VIDEO ANALYSIS OF THE MARCH 2011 TSUNAMI

IN JAPAN’S COASTAL CITIES

Nguyen Ngo

and

Ian N. Robertson

Research Report UHM/CEE/12-11

December 2012

2

Abstract

On March 11, 2011, many coastal cities in Japan were struck by a devastating tsunami following an 8.9 magnitude earthquake. Japanese citizens documented this catastrophic event for the world to see using personal video cameras. Recorded video footage was readily available over the internet, providing the main source of raw data for this study. Free and basic software, such as Google Earth and Quicktime, were used to analyze the videos. This freely available source of data was used to study tsunami behavior. By measuring changes in flow depth and speed over the duration of a video, time history graphs for the tsunami flow characteristics were developed. Time history results were successfully obtained for tsunami inundation in Kamaishi, Ofunato, and Haragama Soma. For other locations, including Minamisanriku, Sendai, Shiogama, and Tagajo, discrete measurements were recorded in order to gain more evidence on a regional scale and to demonstrate the procedure and results of this analysis. The specific momentum flux of the tsunami waves was determined as the product of flow depth and the square of flow velocity, (ℎ푢2). Because momentum flux is proportional to drag force, it was used to estimate the design forces for structures. Results for maximum momentum flux obtained through video analysis of discrete locations in three coastal communities are shown in the table below. The corresponding design force per unit length 1 of the structure is determined using the equation 퐹 = 휌 퐶 푏(ℎ푢2), assuming a drag coefficient, C , of 2.0. 퐷 2 푤 퐷 D

Location Max momentum flux Average Pressure m3/s2 kN/m2 (kip/ft2) Kamaishi 56.0 14.0 (0.29) Ofunato 139.5 46.5 (0.97) Haragama Soma 136.8 22.8 (.048)

The momentum flux can be used to determine the maximum forces induced by a tsunami wave. Yeh et al. (2006) present an equation to determine the maximum momentum flux, dependent either on run-up distance or run- up elevation. Results from the video analysis were compared with this prediction in order to determine the validity of the theoretical prediction in real-life cases. Unfortunately, real-world topography presented difficulties in obtaining a clear estimate from the momentum flux envelope, which did not agree with the momentum flux obtained through video analysis.

Video analysis was proven capable of generating reasonable tsunami flow conditions. However, results from the video analysis do not show a strong correlation with the estimation for the maximum momentum flux. Therefore, video analysis can be utilized whenever video records are available in order to obtain reliable results for real tsunami loads. The maximum momentum flux envelope should not be used to estimate design forces for tsunamis. Instead, results from video analysis can be used to provide better estimates for the maximum momentum flux.

3

Table of Contents ABSTRACT ...... 3 1 INTRODUCTION ...... 7 1.1 BACKGROUND ...... 7 1.2 INTRODUCTION ...... 8 1.3 LITERATURE REVIEW...... 10 1.3.1 TSUNAMI INUNDATION AND RUNUP SURVEY BY THE 2011 TOHOKU EARTHQUAKE TSUNAMI JOINT SURVEY GROUP ...... 10 1.3.2 PARTICLE IMAGE VELOCIMETRY ANALYSIS OF TSUNAMIS ...... 12 1.3.3 MOMENTUM FLUX ...... 14 1.4 SUMMARY ...... 15 2 VIDEO ANALYSIS ...... 16 3 ANALYSIS ...... 18 3.1 TIME HISTORY ANALYSIS ...... 19 3.1.1 Kamaishi, Iwate Prefecture ...... 19 3.1.2 Ofunato, Iwate Prefecture ...... 33 3.1.3 Haragama Soma, Fukushima Prefecture ...... 43 3.1.4 Onagawa, Miyagi Prefecture ...... 52 3.1.5 Summary on Momentum Flux Envelopes ...... 60 3.2 DISCRETE EVENT ANALYSIS ...... 63 3.2.1 Minamisanriku, Miyagi Prefecture ...... 63 3.2.2 Sendai, Miyagi Prefecture ...... 68 3.2.3 Shiogama, Miyagi Prefecture ...... 72 3.2.4 Tagajo, Miyagi Prefecture ...... 76 4 SUMMARY AND CONCLUSION ...... 81 APPENDIX ...... 84 A. REPORT FROM THE JAPAN METEOROLOGICAL AGENCY (JMA) ...... 84

4

List of Tables and Figures Table 3.1.1-1: Summary of results for tsunami at Kamaishi Port ...... 31 Table 3.1.2-1: Summary of results for tsunami at Ofunato Port ...... 42 Table 3.1.3-1: Summary of results for tsunami at Haragama Soma ...... 51 Table 4-1: Summary of results ...... 81

