre-run model scenarios and issued only to those coastal locations that are at risk. Based on

the estimated water levels and travel times, the coastal areas are categorized into either being under Warning (Major

Tsunami), Alert (Medium Tsunami), Watch (Minor Tsunami) or No Threat.

Warning Centre disseminates tsunami advisories to various stakeholders through multiple dissemination modes simul-

taneously (Fax, Phone, Emails, GTS and SMS etc.). Earthquake information, tsunami bulletins as well as real-time sea level

observations are also made available on a dedicated website for officials, public and media. Users can also register on

the website for receiving earthquake alerts and tsunami bulletins through emails and SMS. The ITEWC serves not only as

a national tsunami warning centre for India, but also as a Tsunami Advisory Service Provider (TSP) for the Indian Ocean

Tsunami Warning and Mitigation System (ICG/IOTWS), which is responsible for providing Tsunami advisories to all IOTWS

member States. Future work is focussed towards water level inversion, real-time inundation modelling, use of near-field

GNSS measurements for real-time rupture characterisation and 3D mapping of vulnerable coastal areas.

Cascadia Megathrust Rupture Models for Tsunami Modelling

Kelin Wang, Pacific Geoscience Centre, Geological Survey of Canada

Details of megathrust earthquake models are very important for calculating the impact of tsunami waves on local coastal

areas. One issue of concern is fault geometry: a change of a few degrees in fault dip or using a plane to represent a

curved fault gives rise to a large change in predicted seafloor uplift. An equally critical issue is the distribution of slip along

the shallow part of the fault: A sudden termination of slip at a shallow depth, a gradual termination, and a rupture that

breaches the seafloor produce dramatically different seafloor displacement patterns. These details are poorly known for

any subduction earthquake, even for the best observed 2011 M9 Tohoku-Oki earthquake. For assessing local tsunami

hazards associated with a future Cascadia megathrust earthquake, it is necessary to consider a variety of geometrical

and slip scenarios that are consistent with fault mechanics, geological structure, evidence for previous earthquake and

tsunami events, and lessons learned from instrumentally recorded subduction earthquakes elsewhere in the world.

Cascadia rupture models developed in the 1990’s and early 2000’s typically feature a shallow zone of uniform slip (the

27

full-rupture zone) with a deeper zone over which the slip tapers to zero (the transition zone). These simple models are

adequate for calculating tsunami waves arriving on a remote coast, such as the waves in Japan caused by the great

Cascadia earthquake in AD 1700. They are also useful for predicting a general pattern of tsunami waves at local coasts

along the Cascadia margin. Later models, particularly those used by the Oregon Department of Geology and Mineral

Industries (DOGAMI) for tsunami hazard assessment along the Oregon coast, feature more realistic scenarios. The

DOGAMI models consider two types of fault geometry: a megathrust with or without a splay fault which diverts the

shallow rupture upward. The 3D megathrust geometry is based on geophysical imaging, and the location and geometry

of the splay fault are based on interpretations of geological and seismic imaging data. The slip distribution along the

megathrust in the downdip direction is assumed to have a bell shape, with the slip peaking in the central part but tapering

to zero updip and downdip. The assumed updip decrease in coseismic slip toward the deformation front (“trench”) is

based on the classical fault model in which the shallowest segment resists seismic rupture but slips aseismically after the

earthquake. The Tohoku-Oki earthquake which exhibited dramatic trench-breaching rupture challenges the universality

of the classical fault model, and the seismic potential of the shallowest part of the Cascadia megathrust needs to be

reassessed. Most Cascadia rupture models assume a rather uniform slip pattern along strike. Recent microfossil data

suggest strong variations of coseismic slip along the Cascadia margin during the AD 1700 great earthquake, similar to any

other instrumentally recorded subduction earthquakes. The importance of such heterogeneous slip to tsunami modeling

at Cascadia is yet to be understood.

The Activities of the US Tsunami Warning Centers During the Haida Gwaii Tsunami

Dr. Stuart A. Weinstein, NOAA/NWS Pacific Tsunami Warning Center, USA

The Oct. 28, 2012 Haida Gwaii Mw 7.8 earthquake was the largest earthquake to strike Canada since 1958, and the second

largest instrumentally recorded earthquake within Canadian Territory. This earthquake and subsequent tsunami had

characteristics that proved challenging for the US Tsunami Warning Centers (TWCs). For starters, the plate boundary upon

which the earthquake occurred was thought to produce earthquakes with a predominantly oblique strike-slip mechanism.

Hence no DART buoys were deployed in this region. There was also a dearth of near real-time sea-level data from coastal

areas in close proximity to the earthquake. In addition, a delay of around 10-15s in the onset of the main moment release

of the earthquake contributed to the TWC’s underestimating the size of the earthquake in the first three to four minutes.

The initial estimate of the magnitude, 7.1, exceeded the US National Tsunami Warning Center’s (NTWC) threshold for

issuing a warning; hence a warning was issued for an area surrounding the immediate vicinity of the earthquake. After

more analysis both US TWC’s obtained a larger magnitude of 7.7. At this point the NTWC placed the region to the south

of the warning area in an advisory. About two hours later when larger waves than expected were recorded by the DART

28

buoys and models showed the main beam of energy pointing towards Hawaii, the US Pacific Tsunami Warning Center

(PTWC) issued a warning for the state of Hawaii. This was the first time in recent memory where Hawaii was the only

region placed in a warning status for an earthquake occurring on the Pacific Rim by PTWC. The event came to an end when

PTWC issued a warning cancellation for the State of Hawaii some 7.5 hours after the earthquake.

Fortunately, no destructive tsunami was observed in Hawaii or along Canadian and the Northwest US coasts. The maximum

observed amplitude on sea-level stations in Hawaii was 0.8m at Kahului, Maui. However, one of the hardest hit areas, the

North Shore of Molokai did not have any functioning sea-level stations. Observations on the North Shore of Molokai do

suggest runups exceeded one meter in some places. These observations bear out a prediction of the forecast models

suggesting that Maui and Molokai would be the hardest hit coastlines in Hawaii. Indeed, these shores were perhaps the

hardest hit by the tsunami anywhere outside the immediate vicinity of the earthquake itself.

I will present a detailed timeline of the events and challenges faced by the US TWCs during the course of this event.

29

Summary of Discussions Instrumentation

Peter Anderson, Jan Buermans, Tom Dakin, Martin Heesemann, Joe Henton, Garry Rogers, Andreas Rosenberger, Michael

Schmidt, Tummala Srinivasa Kumar

Summary

Tsunami detection instrumentation have two uses. On one hand they provide ground truth data for models. This data can

be compiled into a library of various events, which can be used to evaluate models. On the other hand, the real-time data

of directly measured displacements can also provide the initial conditions and constraints for choosing a pre-run model

to suit the existing event and estimate the impending impacts. As more real-time data comes in from the entire network

sophisticated users can refine the impact assessment over time.

The library of data for models is especially valuable for propagation and shoaling teletsunamis models. This library is

useful for developing the models as well as assessing the accuracy of the models. One important data set for impact

assessment is a map of the seafloor displacement.

This discussion group defined the requirements for the instrumentation as well as the considerations required for correct

detection of events.

Technical requirements

The distribution of the array of sensors is a key part of the detection. In the case of an earthquake, at least three P-wave

detections are needed to estimate the epicenter. Japan is so far the only country with a near field system based on a high

number of strong motion sensors distributed widely, densely and with exceptional reliability for the communications

from the sensors. Their target response is two to three minutes.

In the case of the coast of British Columbia, a combination of land and shore based seismometers would be the ideal

combination for rapid estimation of source parameters. The use of real-time GPS co-located with strong motion sensors

on land would provide the optimal data set of motion and displacement.

Currently Ocean Networks Canada is working on a Web-enabled Awareness Research Network (WARN) system that will

speed up the warning process. In this system, when an event is detected by three sensors calculations are triggered to

detect the epicenter and magnitude of the event. WARN has the ability to look for correlations of both, movement and

30

direction of the co-seismic displacements. Multiple sensors need to be included in this system, such as strong motion

sensors, seismometers, accelerometers, geophones, bottom pressure recorders (BPRs) on both sides of the fault, GPS on

land for ground displacement direction and magnitude, tilt meters and dilatometers for measurements of strain. An ideal

distribution of the instruments would have BPRs and seismometers located in lines perpendicular to the coast with 10km

spacing of sensors in the lines.

Technical considerations

Due to the infrastructure and servicing costs of subsea sensors land based sensors provide detection at a lowest price.