Figure 1-1: Overall map of earthquake epicenter and affected sites reported in this project ...... 7 Figure 1-2: Google Earth image of survey data of inundation and runup values (Mori and Takahashi, 2012) ...... 10 Figure 1-3: Regional analysis of tsunami inundation height (Credit: Mori and Takahashi, 2012). “Regional analysis of tsunami inundation height (unit: m) and distance (unit: km; circle: inundation height, triangle: run-up height, lines: empirical curve).” ...... 11 Figure 1-4: Tsunami height time history for Kesennuma Bay (Credit: Fritz et. al., 2012) ...... 13 Figure 3.1.1-1: Google Earth image of Kamaishi port dated 4/26/2005 ...... 19 Figure 3.1.1-2: Google Earth image of Kamaishi port dated 4/26/2005 ...... 20 Figure 3.1.1-3: Google Earth image of Kamaishi port dated 3/31/2011 ...... 21 Figure 3.1.1-4: Debris #1 ...... 22 Figure 3.1.1-5: Surge #1 (Kamaishi) ...... 23 Figure 3.1.1-6: Debris #2 (Kamaishi) ...... 24 Figure 3.1.1-7: Debris #3 (Kamaishi) ...... 24 Figure 3.1.1-8: Debris #4 (Kamaishi) ...... 25 Figure 3.1.1-10: Debris #6 (Kamaishi) ...... 26 Figure 3.1.1-11: Debris #7 (Kamaishi) ...... 27 Figure 3.1.1-11: Debris #8 (Kamaishi) ...... 28 Figure 3.1.1-12: Debris #9 (Kamaishi) ...... 29 Figure 3.1.1-13: Debris #10 (Kamaishi) ...... 29 Figure 3.1.1-14: Time history diagram for tsunami at Kamaishi (First Wave) ...... 30 Figure 3.1.1-15: Time history diagram for tsunami at Kamaishi (Second Wave) ...... 30 Figure 3.1.2-1: Google Earth satellite image of Ofunato Bay ...... 33 Figure 3.1.2-2: Google Earth image of Ofunato Port on July 22, 2010...... 34 Figure 3.1.2-3: Google Earth image of Ofunato Port on February 21, 2012...... 34 Figure 3.1.2-4: Satellite image of seawall at Ofunato port before (7/22/2010) and after (2/21/2012) ...... 35 Figure 3.1.2-5: Debris #1 (Ofunato) ...... 36 Figure 3.1.2-7: Debris #3 (Ofunato) ...... 37 Figure 3.1.2-8: Debris #4 (Ofunato) ...... 38 Figure 3.1.2-9: Debris #5 (Ofunato) ...... 39 Figure 3.1.2-10: Debris #6 (Ofunato) ...... 39 Figure 3.1.2-11: Debris #7 (Ofunato) ...... 40 Figure 3.1.2-12: Maximum depth of flow was estimated at 6 meters ...... 41 Figure 3.1.2-13: Time history of tsunami at Ofunato Port ...... 42 Figure 3.1.3-1: Google Earth Satellite Image of Haragama Port ...... 43 Figure 3.1.3-2: Post-tsunami image from Google Street View of the reference building which performed well against the wave impact ...... 44 Figure 3.1.3-3: Google Earth image of Haragama port on September 10, 2010...... 45 Figure 3.1.3-4: Google Earth image of Haragama port on April 5, 2011...... 45 Figure 3.1.3-5: Surge #1 ...... 46 Figure 3.1.3-6: Debris #1 (Haragama) ...... 47 Figure 3.1.3-7: Haragama seawall location; the seawall location is not as apparent in Frame 15124 but by comparing it with an earlier frame, the seawall location can be established with some certainty ...... 48 Figure 3.1.3-8: Wave impact and distance travelled by wave ...... 48 Figure 3.1.3-9: Debris #2 ...... 49 Figure 3.1.3-9: Debris #3 (boat) ...... 50 Figure 3.1.3-10: Time history of tsunami at Haragama Soma ...... 50 Figure 3.1.4-1: Elevation profiles of Kamaishi from Google Earth...... 61 Figure 3.1.4-2: Elevation profiles of Ofunato from Google Earth ...... 62 Figure 3.2.1-1: Satellite image of Minamisanriku on June 24, 2010 ...... 64 Figure 3.2.1-2: Satellite image of Minamisanriku on April 5, 2011 ...... 64

5

Figure 3.2.1-3: Surge #1 ...... 65 Figure 3.2.1-4: Surge #2 ...... 66 Figure 3.2.1-5: Debris #1 ...... 67 Figure 3.2.2-1: Satellite image of cameraman location on April 3, 2010 ...... 69 Figure 3.2.2-2: Satellite image of cameraman location on April 5, 2011 ...... 69 Figure 3.2.2-3: Surge #1 (Mitsui Mall) ...... 70 Figure 3.2.2-2: Debris #1 (Mitsui Mall) ...... 71 Figure 3.2.3-1: Satellite image of cameraman location on March 30, 2009 ...... 73 Figure 3.2.3-2: Satellite image of cameraman location on April 5, 2011 ...... 73 Figure 3.2.3-3: Surge #1 (Shiogama) ...... 74 Figure 3.2.3-4: Debris #1 (Shiogama) ...... 75 Figure 3.2.4-1: Satellite image of cameraman location on March 30, 2009 ...... 77 Figure 3.2.4-2: Satellite image of cameraman location on April 5, 2011 ...... 77 Figure 3.2.4-3: Surge #1 (Tagajo) ...... 78 Figure 3.2.4-4: Debris #1 (Tagajo) ...... 79 Figure 3.2.4-5: Debris #2 (Tagajo) ...... 79 Figure 3.2.4-6: Debris #3 (Tagajo) ...... 80

6

1 Introduction

1.1 Background

The 2011 Great Japan Earthquake occurred at 14:46:24 JST (5:46:24 UTC) on Friday, March 11, 2011, with an epicenter 70 kilometers off the coast of Japan (USGS website, 2011). Figure 1-1 shows the epicenter as well as some of the major affected sites analyzed in this study. After the traumatizing magnitude 8.9 earthquake, the people of Japan faced an even greater catastrophe. Tsunami waves flooded Japan’s coastal cities, claiming thousands of lives and causing massive destruction.

Onagawa

Figure 1-1: Overall map of earthquake epicenter and affected sites reported in this project

The earthquake prompted local news crews into action. Helicopter crews captured the historic event as the tsunami waves washed away homes and flooded buildings. Alongside professional media coverage, hundreds of

7

first-person accounts were recorded on video and uploaded on to the Internet. This allowed everyone around the world to witness the enormous scale of destruction from this disaster.

Prior to this tsunami, few videos were available that captured tsunamis on land. The 2009 Samoa Tsunami had only one known video, while the 2010 Chile Tsunami occurred very early in the morning, hindering any meaningful video recording. Limited research was done on the Indian Ocean Tsunami in 2004, which included flow velocity analysis similar to this project (Fritz, et al., 2006). These videos are the first of many that will document tsunami events and may prove to be an invaluable resource for future research.

1.2 Introduction

Currently, the design of structures for tsunami loading is not well established, in comparison to wind and earthquake loading. Studies on tsunami flow characterization on land are few and far between given that tsunamis rarely occur. So when tsunamis do strike, especially one of this magnitude, the event spurs research in tsunami behavior (duration and cycles) and effects (hydrostatic and hydrodynamic forces). Current design provisions for flooding can be found in the ASCE 7 Minimum Design Loads for Buildings and Other Structures. Although some tsunamis do resemble flooding, the provisions presented in ASCE 7 do not address tsunami loads. ASCE 7-05 comments that flood characteristics can be very different between riverine and coastal areas. The Federal

Emergency Management Agency (FEMA) presents an extensive resource on tsunamis in their P-646 Design of

Structures for Vertical Evacuation from Tsunamis. FEMA P-646 represents the current state of structural engineering research on tsunamis and provides the design criteria for structures to resist tsunami loads. However, since P-646 is a guideline written in non-mandatory language, it has not been enacted into law.

An effort is currently underway to develop a chapter entitled ‘Tsunami Loads and Effects’ for the ASCE 7 standard. This research project aimed to harness the vast amount of data provided by first-person video recordings of the 2011 Tohoku Tsunami. This project will attempt to serve as a proof of concept for the analysis of first-person video evidence. Through observing video footage, the characteristics of the tsunami, such as flow speed and depth, will be estimated and discussed. Discussions will be framed around the objective of determining design loads for structures.

Tsunami waves present a lot of different types of forces. These forces include: hydrostatic forces, buoyant forces, hydrodynamic forces, surge forces, impact forces, and breaking wave forces (Yeh, et al. 2005). Most of

8

these forces are dependent on two variables: depth and flow velocity. Depth can be measured in a number of ways.

The use of tidal gauges is a well-established method for obtaining depth with respect to time, but tidal gauges are generally only located in ports or harbors. Watermarks on buildings and trees provide a convenient record of the maximum depth. First-person videos can provide a time history or spot record of flow depth during the tsunami.

On the other hand, measurements for flow velocity are a lot more difficult to obtain. Video recordings provide a solution by capturing the motion of the tsunami and keeping a record for analysis. Although this analysis method is indirect, videos can be inspected on an on-demand basis and can be very precise with respect to time.

There are many research teams contributing to new developments in tsunami structural design. The following section presents a few reports that share similar goals in determining tsunami velocities, depths, and forces. These resources were used as a basis for the research performed for this project, in developing an appropriate procedure for video analysis, and in evaluating the validity of its results.

9

1.3 Literature Review

1.3.1 Tsunami Inundation and Runup Survey by The 2011 Tohoku Earthquake Tsunami Joint Survey Group

In an extremely extensive report, Nobuhito Mori and Tomoyuki Takahashi (2012) collected tsunami data from a large group of surveyors and researchers. The compilation of this survey can be seen in Figure 1-2, which shows the data collected and plotted in Google Earth. Mori and Takahashi analyzed the changing tsunami depth over time, as well as the max depth distribution on a regional scale. With respect to time, numerical models were used to estimate the arrival time of the maximum tsunami height in relation to the distance from the epicenter. With the regional scale, Mori and Takahashi studied the factors from topography and proximity to the epicenter and how it can affect the maximum tsunami height. For a structural designer, the magnitude of tsunami forces is the major concern.

Figure 1-2: Google Earth image of survey data of inundation and runup values (Mori and Takahashi, 2012)

The report defines the tsunami height in three different ways: (1) the height of the tsunami wave before crossing the shoreline, (2) inundation height based on the mean sea level, and (3) the run-up height, which is the elevation of the max run-up distance (Mori and Takahashi, 2012). Large inundation depths, up to 20 meters,

10

occurred at locations near the epicenter of the 8.9 magnitude earthquake. The tsunami reached greater depths in areas that had steep topography. In contrast, flatter areas, like the Sendai plains, had greater run-up distances and lower inundation depths. Figure 1-3 shows a scatter plot of the depth of the tsunami was plotted against the distance inland from the coast. By compiling hundreds of discrete measurements, Mori and Takahashi were able to demonstrate the relationship between tsunami depth and distance from shoreline for each of the affected regions.