However they also provide the minimal warning time. The major cost factor for submarine installations is the infrastruc-

ture, which makes the cost of the sensors look cheap in comparison. Therefore, the cost of the sensors should not be a

factor in choosing technologies for the coast of British Columbia, since the observatories are already in operation, this

significantly reduces the initial investment. Laying out cable is a major technical impediment to submarine operations. Key

measurements should take place along the shelf, in the continental slope for Cascadia.

One limitation related to the use of cabled observatories is the possibility of cable breakage due to the earth motion. The

reliability of a cabled system during major local events is questionable. The rupture expected for this area is hundreds

of kilometers long and on the order of 15 to 20 km wide. The rupture will propagate at a rate of 2 to 3 km/s if everything

moves in the same direction. Another issue for reliability is trawling fisheries in the area. The sensors deployed for early

warning need to have resistant frames to avoid damages in this context.

Event Detection and confirmation

In the event of an earthquake, seismometers are able to detect the primary and secondary waves. The sensors and

infrastructure used for a reliable system need to be distributed in a dense network able to detect the event even during

strong motion. The communication system needs also to be reliable during a catastrophic event. An example of a reliable

communications system is the BGEAN satellite.

The event confirmation relies on multiple instruments including GPS, strong motion seismometers and seafloor geodesy

instrumentation such as tiltmeters, dilatometers that can measure volumetric strain and bottom pressure recorders

(BPRs). The strong motion sensors should be collocated with BPRs or GPS. The installation of some of the sensors on land

may be better for the reliability of the system; however, marine installations are also required for early event detection.

The study of small tsunamis and far field tsunamis is key to calibrate the models.

The team identified different types of sensors based on their use in terrestrial versus marine environments as well as

their use for studying near and far field tsunamis. The major instrumentation discussed is represented on the table below.

31

Near Field Far FieldTerrestrial •  Collocated GNSS (e.g. GPS) and strong motion sensors (real time)

•  Tide gauges (real time)

•  Coastal radar

Tide gauges

Marine •  Strong motion accelerometers and geophones (real time)

•  Bottom Pressure Recorders (BPRs) array

•  Tiltmeters (real time)

•  Hydrophones for landslides

•  Seafloor geodesy (acoustic GPS)

•  Strain meters

BPRs

The displacement in terrestrial environments for near field tsunamis can be measured with GPS and strong motion sensors

such as seismometers or accelerometers. The GPS data needs accumulative measurements. This system can provide

good data in 60-90 seconds and accurate data in 3 minutes. The GPS and strong motion sensors need to be co-located.

The tidal gauges can provide important information for historical events. The coastal radar may provide confirmation of

the predictions. As in every tsunami detection system, it is important to have redundancy of instruments.

In the case of the terrestrial far field detection, the first warnings will come from tsunami centers in the Pacific, and the

sensors from Ocean Networks Canada will provide confirmation. The tide gauges for far field detection need to be located

at critical points along the coast and at the mouth and inland of the inlets to measure the magnification of the wave. The

systems should be distributed along the whole coast of British Columbia in a coarse spacing to help with the assessment.

The wave propagation needs to be measured with tide gauges distributed from North to South along the coast with

sufficient density to supply the model library with wavelength and wave height data. The location of the BPRs and tidal

gauges is critical and need to be located in areas of high impact such as Tofino, Port Alberni and Uclulet.

Potential sources for far field tsunamis for our coast are Hawaii, Alaska, Japan and South America. Two Alaskan areas

are of particular interest since the stresses in these areas have been building without relief. The bathymetry is a key

component of the models and in particular for inlets such as Tahsis, Port Alberni and Port Renfrew. The location of

BPRs and tide gauges at mouth and headwater are important to determine the waveheight multiplier effect of inlets. It

would be necessary to increase the collaboration between NOAA and Ocean Networks Canada, so that both systems can

exchange alerts and real time data from instruments.

In the case of marine far field tsunamis the location of the BPRs is related to their capability of detection from different

locations. In the case of propagation from North to South there is no coverage. Tsunamis propagating from South to

North are detected by the DART buoys. The NEPTUNE cabled observatory is able to detect tsunamis that are propagating

from West to East, and in a similar way, the near field data from the NEPTUNE observatory could be sent to Hawaii.

32

The most relevant seismic frequencies for the detection of these events is mostly below 10 Hz, with far field events from

0.1 Hz to 2 Hz and near field events varying between 0.1 Hz and 60 Hz .

For near field tsunamis in the marine environment tidal gauges, BPRs and mini-observatories with tidal gauges were

identified as the main independent instruments for detection. Several other instruments can be used as part of the cabled

network. These include strain meters (piezometers and fiber strain sensors), hydrophones for underwater landslides,

seismometers, accelerometers, tilt meters, acoustic GPS, geophones and BPRs. The BPRs need to be co-located with

strong motion and tiltmeters in lines perpendicular to shore.

A question was raised during the discussion about the capability of the network to receive data from 100+ km away

from the network nodes in an effort to expand the spatial coverage of sensors. One possibility is the use of long range

low bandwidth acoustic links. Another option is the use of satellites such as Iridium, BJAM and GOOS (used by NOAA) for

terrestrial sensors or where there is a surface expression for sub-sea sensors.

Another factor to be considered in the case of the coast of British Columbia is the detection of landslides. Submarine

landslides can be detected in real time with hydrophones. A team at Natural Resources Canada is studying slip failure and

multibeam data needs to be collected after a submarine landslide to ground truth the event.

Earthquake Source Definition

Jörn Behrens, Herb Dragert, Tania L. Insua, Dmitry Nicolsky, Roe Markham, Kelin Wang

Summary

One of the discussion groups during this workshop has focused on definition of the earthquake sources near British

Columbia. The group covered three major components of the earthquake source definition. Namely, the fault mechanics,

rupture scenarios and real time tsunami source estimation were analyzed. Uncertainties and the need of further research

were pointed out when the group considered that this was a limitation to forecast tsunamis in the area.

Fault mechanics

Paleoseismic records reveal that great tsunamigenic earthquakes repeatedly occur in the Cascadia subduction zone with

irregular intervals averaging about 500 years (Atwater, 1987), often accompanied by a tsunami. The latest trans-Pacific

tsunami generated by an earthquake at Cascadia occurred in January 1700 (Satake et al. 1996; Atwater et al. 2005). Multiple

models of the Cascadia zone rupture are suggested by Satake et al. (2003) and Priest et al. (2009) and in references

33

therein. These models describe hypothetical coseismic displacement fields of the Cascadia rupture, with various levels

of detail. Tsunami heights in Japanese historical records can constrain the slip distance of the 1700 Cascadia earthquake

(Satake et al, 1996), but do not well constrain the downdip limit of the rupture (Wang et al, 2003).

Although the Cascadia subduction zone ruptured about 300 years ago, it still has the great potential to generate tsunamis

along shore of British Columbia. The discussion group indicated the lack of sampling in the area to constrain the physics

behind the megathrust mechanics. The Nootka Fault and the Explorer plate in the Cascadia subduction zone are still

unknown areas. The rupture behavior of the shallowest part of the megathrust is one of the research topics that should

be addressed as soon as possible in order to have a better understanding of the source. The downdip limit of the rupture

zone is not well defined for Cascadia, and estimations vary from deep to shallow rupture on a broad range. The uncertainty

on the slip distribution limits the model complexity that can be used for the source, and therefore it affects the outputs

from the tsunami modeling. Complex models require more detailed inputs that come from research that needs to be

developed for the British Columbia coast.

The group also emphasized slow ruptures that typically happen far offshore, and their prevalence in the B.C. area is still

unknown. They are, in general, rare events such as a 2006 Mw 7.7 event that generated the devastating tsunami in Java.

Thus, earthquake generated tsunamis in the British Columbia coast require near-shore tsunami monitoring and the wave

generated may be larger than expected based on the magnitude of the earthquake. More in-depth knowledge in the

sediment and fault zone reology and fluid pressure is needed for Cascadia. The fault zone structure, dynamic weakening

and the time for the healing of faults in this area are not well defined.

The sensitivity of the different models should be tested in advance to identify and prioritize major areas of research

that need to be addressed. The instrument sensitivity to other signals should also be modeled, for example, cryogenic

gravimeters were able to detect a Haida Gwaii hydrogeological effect.

Along the process of determining the earthquake source definition there are multiple uncertainties that add to each

other such as the variability of the base data, the modeling uncertainty, etc. The sensitivity of the rupture mechanisms is

complex and the outputs for tsunami modeling are dependent on the particular case. Fully dynamic rupture models with

a solid fluid interaction can define the system and help to address the sensitivities. These models can also be useful to

validate simple models used for real time predictions (see ASCETE project: http://www.ascete.de).