The solid and dashed lines are functions of exponential decay at different degrees, which can be used to estimate the tsunami depth. This information can be particularly useful for a structural engineer because it allows the engineer to easily determine the depth for a design tsunami based on the structure’s distance from the coast. The plot also indicates the run-up distance. Areas with low tsunami depth, but long run-up distances may still pose great risk for one- and two-story residential homes.

Figure 1-3: Regional analysis of tsunami inundation height (Credit: Mori and Takahashi, 2012). “Regional analysis of tsunami inundation height (unit: m) and distance (unit: km; circle: inundation height, triangle: run-up height, lines: empirical curve).”

11

In Mori and Takahashi’s report, the maximum inundation and its location were documented across the affected regions. Both inundation depth and flow velocity have a large effect on the tsunami force for structures.

Hydrostatic pressure increases as depth increases and hydrodynamic pressure increase with flow depth and the square of flow velocity. The next section will review an research of flow velocity of tsunami waves using video analysis.

1.3.2 Particle Image Velocimetry Analysis of Tsunamis

A research team led by Dr. Hermann Fritz has done extensive work in post-tsunami analysis using video evidence. In a survey of the 2004 Indian Ocean tsunami, Dr. Fritz’s team collected real-world coordinates of control points which could be referenced to a video of the tsunami at Banda Aceh, Indonesia. Using particle image velocimetry (PIV) analysis on the video file, the research team measured the motion of the fixed structures in the video and related this result to the motion of the camera. The team then transformed the video plane coordinates to real world coordinates using a process called direct linear transformation (DLT). This caused the video frame to be warped in order to isolate the motion of the debris flowing in the tsunami. Lastly, PIV analysis was used to measure the overall motion of the debris in the tsunami and the difference between the results of the two PIV analyses returned the net velocity of the tsunami debris. The tsunami flow speeds at Banda Aceh ranged from 2 to 5 m/s

[Fritz et al., 2006].

After the 2011 Japan Tsunami, Dr. Fritz’s team repeated the procedure at Kesennuma Bay. The team implemented LiDAR (Light Detection and Ranging) scans to capture more accurate three dimensional coordinates during their survey trip. With several LiDAR scans coupled with GPS control points, a three dimensional coordinate system was established to be used to transform the reference video. PIV analysis was performed similarly to the procedure for the Indian Ocean tsunami. The analysis was used to determine flow depth and velocity in the center of the harbor channel. The maximum velocity of the tsunami flow was measured at 11 m/s during drawdown [Fritz et al., 2012]. The team also produced a graph of the tsunami depth time history, which included the initial inflow phase and the drawdown phase, seen in Figure 1-4.

12

Figure 1-4: Tsunami height time history for Kesennuma Bay (Credit: Fritz et. al., 2012)

There are many advantages to PIV analysis. Results are very accurate and precise. Velocity vectors are recorded in both x- and y-components on a horizontal plane. Flow velocity can be recorded instantaneously providing a more vivid picture of the changes in hydrodynamic forces. However, this procedure is highly technical and laborious. Similar to the analysis on which this report is based, the results are limited by the scope of the video footage and therefore can only produce data for a specific location and for the duration of the video. Results from video analysis may not be an accurate representation of critical tsunami loads. PIV analysis is extremely effective at tracking a large amount of debris, but in many cases, it is limited by the availability of suitable floating debris objects. As seen in Figure 1-4, even though a complete time history was generated for flow depth, only isolated estimates were obtained for flow velocity.

From the survey on tsunami depths by Mori and Takahashi and the research on flow velocity by Fritz, there is a substantial amount of data to be used to compare new results from this project. Structural engineers are not as concerned with flow depth and velocity as they are about the resulting forces. The next section will cover a procedure for estimating forces from depth and velocity and for estimating maximum forces from given topography data.

13

1.3.3 Momentum Flux

In 2006, Dr. Harry Yeh published a report titled “Maximum Fluid Forces in the Tsunami Runup Zone”, where he concluded that maximum, or design, tsunami forces can be determined when momentum flux is known.

Momentum flux is the product of the depth of tsunami flow multiplied by the square of the tsunami flow velocity.

With momentum flux, the drag force on structures can easily be determined. The basic equation for drag force is as follows,

1 퐹 = 휌퐶 퐴푢2 (Eqn. 1) 퐷 2 퐷 where ρ is the density of water, CD is the drag coefficient, A is the wetted area whose plane is perpendicular to the flow direction, and u is the speed of the flow. In terms of momentum flux, the drag force can be written as,

1 퐹 = 휌퐶 푏(ℎ푢2) (Eqn. 2) 퐷 2 퐷 where b is the breadth of the structure (perpendicular to the flow direction), h is the flow depth, and hu2 is the momentum flux. This equation provides the tsunami design force if inundation depth and flow velocity are known.

For example, by using the results for inundation depth and flow velocity from the particle velocimetry analysis by

Fritz et al. (2012), the maximum momentum flux observed in the video was determined. The momentum flux reached a maximum of 338.8 m3/s2 during drawdown. Assuming CD = 2.0 for a rectangular structural element, this corresponds to a lateral force of 338.8 kN/m or 23.2 k/ft.

In many cases, given a particular site, these parameters cannot be directly determined. Just like with wind design and seismic design, tsunami forces should be determined based on the geography of the site. Yeh et al.

(2006) studied various wave forms in order to develop such a procedure.

To determine an appropriate design momentum flux, Yeh defines the envelope for the maximum flux, which requires only a few known parameters. Yeh et al. (2006) references a number of other independent studies to support his conclusion. Carrier (2003) was credited for his work with the shallow-water theory and different initial wave forms. Carrier studied different initial wave forms and their resulting momentum flux and found that the maximum momentum flux occurs at the initiation of the drawdown phase. Drag force is proportional to momentum flux and can be easily calculated when the depth and flow velocity are known. Yeh also references independent studies from Ramsden (1993) and Arnason (2005), both coming to similar results for the maximum surging force.

By either eliminating or quantifying the effects of the surging force, a comprehensive force estimate can be

14

determined directly from momentum flux. Yeh added several models to Carrier’s and developed an envelope for the maximum momentum flux, given by the equation,

ℎ푢2 = 0.11(푥/푙)2 + 0.015(푥/푙) gα2푙2

(Eqn. 3) where α is the beach slope, l is the inundation distance inland from the shoreline, g is the gravitational constant, x is

2 the distance of the structure from the inundation distance, and hu is the momentum flux. The drag coefficient, CD, varies depending on the shape of the structure and is at a maximum of 2 for square columns. Yeh suggests that with a conservative drag coefficient value, reasonable tsunami design forces can be determined. From the FEMA P-55

Coastal Construction Manual, Yeh presents an alternate equation for the momentum flux envelope,

ℎ푢2 푧 푧 2 = 0.125 − 0.235 + 0.11 ( ) gR2 푅 푅

(Eqn. 4) where R is the elevation of the maximum runup distance and z is the elevation of the site of interest.

1.4 Summary

From the literature review, several methods for the determination of tsunami forces were presented. Should tsunami design forces be entirely dependent on maximum inundation depth? Tsunami velocities vary from highs of

10 m/s to 20 m/s and are sometimes unpredictable due to local topography in urban areas. It may be sufficient to assume a flow velocity of 20 m/s for all tsunami resistant buildings, especially those designated for tsunami evacuation. Max inundation depths are much easier to map than max flow velocity. Furthermore, for a certain location, max flow velocity is likely to have a larger range among different tsunami events, whereas inundation depth will likely reach the same maximum depth.

On the other hand, maybe recent concepts like momentum flux or tsunami fragility (Suppasri et al., 2010) are more appropriate for tsunami resistant design. [Expand on momentum flux] “Video […] data will be helpful to estimate the necessary dynamic information” (Mori, 2012). This project will explore the validity of these methods by analyzing the results obtained from video analysis.

15

2 Video Analysis

For this project, both depth and flow velocity were measured using video analysis. These results were used to also determine the momentum flux of the tsunami in order to determine a design force.

Because concrete structures have a lot more weight in comparison to wood structures, concrete buildings performed relatively well under tsunami loading. Many concrete buildings survived the tsunami and so they made good references to estimate the changing depth of the tsunami. A common metric used were story heights, which was typically estimated at 3 meters. A depth measurement and the respective time frame were recorded every time the depth would increase or decrease by a meter. Measurements for depth were also recorded for every surge velocity or debris velocity result. Video evidence can only be analyzed from the camera’s perspective and so, the accuracy of estimating the tsunami depth is limited. Google Earth’s function, Street View, was used to gain better viewpoints. Because these results are estimated, they will be checked with the results from Mori et al. (2011). The depth measurements from Mori et al. (2012) are from post-event surveys and are fairly accurate and precise.

A very basic approach was taken to determine the speed of the tsunami flow. Using Google Earth, satellite images were matched to the locations of the videos. Geographical landmarks were located and were used to establish distances. To measure time, still images from the videos were saved and their respective frame number was recorded. The selected images captured the moment when a wave or a piece of debris crossed a landmark. The precision of time was dependent on the frame rate of the video, typically 30 frames per second. Compared to the analysis done by Fritz et al. (2011), this video analysis is less systematic and instead is very dependent on engineering judgment. Engineering judgment dictated several factors; the choice of debris to track to represent the flow velocity, the timeframe for which the debris was tracked, and position of geographical landmarks based on the perception of the camera view. Therefore, the results may be a lot less precise and need to be compared with typical flow velocities.

For videos that captured uninterrupted footage of complete cycles of the tsunami, the results were compiled into a time history plot. The frame that corresponded to the time of the initial run-up was used to establish the start of the tsunami. The Japan Meteorological Agency (JMA) issued a report that indicated the start time of the tsunami in the different cities along Japan’s coast. Time history graphs are useful because they illustrate the changes in the tsunami over time. Flow velocity and depth typically increase sharply causing a much larger impact on structures compared to changes that occur at a constant rate. On these graphs, momentum flux values were also plotted.