Rupture scenarios

Tsunamigenic rupture scenarios are geological plausible rupture scenarios that specify the rupture dimensions (up-dip

and down-dip depths), maximum and average slip as well as the slip distribution.

There are some observational constraints to this such as the regional framework and seismotectonics of the area, which

may not be well defined as in the case of the Explorer plate area. Paleoseismic evidence can help to reduce uncertainties

34

in this framework as well as interseismic long term GPS records that would indicate potential patterns of the co-seismic

deformations. Turbidite records can indicate the recurrence of these events, although they do not provide much

information about the rupture. Geophysical imaging can provide geometry of the megathrust structure and address the

possibility of splay faulting affecting ruptures in this area.

Different models can be applied to generate a source from rupture scenarios. Static models can be done using an elastic

half-space but they cannot justify more complex geometries which require more extensive input data. Finite element

models can be used to produce a more sophisticated output based on multilayer applications. Another catagory of models

that can be used is kinematic time dependent rupture models.

Real time source definition

When fault geometry is assumed for the generation of the source signal, the inverted model can be done in real time

to estimate rupture size and slip magnitude. Based on these parameters, deformation can be determined. However

numerous assumptions are made that can affect the uncertainty of the source. When a wave is measured more directly,

as for example by GPS buoys or pressure sensors, some of these uncertainties are directly addressed.

The discussion group mentioned that there are multiple measurements that can be used to define the source. It is

possible to use seismic networks, real-time GPS, seafloor pressure sensors, borehole strainmeters and pore pressure

sensors. A number of these sensors are currently operational along the coast of British Columbia, including multiple

real-time GPS stations from Natural Resources Canada, several borehole strainmeters from UNAVCO , together with the

Ocean Networks Canada sensors. A lack of coordination for the combined use of all these sensors in a tsunami detection

system is apparent. Arrival times from different instruments can help to invert back to the source, however, the required

calculations can take over five minutes. This can be a limitation for the system that ONC is trying to test.

There are different real-time inversion techniques for this problem. One technique is based on unit-source based linear

inversion relying on off-shore wave gauges proposed by Wei et al. (2008). This approach is very successful in far field

tsunami forecasting and can beneficially be combined with real-time local inundation modeling. The other approach by

Behrens et al. (2010) uses a rigorous uncertainty model to reduce forecast error in very short time based on diverse (but

initially few) data records. This approach is bound to pre-computed scenarios, but is well-suited for near-field tsunami

early warning.

Projects under development currently that could be included in an ideal system to detect and forecast tsunamis include

the seafloor tilt-meters being developed by Woods Hole Oceanographic Institution. These sensors could be connected to

the NEPTUNE observatory. Fiber optic cable strainmeters developed by a team from University of Miami and University of

Ottawa with industry partners are currently in testing phase and could provide more insights on source definition.

35

HF radar based wave-detection stations can provide reliable information about sea level anomalies and could readily

be included into the multi-sensor forecasting approach. HF radar can be utilized in a multitude of applications and can

easily be used for ocean current observations, tidal current monitoring and even ship traffic monitoring outside of the

(rare) tsunami hazard event. Algorithmic developments for accurate wave pattern detection and derivation of amplitude

information would be necessary to maximize the gain from this instrument class.

Combined GPS/acoustic geodetic measurements at sea (Burgmann and Chadwell, 2014) are also under development.

Surface GPS systems can monitor the position and dynamics of gliders on the ocean surface while simultaneous acoustic

ranging is carried out to seafloor transponders. These systems could measure motion and kinematics of the seafloor at

the time of the event. GPS located on buoys provide information strictly about the tsunami, not the rupture, although data

processing can help to infer the rupture. GPS located on boats can provide wave information in a similar fashion to a buoy.

Other signals that could be considered for future phases of the tsunami detection system are satellite altimetry and the

atmospheric signature from earthquakes.

A major concern that is still not well addressed for early warning tsunami systems is the presence and effect of submarine

landslides.

References

Atwater, B.F., 1987, Evidence for great Holocene earthquakes along the outer coast of Washington state: Science, v. 236, p. 942–944.

Atwater, B.F., Musumi-Rokkaku, Satoku, Satake, Kenji, Tsuji, Yoshinobu, Ueda, Kazue, and Yamaguchi, D.K., 2005, The orphan tsunami of 1700—Japanese clues to a parent earthquake in North America: U.S. Geological Survey Professional Paper 1707 (prepared in cooperation with the Geological Survey of Japan, the University of Tokyo, and the University of Washington, and published in association with University of Washington Press), 133 p

Behrens, J., Androsov, A., Babeyko, A. Y., Harig, S., Klaschka, F., and Mentrup, L. (2010): A new multi-sensor approach to simulation assisted tsunami early warning, Nat. Hazards Earth Syst. Sci., 10:1085-1100, doi:10.5194/nhess-10-1085-2010.

Burgmann, Roland, and Chadwell, David, 2014, Seafloor Geodesy, Annu. Rev. Earth Planet. Sci. 42:509–34, doi:10.1146/annurev-earth-060313-054953.

Satake, Kenji, Shimazaki, Kunihiko, Tsuji, Yoshinobu, and Ueda, Kazue, 1996, Time and size of a giant earthquake in Cascadia inferred from Japanese tsunami records of January 1700: Nature, v. 379, no. 6562, p. 246–249.

Satake, Kenji, Wang, Kelin, and Atwater, B.F., 2003, Fault slip and seismic moment of the 1700 Cascadia earthquake inferred from Japanese tsunami descriptions: Journal of Geophysical Research, v. 108, no. B11, p. 2,535–2,551, doi:10.1029/2003JB002521.

Priest, G.R., Goldfinger, Chris, Wang, Kelin, Witter, R.C., Zhang, Yinglong, and Baptista, A.M., 2009, Confidence levels for tsunami-inundation limits in northern Oregon inferred from a 10,000-year history of great earthquakes at the Cascadia subduction zone: Natural Hazards, v. 54, no. 1, DOI 10.1007/s11069-009-9453-5.

Wang, Kelin, Wells, R.E., Mazzotti, Stephane, Hyndman, R.D., and Sagiya, Takeshi, 2003, A revised dislocation model of interseismic deformation of the Cascadia subduction zone: Journal of Geophysical Research, v. 108, no. B1, p. 2,026–2,038, doi:10.1029/2001JB001227, http://www.agu.org/pubs/crossref/2003/2001JB001227.shtml.

Wei, Y., Bernard, E. N., Tang, L., Weiss, R., Titov, V. V., Moore, C., Spillane, M., Hopkins, M., and Kanoglu, U. (2008): Real-time experimental forecast of the Peruvian tsunami of August 2007 for U.S. coastlines, Geo. Res. Let., 35:1-7, doi:10.1029/2007GL032250.

36

Tsunami modeling

Ali Abdolali, Marie Eble, Josef Cherniawsky, Isaac Fine, Jannette Frandsen, Jeffrey Harris, Maia Hoeberechts, Jim Kirby,

Hannan Lohrasbi, Scott McLean, Steve Mihaly, Alexander Rabinovich, Volker Roeber, Fred Stephenson, Elena Tolkova,

Christina Waddle, Onat Yazir

Summary

The tsunami modelling breakout group had an open discussion focused in answering multiple questions about the imple-

mentation of the models, including limitations and requirements. The views of the different experts included in this group

are captured on this summary.

The group discussed the need for improvement on detection of tsunami, wave propagation, and inundation maps as well

as on modelling and computational requirements.

General conclusions indicated the need for accurate topography and bathymetry as a key step in the process of getting

better models and as a particularly limiting factor for the coast of British Columbia. Currently, the propagation models

are not a limitation for tsunami forecasting in deep waters, where linear approximations can be used, but the source

modelling and the propagation during inundation are more complex. Inundation models are particularly important in

near-field tsunamis like the one expected along our coast. The effect of shallow water and small basins on the inundation

is an important research area that affects the results of inundation models.

Ocean Networks Canada as a leader of this initiative needs a repository of models, algorithms, data input types and

output types. These pieces are the base to test and build an environment where we can use the model on our coast. The

group proposed to compare actual data with model outputs from different research teams over the world that could help

to validate different models against each other. The real time observations such as the ones obtained by the NEPTUNE

observatory, should fit into the models to achieve a better knowledge of the propagation of the wave. The benchmark

testing could be referred to a specific case (e.g. Haida Gwaii) or be a generic study that could be tested in different areas.

The team also discussed the best way to accumulate the data not only to verify the accuracy of models with past events

but also to best forecast future events. Existing NEPTUNE sensors provide good information on wave paths for far field

tsunamis; however, the network needs new approaches and an increase in sensors for near-field tsunami detection. The

determination of the source is one of the major limitations in near field events.

37

What requirements should each model fulfill in order to implement it for the BC coast?

A discussion was centered in the need to improve benchmark standards, since these standards are not sufficient

with testing being too lax, especially if there is a high level of complexity. Two major research fields were identified as

challenging for tsunami forecasts: velocity and wave propagation in complex coastlines where reflection takes a major

role. Benchmark testing against real data is a difficult task and normally lab experiments are set up for benchmarking

models where simplified cases are studied.

When a model is transferred from one area to another, the equations are the same but the application to a complex coast

like British Columbia needs to be tested. However, the generic nature of most of the models should make them useful for

this coast. Bathymetry is a key component for the success of this adaptation. Land based data is also important but easier

to acquire and therefore it is not as limiting as bathymetry data.

Modellers need to agree on criteria to evaluate the performance of models for a benchmark test. Models can be plugged

in a test bed and compared against actual data. The initial objective for a warning from this workshop was the ability

of generating a warning in 2 minutes. The experts in the workshop identified this objective as very ambitious, since the

existing systems take longer time to evaluate the event.

What inputs and quality of data are needed for these models?

The team of experts attending the workshop indicated that a 7 m (~1/3 arc second) grid resolution for the bathymetry

off the coast of British Columbia would cover the needs but it would be a grid of a considerable size for the analysis. A

modification to this approach would be to use coastal relief models such as the ones used by NGDC (http://www.ngdc.

noaa.gov/mgg/coastal/crm.html) at a 3 seconds resolution. All the data available for bathymetry needs to be gridded up

and referred to a consistent datum. In the case of topography, mostly based on LIDAR, less than 1 m resolution is needed

to have accurate tsunami forecasts. The data necessary for the project should at least cover 50 km off the coast for

detailed bathymetry The entire width of the continental shelf should be available with at least a 10 seconds grid. The team

indicated that a good review in general terms of the necessary inputs is available in Yeh (1998).

The experts indicated the need of benchmark testing for structures and also the difficulty of knowing the cause of inac-

curacies in the inundation models. Wave records from previous events are necessary to evaluate models.

Changes in the bathymetry can alter the accuracy of the models and therefore data needs to be updated and reviewed

regularly. For example, a sand bar area that is dynamically changing should be regularly revisited. The workshop attendees

recommend doing a sensitivity test for the bathymetry versus the topography for the inundation maps. The effect of a

smaller grid in bathymetry may be less important than the effects of the topography after certain requirements for grid

size are fulfilled for the bathymetry. The team also indicated that a sensor sampling-rate of one second was good for the

models that they are using.

38

What software, hardware, or systems would be needed to implement these models?

Most of the models used currently in tsunami science require relatively conventional hardware, except those using GPU

clusters. Programming languages are variable among the different teams that met during this workshop. They offered to

provide more detailed information on each model after this meeting. It was suggested that ONC develop a questionnaire

to send to modellers for this information.

How could these models be integrated with the existing and future real time sensors from Ocean Networks Canada?

Tsunami events have been detected by the Ocean Networks Canada infrastructure in the past. In particular, the current

meter located at Barkley Canyon and multiple Bottom Pressure Recorders (BPRs) detected the Tohoku tsunami in 2011.

The team of experts suggested the addition of a current meter to the mini-observatory in the Alberni Inlet. The Alberni

Inlet area is a generally calm area that has shown the importance of coastal features. The modelling discussion group

requested additional sensors to be added to this area in order to get better Acoustic Doppler Current Profiler (ADCP) and

hydrophone measurements. These new sensors would be used not only for validation of models but also for real time

detection.

Other instrumentation such as gravimeters were also considered as a useful tool. Ocean Networks Canada has one

available that could be deployed at Cascadia, adding to the network to detect tsunamis.

The current sampling for the ADCPs and BPRs is one second, and their capability of detection is 0.1mm at 1Hz. These

sensors have been able to detect tsunami waves from far field tsunamis in the past.

Global Positioning Systems (GPS) would be a good addition to the Ocean Networks Canada system. GPS systems have

demonstrated strong capabilities for the detection of horizontal motion in the case of the Tohoku earthquake and tsunami.

There is currently an existing GPS network on Vancouver Island, but this system is not operative for tsunami detection.

The workshop attendees expressed concern on the use of the GPS technology for near field tsunami, but they indicated

the necessity of considering this option.

Temperature measurements are necessary to correct the pressure variations in the ocean. This is available on the existing

sensors of the Ocean Networks Canada.

39

What areas of research require new modelling approaches for the BC case? What are the technical challenges to developing these new approaches?

The group identified major research areas as landslides (aerial, subaerial, lake and submarine) and meteorological

tsunamis.

Meteorological tsunamis due to atmospheric conditions are potentially damaging and seem to be relatively common

on the coast of British Columbia. In the past ten years, 6 six meteorological tsunamis have been detected. They are

difficult to model since the measurements of the source are complicated, however, they can be detected with HF radar.

Measurements of the wind field can help when forecasting meteorological tsunamis. There are multiple technologies

to take these measurements, including AXYS technologies measurements of wind speeds 100m above surface or buoy

measurements ~3m above sea surface. These measurements need to be standardized for a distance over the surface and

benchmarking strategies need to be planned for this type of tsunami.

Submarine landslides are also recurrent events off the coast of British Columbia (e.g. Kitimat submarine landslide).

Mapping for unstable sediment bodies and geotechnical studies of the sediment can help to forecast these events.

Obtaining the geotechnical information for underwater areas is challenging.

Citizen science is an interesting option for the forecasting of these events. The use of barometers by citizens and schools

could aid on the need of measurements for this project. To date, weather measurements available from schools are not

common enough to have a reliable network. There are also some limitations on where the instrumentation could be

located, for example, barometer measurements in boats are probably affected by the aerodynamics of the ship rather

than the surrounding weather. The option of using high-resolution satellite live feeds should be also investigated for this

purpose. Wind variations would need to be measured in multiple locations and in particular in shallow water areas since

they are more related to meteorological tsunamis.

Hecate Straight, off the coast of British Columbia, is an area susceptible to meteorological tsunamis. This area could

be monitored with CODAR in Sandspit and Vanilla Island and existing buoys from Environment Canada, and additional

instrumentation could also be used in the area.

What are the challenges to integrating these models into an early warning tsunami system (i.e. real time optimization of the model based on changing input from instruments)?

The major challenges for the integration of models from different people in the discussion group are the time necessary

to integrate all the data available and making this data accessible for inundation modelling. The data requirements,

discussed above indicate the necessity of accurate bathymetry and topography.

40

What type of ‘modelling studio’ could ONC provide for you to develop your models in an ideal world?

One of the major objectives of Ocean Networks Canada is to provide a modelling studio that could be used as a test-bed

for different tsunami models. We asked the experts attending our workshop to give a description of what would be the

assets of an ideal system for them.

There is a clear need of a graphical user interface that allows for the plugging in of models as well as allowing the user into

access points. The modelling desktop should be standardized and the user should be able to run the models. Since one

of the major objectives of the studio is to be able to compare models there needs to be a base grid to be able to compare

models.

Multiple questions arose on the best way to architect the studio. One point to be determined on the definition of the

studio is to decide if all users can log in and see the rest of the models or if a graphical interface should be used for the

comparison among models. Another parameter to be defined is if the user can replace models as an end user or only

view the ones uploaded into the system. These two parameters will need to be defined as the project advances. The user

should be able to define the number of CPUs to be used in the studio and the output data should be transferred back to

the users.

To guarantee the success of this studio, Ocean Networks Canada should start with a working group that would aid on the

first trials. ONC should also evaluate the resources that it will take to maintain the model studio.

41

Attendees BiographiesAli Abdolali - University of Rome, Italy

Ali obtained his bachelor degree in Civil Engineering and Master of Science in Hydraulic

Engineering from Amirkabir University of Technology (Tehran Polytechnic).

As part of his MSc project, he investigated the flow pattern near floating breakwaters. Later, he

joined University of Roma Tre, Italy to do his Ph.D in coastal and maritime engineering. His field

of research mainly focused on numerical modeling of weakly compressible fluid in the field of

large scale tsunami modeling. Besides works on tsunami early warning systems (TEWS), he has

been involved in a European project entitled MERMAID (Innovative Multi-purpose offshore

platforms: planning, design and operation) working on numerical modeling of WECs i.e. Venice

Gates and Oysters.