16

Because momentum flux is theoretically proportional to drag force (Yeh et al., 2006), the time history will also show the changes of forces induced by the tsunami. These results will be discussed to judge if momentum flux is an appropriate approach for determining tsunami forces. Also, the equation for the momentum flux envelope presented by Yeh et al. (2006) will be compared to to see how well the envelope covers the load cases from real-life recorded events.

17

3 Analysis

Time history analysis includes data on depth and velocity of the tsunami versus time for a certain location.

In order for a video to provide sufficient data for a time history graph, the video must have a central area of interest, which is in clear view for the majority of the video. The video also has to be long enough to capture the different stages and cycles of a tsunami wave in order to compare and discuss the behavior of the tsunami. Time history graphs were produced for three locations where videos were available that met these requirements: Kamaishi Bay in

Iwate Prefecture, Ofunato Bay in Iwate Prefecture, and Haragama Soma in Fukushima Prefecture.

Many other videos do not meet the criteria needed to determine the time history. This is understandable due to the state of emergency under which the videos were recorded. For these cases, discrete events were recorded to demonstrate the method of this analysis. These results can be used as a comparison to other studies and tsunami models. The sites reviewed in this project can be seen in Figure 3-1.

Onagawa

Figure 3-1: Locations of sites analyzed in this project

18

3.1 Time History Analysis

3.1.1 Kamaishi, Iwate Prefecture

The first-person video used for analysis of flow characteristics at a site in Kamaishi has the following properties:

Reference Locations: Rikuchu-Kaigan Grand Hotel, Ito En Kamaishi Branch office building

Coordinates: 39°16’07”N; 141°53’06”E

From Japan Meteorological Agency (JMA) report (March 13, 2011):

Tsunami Arrival Time: 14:45 JST

Maximum Tsunami Depth/Time: 4.1 m or more/ 15:21 JST

From Mori and Takahashi (2012):

Maximum Run-up Elevation: 8.64 m

Figure 3.1.1-1: Google Earth image of Kamaishi port dated 4/26/2005

The video footage of the tsunami at Kamaishi Bay is credited to an evacuee who was located on top of a building off National Highway No. 45 and adjacent to the Kashigawa Estuary, seen in Figures 3.1.1-2 and -3. The

19

video contains over one hour of footage capturing the initial inflow, the following drawdown, and also the second inflow. The video starts off with the cameraman observing the rising sea level. Approximately 18 minutes into the video (Frame 32670) and half a kilometer away, the sea level had reached the top of the pier. This point in the video will be referred to as the start of the tsunami. The Japan Meteorological Agency reported that the initial tsunami began at 14:45 JST.

Figure 3.1.1-2: Google Earth image of Kamaishi port before the tsunami, dated 4/26/2005

20

Figure 3.1.1-3: Google Earth image of Kamaishi port after the tsunami, dated 3/31/2011

At about 6.5 minutes from the start of the tsunami (Frame 44099), the tsunami began flowing into the city.

The first measurement recorded was for a silver hatchback vehicle flowing along the nearby highway bridge. The footage captured the flow carrying the car a distance of 19.9 meters at a speed of approximately 2.49 m/s, shown in

Figure 3.1.1-4. By observation, the depth was estimated to be roughly 0.5 meters. The location of this reading was further away from the main foreground of the video, so the depth was not included in the overall time history.

However, this demonstrates the range of results possible at the same site.

21

Debris #1 – Silver Car

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 44100 44340 8.00 19.9 2.49

1 2

Figure 3.1.1-4: Debris #1

Soon after (Frame 44479), a violent surge rushed through the nearby intersection, seen in Figure 3.1.1-4.

The surge velocity was measured as it made its way across the intersection. The leading edge moved at a rate of approximately 9.56 m/s. The speeds of the floating debris were also measured to gain a more accurate reading of the average flow of water. Using the same distance markers, the speed of a floating green vehicle was calculated to be 7.43 m/s. A stream of small debris followed right behind and was calculated to be 7.87 m/s. The estimated depth for these measurements was 0.5 meters.

22

1 – 44570F Surge #1 Frame Start Frame End Time (sec) Distance (m) Velocity (m/s) 44570 44640 2.33 22.3 9.56

2 – 44640F

1

22.3 m 2

Figure 3.1.1-5: Surge #1 (Kamaishi)

Debris #2 following Surge #1

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 44640 44730 3.00 22.3 7.43

1

22.3 m

2

23

Figure 3.1.1-6: Debris #2 (Kamaishi)

Debris #3 Frame Distance Velocity Start Frame End Time (sec) (m) (m/s) 44655 44740 2.83 22.3 7.87

1

22.3 m

2

Figure 3.1.1-7: Debris #3 (Kamaishi)

The depth of the tsunami quickly rose to 3 meters. Debris #4 was measured over a distance of 50 meters at a speed of 3.88 m/s. This result shows that the speed of the flow has drastically decreased, most likely due to the flow spreading out because many of the structures were uprooted.

24

Debris #4

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 45440 45827 12.90 50 3.88

1

2

Figure 3.1.1-8: Debris #4 (Kamaishi)

At about 8.4 minutes from the start of the tsunami, the depth reached the bottom of the green sign, about 4 meters. This was the max depth observed in the video. This result matches the JMA report, which estimated the maximum water elevation at 4.1 m. From Mori et al., the maximum depth for this site, based on the survey point nearest to this site, was 8.64 meters (2012) with respect to the mean sea level. Google Earth shows that the elevation of this site ranges from 3 to 5 meters. Therefore, the estimation of the depth through video observation appears valid.

Debris #5, #6, and #7 share similar distance markers along the green sign. Their speed measurements were

3.74, 2.95, and 1.73 m/s, respectively. These results show the typical behavior for a tsunami approaching maximum inundation. Drawdown started at approximately 10.2 minutes from the start of the tsunami.

25

Debris #5

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 47718 47836 3.93 14.7 3.74

1 2

Figure 3.1.1-9: Debris #5 (Kamaishi)

Debris #6

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 48670 48869 6.63 19.6 2.95

1 2

Figure 3.1.1-10: Debris #6 (Kamaishi)

26

Debris #7

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 49967 50048 2.70 4.68 1.73

1 2

Figure 3.1.1-11: Debris #7 (Kamaishi)

Drawdown ran for approximately 25.1 minutes as the depth of the tsunami in the main foreground of the video approached zero. The drawdown period did not prove to be the critical loading because many of the structures had failed during the initial inflow. The second cycle started with the sea level near the top of the pier once again.

The secondary inflow continued for about 6.28 minutes and reached a maximum depth of about 2 meters. The video ended during the second drawdown period.

27

Debris #8 Frame Frame Distance Velocity Start End Time (sec) (m) (m/s) 53030 53432 13.40 18.6 1.39

2 1

Figure 3.1.1-11: Debris #8 (Kamaishi)

Debris #9 Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 55235 55415 6.00 10.7 1.78

2 1

28

Figure 3.1.1-12: Debris #9 (Kamaishi)

Debris #10 Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 102807 102910 3.43 10.7 3.12

1 2

Figure 3.1.1-13: Debris #10 (Kamaishi)

29

Kamaishi First Wave Time History

12 60

Depth (m)

)

10 Velocity (m/s) 50 2 /s

Froude Number 3 (m

8 Momentum Flux 40 2

6 30

4 20

2 10

0 0

-2 -10 Specific Momentum Flux, hv Flux, Momentum Specific

-4 -20 Depth (m) or Velocity (m/s) or Froude Number Froude or (m/s) or Velocity (m) Depth 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 Time (minutes)

Figure 3.1.1-14: Time history diagram for tsunami at Kamaishi (First Wave)

Kamaishi Second Wave Time History

12 60

Depth (m)

10 Velocity (m/s) 50

) 2

Froude Number /s 3 Momentum Flux

8 40 (m

2

6 30

4 20

2 10

0 0

-2 -10 hv Flux, Momentum Specific Depth (m) or Velocity (m/s) or Froude Number orFroude (m/s) orVelocity (m) Depth -4 -20 35.0 36.0 37.0 38.0 39.0 40.0 41.0 42.0 43.0 44.0 45.0 Time (minutes)

Figure 3.1.1-15: Time history diagram for tsunami at Kamaishi (Second Wave)

30

The time history in Figure 3.1.1-14 and 3.1.1-15 were plotted using the results from the video analysis.

Overall, the results match the typical behavior of a tsunami. The tsunami started at 18 minutes from the start of the video. Max flow velocity occurred 7 minutes after the tsunami breached the port at a speed of 9.56 m/s. This maximum is a result of the neighboring buildings causing a focused flow, resulting in high velocity flow. The flow velocity quickly dropped to 3.88 m/s and decreased steadily as the tsunami reached its maximum inundation depth.

Inflow lasted 10.2 minutes reaching a maximum inundation depth of 4 meters. This was followed by the drawdown period, which lasted 25.2 minutes. The maximum drawdown flow speed observed was 1.78 m/s.

Secondary inflow reached a maximum depth of 2 meters with an observed maximum flow velocity of 3.12 m/s.

Table 3-1 lists the time history values for the study location in Kamaishi. The maximum momentum flux

3 2 3 was about 56 m /s . Assuming a drag coefficient of CD = 2.0 and seawater density of 1024 kg/m , the equivalent average pressure on the front face of a structure would be (enter value).

Table 3-1: Summary of results for tsunami at Kamaishi Port

Average Fr E Time Depth Velocity Momentum Flux Pressure min m m/s m3/s2 kN/m2 (k/ft2) ft 0.0 0 0 0 0 Start of Tsunami 6.6 0 0 0 0 Start of Surge 6.6 0.5 9.56 45.70 91.4 (6.26) 4.317 5.160 Surge #1 6.7 0.5 7.43 27.60 55.2 (3.78) 3.355 3.315 Debris #2 6.7 0.5 7.87 30.97 62.0 (4.24) 3.553 3.658 Debris #3 6.7 1 1

6.8 2 2

7.1 3 3.88 45.16 15.4 (1.05) 0.715 3.768 Debris #4 8.4 4 3.74 55.95 14.0 (0.96) 0.597 4.713 Debris #5 8.9 4 2.95 34.81 8.7 (0.595) 0.471 4.444 Debris #6 9.6 4 1.73 11.97 3.0 (0.205) 0.276 4.153 Debris #7 10.2 4 4 Drawdown starts