Peter S. Anderson - Simon Fraser University, CanadaPeter Anderson is Director of the Telematics Research Lab and Associate Professor of

Communication at Simon Fraser University. He has an international background in research and

teaching in the fields of telecommunications, media, information systems and communication

policy. During the past twenty-five years he has participated in the design and implementation of

electronic communication and information systems for disaster management in collaboration

with the United Nations, NATO, scientific, government and non-government disaster management

organizations and representatives and is frequently called upon to assist during emergency and

disaster events. He is currently collaborating with Canadian federal, provincial and territorial agencies, local authorities and

responders on new methods for improving intra and interagency communications for mission critical operations, public

warning and situational awareness. In 2013, he was awarded the Order of British Columbia and the Pacific Northwest

Preparedness Society Lifetime Achievement Award in recognition of his contributions.

Jörn Behrens - University of Hamburg, GermanyEducated as an applied mathematician with PhD (Dr. rer. nat.) from Bremen University and

Habilitation from Technische Universität München, Munich, Jörn develops numerical methods

for atmospheric and oceanic simulation. In 2006 following the 2004 Sumatra-Andaman Tsunami,

he became head of the tsunami modeling group at Alfred Wegener Institute, Bremerhaven, and

developed the simulation component of the German-Indonesian Tsunami Early Warning System

(GITEWS). After delivering the system, he accepted a professorship for numerical methods in

geosciences at the Center for Earth System Research and Sustainability (CEN) of University of

Hamburg in 2009. He coordinates the research project ASCETE (Advanced Simulation of Coupled

Earthquake Tsunami Events), and serves as Co-Chair of UNESCO Intergovernmental Coordination Group for setting up a

Tsunami Early Warning System for the Mediterranean, North-East Atlantic and Connected Seas (NEAMTWS) Working Group 1

(Hazard Assessment and Modeling).

42

Alison Bird - Natural Resources CanadaAlison Bird began her work as earthquake seismologist in 1997 with the International Seismological

Centre in England, returning to Vancouver Island in 2000 to work with Natural Resources Canada.

Alison analyses western Canadian earthquakes and routinely responds in the wake significant

events (giving over 500 interviews over the past decade). She lectures for part of a quarterly

course in Light Urban Search and Rescue, and is involved in several outreach activities as part of

her participation in the Public Safety Geoscience programme. Alison is passionate about seismic-

resistant engineering, mitigation techniques for at-risk communities and encouraging awareness

of earthquakes and their associated hazards.

Jan Buermans - ASL Environmental Sciences Inc., CanadaJan Buermans, PEng. has a degree in Mechanical Engineering from Queens’ University. He has

been working with ASL Environmental Sciences Inc. since 2002. He is responsible for business

development for ASL’s Product Division. Jan also markets the WERA phased-array HF-radar

system manufactured by Helzel Messtechnik, Kaltenkirchen, Germany in North America. The

shore-based WERA provide reliable ocean current measurements with outstanding spatial and

temporal resolution suitable for the detection of near-shore Tsunamis.

Josef Cherniawski - InsLtute of Ocean Sciences, Canada Josef Cherniawsky is a Research Scientist at the Institute of Ocean Sciences, Department of

Fisheries and Oceans, Sidney, BC. Dr. Cherniawsky has about 30 years of experience in Physical

Oceanography, reflected in his publications (https://www.researchgate.net/profile/Josef_

Cherniawsky/contributions?ev=prf_act). In the past 10-15 years he has been processing and inter-

preting observations of sea level and tides from satellite altimeters, as well as working on

numerical models of tsunami waves. His paper “Numerical Simulations of Tsunami Waves and

Currents for Southern Vancouver Island from a Cascadia Megathrust Earthquake” (Pure Appl.

Geophys. 2007) was the first attempt to produce detailed models of tsunami waves at several

locations on the south coast of British Columbia that may be generated after a future megathrust (Mw=9) earthquake.

Tom Dakin - Ocean Networks CanadaTom started in physics and physical oceanography in 1981. He spent 12 years in the Military as

both a Maritime Engineer and Aerospace Engineer with a focus on underwater acoustics. He

spent 17 years at AML Oceanographic as the R&D manager.. He is best known for the time of

flight sound speed sensor which is the industry standard in hydrographic surveying. Other

designs include geothermal heat flux, conductivity, pressure, temperature, speed log, underwater

mass spectrometer, oil spill detector buoy, XSV, boric acid concentration, diver ascent rate meter

and a VLF hydrophone calibration system. Tom is now the Sensor Technologies Development

Officer at Ocean Networks Canada, providing test, engineering and marketing support to the

ocean instrumentation manufacturing industry to promote industry growth.

43

Richard Dewey - Ocean Networks Canada Richard is the Ocean Networks Canada Associate Director, Science. Richard is responsible for

coordinating and assisting all scientists and researchers using the observatories, from planning

to publication. He works with the Staff Scientists to support the science community. He has a B.

Sc. in Physics from UVic and a Ph.D. in Oceanography from UBC. His research interests are coastal

flows, mixing, turbulence, waves, and tides. He has conducted research throughout the Pacific

from Japan to California, and along the B.C., Alaskan, and Arctic coasts. He has used a variety of

profilers and ROVs, and deployed more than 150 moorings on over 100 oceanographic

expeditions. He is author of the Mooring Design and Dynamics MATLAB package, and specializes

in time series analysis.

Herb Dragert - Natural Resources Canada Herb Dragert is an emeritus research scientist with the Geological Survey of Canada and also

holds a visiting professor position with the School of Earth and Ocean Sciences at the University

of Victoria. His principal area of research has been the study of crustal deformation on the west

coast of Canada using geodetic techniques such as leveling, precise gravity, laser-ranging trilat-

eration, and GPS. Under his direction, the Geological Survey of Canada established the Western

Canada Deformation Array, the first continuous GPS network in Canada for the express purpose

of monitoring crustal motions. Data from this network led to the discovery of slow earthquakes

and “Episodic Tremor and Slip” in the Cascadia Subduction Zone. He served on the UNAVCO

Board of Directors from 2003–2006 and he was the Canadian Geophysical Union’s J. Tuzo Wilson

medalist for 2007. He was elected a Fellow of the American Geophysical Union in 2013. Dr. Dragert received his B.Sc. in

mathematics and physics from the University of Toronto and his M.Sc. and Ph.D. degrees in geophysics from the University of

British Columbia.

Marie Eblé - NOAA/Pacific Marine Environmental Laboratory, USA Marie Eblé graduated from Texas A&M University in 1984 with a MS in Physical Oceanography.

She was employed for two years with the consulting firms of Northern Technical Services and

Engineering Hydraulics, Inc. While in consulting, Marie was responsible for modeling and current

studies in support of project awards including Sewage Outfall siting for the cities of Seattle and

Tacoma and the U.S. Navy Homeport establishment in Everett, Washington. She became

proficient in the use of the Delft hydraulics ‘Hydro Delft’ and the NOAA ‘On Scene Spill’ models and

developed time series processing and analysis capabilities used for a varied number of contract

projects. In 1986, Marie began her career with the NOAA Pacific Marine Environmental Laboratory

in Seattle, Washington. She has been involved with development, testing, and deployment of

bottom pressure technology for observing tsunamis in the deep-ocean. Marie remains engaged

in tsunami research with applicability to operational forecasting.

44

Isaac Fine - Institute of Ocean Sciences, CanadaIsaac Fine (Isaak Fain) is a contractor at the Institute of Ocean Sciences, DFO Canada, Sidney, BC,

since 2004. He is originally from Russia, where he coursed his bachelor in the Byelorussian State

University, Minsk, USSR. In 1985 he finished his PhD in Geophysics. He worked in the Tsunami

department of Institute of Marine Geology and Geophysics, Yuzhno-Sakhalinsk, Russia, from

1976 to1989, and on the Heat and Mass Transfer Institute, Minsk, Belarus afterwards until 2004.

His main interests include the development of 2D finite-difference and finite-elements numerical

models for studying tsunamis, submarine landslides, and meteotsunamis. He works on data

processing, 3D ocean modelling.

David Fissel - ASL Environmental Sciences Inc., CanadaAfter completing an M.Sc. in physical oceanography at the University of British Columbia in 1975,

David worked as a research oceanographer at the Institute of Ocean Sciences of the Department

of Fisheries and Oceans (DFO) in Sidney B.C. In 1977, he co-founded ASL Environmental Sciences

Inc. (originally Arctic Sciences Ltd.), and has held a number of senior positions in the company.

David Fissel has managed hundreds of oceanographic projects, involving studies of ocean

currents, waves, and sea-ice in various parts of all three oceans bordering Canada as well as

overseas projects. Most of these projects involved input to the design of offshore oil and gas

facilities, port development, or environmental assessment and monitoring for coastal and

deepwater developments. David has served on the many Boards and advisory committees including Ocean Networks Canada

at the University of Victoria; and the Marine Environmental and Observation Prediction and Response (MEOPAR) National

Centre of Excellence (NCE) at Dalhousie University.

Jannette Frandsen - InsLtut NaLonal de la Recherche Scientifique, CanadaJannette Frandsen joined the Institute National de la Recherche Scientifique (Quebec, Canada) in

2013, as professor and chairholder of the research chair in coastal and river engineering. Her

research interests are in the areas of fluid dynamics, nearshore hydrodynamics, natural hazards,

fluid-structure interactions and vibrations. Part of her responsibilities are activities in the large

wave flume facility (http://lhe.ete.inrs.ca) and to establish a coastal science program. She received

her B.Sc. from the Technical University of Denmark, M.Sc. from Imperial College London and the

doctorate at Cambridge, engineering science (aeroelasticity) in 2000. She was appointed a

Departmental Lecturership at Oxford and concurrently held a Fellowship at Oriel College (1999).

From 1999-2008, she has taught and undertaken research in U.K., USA and Australia. Underpinning

her academic experience, she spent several years working in industry, where she undertook analysis and design of fixed and

floating offshore platforms, bridges, harbors, offshore wind farms and wave energy devices.

45

Ken Halcro - Canadian Hydrographic Service, Pacific Region, Fisheries and Oceans Canada Ken joined the Canadian Hydrographic Service (CHS), Fisheries and Oceans Canada in 1980 as a

hydrographic surveyor and graduated that same year from BCIT with Survey diploma. He

received Canada Land Surveyor commission in 1988 and graduated from UVic with a BSc in

Computer Science in 2000. Presently managing the data acquisition division of CHS Pacific, with

which the running and monitoring of the Canadian Tsunami Warning system falls under.

Jeffrey C. Harris - University Paris- Est, France Jeffrey Harris is a researcher at Ecole des Ponts (ENPC) with the Saint-Venant Hydraulics Lab (a

joint research lab between EDF R&D, CEREMA, and ENPC). His primary research interests are

ocean wave modeling and high performance computing. Previously he worked as a postdoctoral

researcher at the University of Rhode Island conducting tsunami hazard simulations that were a

part of the east coast component of the US National Tsunami Hazard Mitigation Program

(NTHMP). He presently is working on the development of a numerical wave tank for the modeling

of offshore structures such as wave energy converters, as well as contributing to other wave

modeling projects, including tsunami modeling, with a particular focus on parallel processing. Dr.

Harris received B.S. degrees in both mathematics and oceanography from the University of

Washington, and M.S. and Ph.D. degrees in ocean engineering from the University of Rhode

Island.

Martin Heesemann - Ocean Networks Canada Dr. Martin Heesemann joined NEPTUNE Canada as a Research Theme Integrator for Plate

Tectonic Processes and Earthquake Dynamics in July 2010. In 2012 his job title was changed to

Ocean Networks Canada Staff Scientist. Martin’s Ph.D. thesis at the University of Bremen (2008)

focused on the acquisition, processing and interpretation of marine temperature and pressure

data in the context of subduction zones. Most of his research is related to studying plate tectonic

processes and earthquake dynamics using numerical models and time-series analyses of CORK

borehole observatory, bottom pressure recorder, and seismometer data.

46

Joe Henton - Natural Resources Canada Joseph Henton is the leader of Natural Resources Canada’s “Plate Boundary Earthquake

Geohazards” activity and a Research Scientist with NRCan’s Canadian Geodetic Survey. Through

the application of satellite-based geodesy and absolute gravimetry, he investigates the time-

evolution of national and regional spatial reference frames and gravity datums as well as the

processes driving the observed change. In particular his research interests include crustal

deformation studies varying from broad-scale glacio-isostatic adjustment to regional earthquake

hazard studies at both active margins and continental interiors to local monitoring of surficial

doming near a CO2 sequestration injection site. His work investigates processes that operate on

time scales from long-term secular deformation, to transient geodetic signals that have durations of several weeks (such as

Episodic Tremor and Slip), and to “GPS seismology” studying surface coseismic displacements and surface waveforms.

Maia Hoeberechts - Ocean Networks Canada Dr. Maia Hoeberechts is the Associate Director of User Services at Ocean Networks Canada (ONC),

based at the University of Victoria. She manages the Data, Communications and Learning teams,

the groups responsible for creating products and services which meet the needs of the scientific

and broader user community of ONC. Maia joined ONC in 2010 as the Staff Scientist for Computer

Science and Engineering, a role which she held for three years before assuming her current

position in October, 2013. She continues her research as Adjunct Professor in the Department of

Computer Science at the University of Victoria focusing on interdisciplinary connections between

ocean science and computer science, principally in the areas of computer vision, automated

event detection and data visualization. Maia holds a PhD in Computer Science in the area of

computability theory and a BSc in Philosophy and Computer Science from Western University.

Tania L. Insua - Ocean Networks CanadaTania is the Ocean Analytics Program Manager at Ocean Networks Canada. She obtained her

bachelors degree in Marine Sciences and her Diploma of Advanced Studies in Marine Biology and

Aquaculture from University of Vigo. After several years working in the biological sciences she

became interested in engineering and modeling. She obtained MS and PhD degrees in Ocean

Engineering from the University of Rhode Island with research focusing on applying models of

sediment physical properties to past and present climate change. She has been an active

participant of several Integrated Ocean Drilling Program, sailing as a physical properties specialist

or ocean engineer.

47

Christian Janssen - Hamburg University of Technology, GermanyChristian F. Janßen is a post-doctoral research fellow at the Institute for Fluid Dynamics and Ship

Theory, Hamburg University of Technology (TUHH). He studied Civil Engineering and Computational

Sciences in Engineering (CSE) in Braunschweig, Germany. In 2010, he received his PhD from

Braunschweig University of Technology for his dissertation on kinetic approaches for the

simulation of non-linear free surface flow problems in civil and environmental engineering.

During the subsequent postdoc stay at University of Rhode Island, Dr. Janßen worked on advanced

Lattice Boltzmann multiphase models and efficient implementations of shallow water models on

graphics processing units for the simulation of long wave propagation and Tsunami events. In

2012, he joined the Institute for Fluid Dynamics and Ship Theory in Hamburg, where he is

responsible for the development of the efficient lattice Boltzmann code elbe. The code aims at

interactively monitored nonlinear free surface flow simulations in real-time and has already been

successfully applied to wave impact and wave-structure interaction problems in complex three-dimensional topologies.

Kim Juniper - Ocean Networks Canada Dr. Juniper has been a Professor in the School of Earth and Ocean Sciences and the Department

of Biology at the University of Victoria, and holder of the BC Leadership Chair in Ocean Ecosystems

and Global Change since 2006. He came to UVic from the Université du Québec à Montréal where

he was Professor of Biology and Director of the GEOTOP Research Centre. He received his BSc

from the University of Alberta (1976) and a PhD from Canterbury University in Christchurch, New

Zealand (1982). The primary focus of his research has been the biogeochemistry and ecology of

submarine hydrothermal systems. His interdisciplinary publications on deep-sea vents

encompass the fields of microbial ecology, biomineralization and benthic ecology. Other research

areas have included the microbial ecology of deep-sea sediments, and the seasonal dynamics of

arctic sea-ice microbial communities. Juniper previously served the NEPTUNE Canada project as Co-Chief Scientist in 2004-2006,

and was President of the Canadian Scientific Submersible Facility from 2001 to 2011.

James T. Kirby - University of Delaware, USAJim Kirby received his B. S. and M. S. from Brown University in 1975 and 1976, and his Ph.D. from

the University of Delaware in 1983. He has served on the Faculty of SUNY Stony Brook, the

University of Florida and the University of Delaware (1989-present). Recent research has focussed

on the mechanics of breaking wave crests, rip current dynamics, wave-current interaction in

strongly sheared flows, tsunami generation, propagation and inundation, and morphology of

tidal marshes. He has co-authored a number of public domain software packages including REF/

DIF, FUNWAVE, NearCoM and NHWAVE, and is author of 99 refereed journal articles. He received

the Walter L. Huber Civil Engineering Research Prize (1992) and the Moffatt-Nichol Port and

Coastal Engineering Award (2011) from ASCE. He has served as Editor or Editor in Chief for two

ASCE and one AGU journal. He presently serves on the Coordinating Committee for the National

Tsunami Hazard Mitigation Program.

48

Lucinda Leonard - University of VictoriaLucinda Leonard is an Assistant Professor in the School of Earth and Ocean Sciences at the

University of Victoria. She holds a Bachelors degree in Geology from Trinity College Dublin, and a

PhD in Geophysics from the University of Victoria. Lucinda works closely with scientists from

Natural Resources Canada at the Pacific Geoscience Centre, where she previously held a Visiting

Fellowship postdoctoral position. Her research interests include the study of large earthquakes

and tsunamis, including both recent and pre-historic events, as well as assessment of the seismic

and tsunami hazard related to future events. Recent work includes a preliminary tsunami hazard

assessment of Canada, and field studies following the 2012 Haida Gwaii tsunami.

Gwyn Lintern - Natural Resources CanadaDr. Lintern researches sediment hazards in the marine environment. He has worked on issues on

all three coasts, including submarine slope stability, dredging and port expansion, coastal erosion,

and storm surge modelling. The sediment issues Dr. Lintern studies often arise in infrastructure

and development projects. Sediment transport can cause damaging scour around structures

(e.g. pipelines, undersea cables), coastline erosion, and infill of navigation channels. He also

advises on activities such as dredging and trawling which can have damaging siltation effects.

Hannan Lohrasbipeydeh - Univeristy of VictoriaMr. Hannan Lohrasbi is a PhD candidate in electrical engineering at University of Victoria. His

major research interest is focused on the underwater acoustic signal processing specially source

detection and localization. He has collaboration with Ocean Network Canada since 2012. He was

a winner of IEEE Scholarship in 2012 and 2013. Due to his publication in underwater marine

mammal acoustics, he was appointed as chair of bio acoustic section in IEEE Ocean Conf. 2013 at

San Diego. His current research is focused on the underwater landslide detection and localization

which is considered as one of the trigger of the tsunami.

Roe Markham - Ocean Networks CanadaRoe is a Strategic Planning Officer for the Ocean Networks Canada Innovation Centre. He is a

management consultant with over 30 years of experience in program development and

management, business process design, and information systems. Roe’s background is in strategic

planning, building and implementing $100M programs in multi-stakeholder environments, and

executive sales and marketing. Industry subject areas include: IT services, public sector, defense,

airlines, banking, insurance, brokerage, manufacturing, ERP systems, HR systems, warehousing

and logistics systems, procurement, printing and distribution systems.

49

Scott McLean - Ocean Networks Canada Scott, a professional electrical engineer, brings over 21 years of ocean technology development

experience to ONC from Halifax, where he worked for eight years as chief technology officer and

vice-president of research and development at a high-tech oceanographic company. Scott’s areas

of expertise include sensor development, observing system design and sensor integration into

observing systems. From his experience in product development from concept through to

creation, Scott has the proven ability to turn partnerships and technology transfers from Canadian

and international groups into successful commercial products. Scott is the Ocean Networks

Canada Innovation Centre Director, and also the Business Development Officer for Ocean

Observing Technology.

Steven Mihaly - Ocean Networks Canada As a staff scientist for Ocean Networks Canada (ONC), Dr. Steven Mihaly focuses on supporting

interdisciplinary research within the theme “Ocean/Climate Dynamics and their Effects on Marine

Biota”. Steve holds Bachelor of Engineering (Mechanical) from Dalhousie University and a PhD in

Physical Oceanography from the University of British Columbia. He comes to ONC from Fisheries

and Oceans Canada where he was a research associate specializing in observational science

using moorings and ship-borne hydro water surveys in the Salish Sea, the west coast of Vancouver

Island and at rough topography associated with the Juan de Fuca Ridge.

Teron Moore - Emergency Management BCTeron Moore is the Provincial Seismic Specialist for Emergency Management BC. His portfolio

includes provincial leadership for earthquake, tsunami, and volcano hazards with an emphasis

on preparedness, mitigation, response, and recovery planning. Among other roles, Teron serves

as the Vice President of the Cascadia Region Earthquake Workgroup, is an executive member of

the BC Seismic Safety Council and co-leads the BC ShakeOut organizing committee. He holds a

Master’s degree in Disaster and Emergency Management from Royal Roads University in Victoria,

BC with a research focus on the evaluation of international earthquake and tsunami projects. He

also holds a professional designation in Project Management from the Project Management

Institute.

50

Dmitry Nicolsky - University of Alaska Fairbanks, USADmitry Nicolsky his B. S. from St. Petersburg State University, Russia in 2000, and his M.S. and

Ph.D. from the University of Alaska Fairbanks in 2003 and 2007. Dmitry’s recent research is

focused on development of the statewide tsunami inundation maps for emergency planning in

Alaska. Dmitry is the State Science Representative on the Coordinating Committee of the National

Tsunami Hazard Mitigation Program (NTHMP), a partnership between federal and state govern-

mental agencies designed to reduce the impact of tsunamis through hazard assessment, warning

guidance, and mitigation.

Benoit Pirenne - Ocean Networks CanadaBenoît Pirenne is Associate Director, Digital Infrastructure, at Ocean Networks Canada ONC). He

joined NEPTUNE Canada in October 2004 after having spent about 18 years at the European

Southern Observatory (ESO), a leading organization for astronomical research where he assumed

a number of scientific and technical (IT) positions in the area of data management. Now at the

University of Victoria, Benoît heads the Digital Infrastructure (DI) department that builds and

operates “Oceans 2.0”, the data management and archiving system for ONC. Oceans 2.0

comprises the software and systems infrastructure necessary to link world-wide user communities

to the NEPTUNE Canada and VENUS observatories. Benoît holds a Masters degree in Computer

Science from the University of Namur, Belgium (1986) and graduated in Computer Science from

the Institut St-Laurent in Liège, Belgium in 1983.

Alexander B. Rabinovich -- Shirshov Institute of Oceanology, RussiaDr. Alexander B. Rabinovich has more than 40 years of research experience in ocean dynamics,

coastal engineering, tides and currents, seiches, sea level changes and time series analysis. The

main focus of his research is tsunamis and is a world-leading specialist in this field. He received

his B.Sc. degree from the Moscow State University (Russia), his Ph.D. from the Marine Hydrophysical

Institute (Sevastopol) and his D.Sc. from the P.P. Shirshov Institute of Oceanology (Moscow). For

20 years he worked in the Russian Far East, focusing his research mainly on tsunamis, tides,

harbour oscillations and storm surges. For the last 20 years he has been working at the Shirshov

Institute of Oceanology in Moscow and also at a number of foreign research institutes and uni-

versities; he has recently immigrated to Canada. He has published more than 180 papers in Russian and international journals

on various aspects of tsunamis, tides and ocean dynamics. He has held academic positions at: Universitat de les Illes Balears

(Spain), Universita di Genova (Italy), Seoul National University (Korea), University of Alaska at Fairbanks (USA) and the University

of Tokyo (Japan). He has participated in a variety of applied projects, including tsunami hazard estimation, sediment transport,

pollutant dispersal, tidal analysis and storm surges.

51

Volker Roeber - Tohoku University, Japan Dr. Volker Roeber is an Assistant Professor at the Laboratory of Technology for Global Disaster

Risk at the International Research Institute of Disaster Science (IRIDeS) at Tohoku University,

Japan. Originally from the Lake Constance region in southern Germany, Volker obtained a Masters

degree in Coastal Geosciences & Engineering from Kiel University in 2001. He then spent two

years in Brazil where he worked for the oil and gas company Petrobrás. Volker then joined the

University of Hawaii where he graduated with a PhD in Ocean & Resources Engineering in 2010.

His PhD and subsequent postdoctoral research focused on numerical computations and

laboratory experiments with respect to nearshore wave processes. At IRIDeS, he is continuing his

research on numerical modeling and disaster mitigation of storm waves and tsunamis.

Garry Rogers - Natural Resources CanadaDr. Garry Rogers is a Senior Research Scientist with the Geological Survey of Canada specializing

in earthquake and tsunami studies. He is also an adjunct professor at the University of Victoria

where he is Principle Investigator for the seismograph network attached to Ocean Networks

Canada’s offshore NEPTUNE cabled observatory. For the past decade he has been representing

Canada on the UNESCO/IOC Intergovernmental Coordination Group for the Pacific Tsunami

Warning and Mitigation System. He also currently is a member of the National Research Council’s

Standing Committee for Earthquake Design, which is responsible for earthquake provisions in

the National Building Code of Canada. In recent years he has served on multi-year scientific

advisory committees to the Earthquake Program of the US Geological Survey, the Southern

California Earthquake Center and NEPTUNE Canada.

Andreas Rosenberger - Ocean Networks CanadaAndreas Rosenberger graduated from University of Hamburg, Germany, in Geophysics, applied

Mathematics and Physics in 1988. He earned a doctoral degree from Alfred Wegener Institute for

Polar and Marine Research and University of Bremen (Germany) in 1991. Before moving to

Canada in 1999 he spent six years teaching as an assistant professor, department of marine

technology, at the University of Bremen. In Canada, he initially worked as an R&D scientist for

Quester Tangent Corp. on marine sonar systems and consulted with the Geological Survey of

Canada (GSC) on the design of a new type of real-time reporting seismograph system. In 2001 Dr.

Rosenberger joined the GSC where his research focused on problems in real-time seismology. He

designed and implemented the software and IT infrastructure of the GSC’s Internet accelerome-

ter network and collaborated with the British Columbia Ministry of transportation and

Infrastructure for the British Columbia Smart Infrastructure Monitoring System (BCSIMS) project.

On leave from the GSC in 2006/2007, Dr. Rosenberger worked for the German Marine Research Consortium (KDM) as

coordinator, marine instrumentation, for the German Indonesian Tsunami Early Warning System (GITEWS) project. Since

January 2014 he is consulting Ocean Networks Canada on the WARN project.

52

Michael Schmidt - Natural Resources Canada Mike Schmidt, P.Eng. was until recently head of the Canadian Crustal Deformation Service at the Geological

Survey of Canada’s Pacific Geoscience Centre. He has been involved with the application of GPS to crustal

deformation, earthquake research and tsunami early warning for over 20 years, developing the Western

Canada Deformation Array a network of continuously Operating GNSS Reference Stations in western

Canada. He has worked with GPS since 1980, developing early algorithms for the analysis of GPS data,

managed several major research projects and field operations in western Canada and the Arctic. Along

with colleagues at the GSC he co-developed one of the first prototype GPS augmentations to tsunami early

warning systems. He is currently a consultant based in North Saanich, B.C.

Elena Tolkova - NorthWest Research Associates, USAElena I. Tolkova graduated from Nizhny Novgorod State University (NNSU) in Russia in 1985 and stayed

with the University for 15 years after the graduation, as a research scientist and later as a professor of

General Physics Department, specializing in optical interferometry, image processing, and time-series

analysis. She received Kandidat Nauk (Ph.D equivalent in Russia) degree in 1994 in the NNSU. In 2000, she

moved to the USA with her husband, son, and daughter. Elena attained experience with tsunami modeling

and forecasting while working in the University of Washington for NOAA’s Center for Tsunami Research in

2005-2013. Her major research interests are tsunami intrusion into rivers, tsunami dynamics in natural

resonators, forecasting techniques, and computer code development for tsunami simulations. Since 2013,

she pursues these areas with NorthWest Research Associates, Inc.

Tummala Srinivasa Kumar - Indian National Centre for Ocean Information Services, IndiaDr. Srinivasa Kumar Tummala is a Scientist heading Advisory Service & Satellite Oceanography Group at

the Indian National Centre for Ocean Information Services (INCOIS) and also in-charge of the Indian

National Tsunami Early Warning Centre. As Project manager, he contributed immensely to the establish-

ment of state-of-the-art facilities at the Indian Tsunami Early Warning Centre. Recognizing his contribu-

tions for establishing the Indian tsunami early warning centre, the Indian Ministry of Mines conferred him

with the prestigious “National Geoscience Award” in 2010. He also contributed to the Indian Ocean Tsunami

Warning and Mitigation System (ICG/IOTWS) in various capacities. Currently he is a Vice Chair the ICG/

IOTWS and Chairman of TOWS Working Group’s Inter-ICG Task Team on Tsunami Watch Operations. He

holds a Master’s degree in Marine Biology & Oceanography and Ph.D. in Marine Science. He authored several publications in interna-

tional peer reviewed journals.

53

Carlos E. Ventura - University of British Columbia, CanadaCarlos Ventura is a Civil Engineer with specializations in structural dynamics and earthquake

engineering. He has been a faculty member of the Department of Civil Engineering at the

University of British Columbia (UBC) in Canada since 1992. He is currently the Director of the

Earthquake Engineering Research Facility (EERF) at UBC, and is the author of more than 400

papers and reports on earthquake engineering, structural dynamics and modal testing. He is a

member of the Canadian Academy of Engineering and Fellow of Engineers Canada, also a

member of several national and international professional societies, advisory committees and

several building and bridge code committees. Dr. Ventura has conducted research about

earthquakes for more than thirty years. His current research includes the development and

implementation of performance-based design methods for seismic retrofit of low rise school

buildings, novel techniques for regional estimation of damage to structures during earthquakes,

and on structural health monitoring of building, bridges and dams. His consulting practice includes projects with companies

and government institutions in North America, Central America, South America, Asia and Europe.

Kelin Wang - Natural Resources CanadaKelin Wang is a senior Research Scientist with the Geological Survey of Canada. Most of his current

research is on geodynamics of subduction zones and related earthquake and tsunami hazards,

but he has also worked on a range of other topics regarding the thermal, mechanical, and hydro-

geological processes of Earth’s lithosphere. He obtained his B.Sc. degree at the Peking University

in 1982 and Ph.D. at the University of Western Ontario in 1989. He is an Adjunct Professor at the

University of Victoria and an Honorary or Guest Professor for several other scientific institutions.

He was or still is on the Editorial Boards of a number of scientific journals such as Journal of

Geophysical Research, Geology, Journal of Geodynamics, Science in China (Earth Science), and

Earthquake Science.

Tim Webb - ESSA Technologies, CanadaTim is an analyst, simulation modeler, and emergency management professional with over 33

years of experience designing, developing, and deploying decision support tools. His work has

focussed on providing mission critical information to managers and planners in the fields of

emergency preparedness and response and natural resources management. Tim has directed

projects around the world and been the technical visionary on a number of major systems

projects in North America, the Middle East and Australia. His recent work has focussed on the

development of tools for natural hazard damage forecasting and information management

systems to fuse data from many different sources into actionable intelligence for emergency

planners and responders. Tim is a resident of Tofino and is active in the volunteer community on

the west coast of Vancouver Island; he has been a SAR volunteer for over fifteen years serving in the role of SAR Manager and

president of the Westcoast Inland Search and Rescue Society. Tim served as deputy Emergency Coordinator for Tofino for a

number of years and as director and co-chair on the board of the Clayoquot Biosphere Trust for six years.

54

Stuart Weinstein - NOAA/NWS Pacific tsunami Warning Center, USAStuart Weinstein is the Assistant Director of the Pacific Tsunami Warning Center (PTWC) located

in Ewa Beach Hawaii, USA. He has held this position since 2005, and has been with the PTWC

since 1998. He oversees the day-to-day operations of the warning center and conducts tsunami

training with the International Tsunami Information Center based in Hawaii. Stuart received

his Ph.D. in 1991 from the Johns Hopkins University (Maryland) for his dissertation on thermal

convection in planetary mantles. He continued in this area of research with a National Science

Foundation Postdoctoral Fellow in Earth Sciences grant at the University of Michigan in 1991 and

at the University of Hawaii as the SOEST Young Investigator in 1993. Stuart worked for Bloomberg

L.P. New York, from 1996-1997 writing financial analysis software. New York reminded Stuart

how much he appreciated Hawaii’s climate. This prompted Stuart’s move back to Hawaii in 1998.

Onat Yazir - University of Victoria, CanadaDr. Yagiz Onat Yazir completed his Ph. D. in Computer Science in 2011, and is currently a post-

doctoral fellow at the University of Victoria under the supervision of Dr. Yvonne Coady. His

research mainly focuses on autonomous control of various scales of distributed infrastructures

with particular emphasis on distributed computing, discrete event simulation platforms, resource

allocation management in the cloud, adaptive routing in mobile ad hoc networks, and multiple

criteria decision analysis. Throughout his Ph. D. and his post-doctoral fellowship, Dr. Yazir has

participated in a number of scientific projects that involve both purely academic work and indus-

try-academia research collaborations.

55A N I N I T I A T I V E O F

Ocean Networks Canada is developed through: A consortium of 12

Canadian universities, government labs and international institutions.

Major Funders:BC Knowledge Development Fund

British Columbia Ministry of Advanced Education and Labour Market Development

Canada Foundation for Innovation (CFI)

Canada’s Advanced Research and Innovation Network (CANARIE)

Natural Sciences and Engineering Research Council of Canada (NSERC)

Networks of Centres of Excellence (NCE)

University of Victoria (UVic)

In Partnership with:Federal and provincial departments and agencies

US and international academic institutions

Canadian, US and international companies

Contact:

Ocean Networks Canada

Technology Enterprise Facility (TEF)

University of Victoria

PO Box 1700 STN CSC

2300 McKenzie Avenue

Victoria, BC Canada V8W 2Y2 250.472.5400 | info@oceannetworks.ca | www.oceannetworks.ca

@Ocean_Networks | /OceanNetworksCanada | oceannetworks.ca