11.5 2 -1.39 3.86 1.93 (0.132) -0.314 2.099 Debris #8 12.6 0.5 -1.78 1.58 3.16 (0.216) -0.804 0.662

35.3 0 0 Drawdown stops

35.6 0 0 0.00 0 Inflow starts 39.0 1 3.12 9.73 9.73 (0.67) 0.996 1.496

40.5 2 2

41.6 2 0 0 2 Drawdown starts 42.2 1 1

43.0 0.5 0.5 End of Video

31

ASCE-SEI also performed a similar video analysis for the tsunami at Kamaishi. The video captured the initial tsunami wave rushing through a small parking lot located approximately 0.75 km away from the Rikuchu-

Kaigan hotel. The site was densely surrounded by commercial buildings and was located 2 blocks from the pier edge. The results from the ASCE-SEI tsunami report for Kamaishi fall within the range of velocities for this analysis. Estimated velocities were recorded for a leading edge of a surge and a piece of debris at 3.75 m/s and 5.17 m/s, respectively. The flow velocity may not have reached the maximum of 9.56 m/s from the other analysis, but it did sustain its velocity and depth for a much longer time. For the site in this video analysis, behind the Rikuchu-

Kaigan hotel, tsunami velocity started at the maximum and decreased steadily and quickly due to the wide roads and the proximity to the river. Large open areas allowed the flow to spread out. These geographical details may easily attribute to the difference in results. It is important that the designer take these factors into consideration. Sites where roads and alleys are narrow and buildings are closely set, like the site referenced in the ASCE-SEI report, can be subjected to prolonged periods of loading.

Momentum flux reached a maximum of approximately 56.0 m3/s2 while the tsunami was at its maximum inundation depth. The maximum momentum flux occurred very early on, reflecting the explosive arrival of the tsunami. Based on this assessment of momentum flux, structures at Kamaishi experienced the greatest tsunami forces at the onset. This can be confirmed by video observation. Many of the buildings shown in the video are uplifted and destroyed at the start of the first inflow and few are observed to fail during the latter half of the first inflow and the subsequent drawdown and inflow phases.

The momentum flux envelope for this site was calculated using the equations presented by Yeh et al.

(2006) and the technical notes from FEMA P-646. The max momentum flux for this site, based on maximum run- up distance (Eqn 3), is 95.0 m3/s2. Based on maximum run-up elevation (Equation 4), the maximum momentum flux is 63.2 m3/s2. The second equation, based on run-up elevation, provides a fairly acceptable envelope, while the first equation yields a momentum flux far higher than the results from the video analysis. The disparity in the results for maximum momentum flux shows that the geography of this site may be too complex to be generalized by a momentum flux envelope. The equations for the envelope are based on parameters of a cross section and assume a uniform slope with zero friction. Additionally, the equations do not consider the factors presented in the third dimension and parameters out of plane may be controlling the results.

32

3.1.2 Ofunato, Iwate Prefecture

Reference Video: Ofunato Port

Reference Locations: Ofunato Preschool

Coordinates: 39°04’00”N 141°43’15”E

From Japan Meteorological Agency (JMA) report (March 13, 2011):

Tsunami Arrival Time: 14:46 JST

Maximum Tsunami Depth/Time: 3.2 m or more/ 15:15 JST

From Mori and Takahashi Survey:

Maximum Run-up Elevation: 9.4 m

Figure 3.1.2-1: Google Earth satellite image of Ofunato Bay

The cameraman captured this footage from an elevated road near a preschool, seen in Figure 3.1.2-2 and -3.

The location is about 5.5 kilometers from the seawall. The seawall did not survive the tsunami loads as seen in

Figure 3.1.2-3. This site consists of mainly residential properties and small commercial buildings. The main area of observation is the foreground of the camera view at the street adjacent to the storefronts. JMA reported that the

33

initial tsunami began at 14:42 JST. At about 1.25 minutes into the video (Frame 2230), the sea level reached the top of the pier. This time reference was used to match the data given by the JMA for the start of tsunami activity. A concrete building that survived the tsunami was referenced to approximate the depth of the water, shown in Figure

3.1.2-11. Each story was estimated to be about 3 meters high.

Figure 3.1.2-2: Google Earth image of Ofunato Port on July 22, 2010.

Figure 3.1.2-3: Google Earth image of Ofunato Port on February 21, 2012.

34

Figure 3.1.2-4: Satellite image of seawall at Ofunato port before (7/22/2010) and after (2/21/2012)

The first debris velocity logged was a white chest freezer, seen in Figure 3.1.2-5, at about 3.25 minutes from the time of the initial tsunami. Its flow path was obstructed from the camera view by two houses, but the velocity was still able to be determined from the short time that it was in view. The chest was captured travelling

35

10.32 meters at an average velocity of 3.44 m/s. The corresponding depth of the water was about 0.25 meters. The second debris logged was a small white shack seen in Figure 3.1.2-6. The speed and approximate depth measured were 3.64 m/s and 2 meters, respectively. The third debris velocity logged was a pile of driftwood flowing along the same path. The resulting speed and depth were 3.92 m/s and 2.5 meters, respectively. These results show the tsunami increasing in depth and speed. Indicating that the tsunami is early in its inflow phase.

Debris #1 – 8120F-8210F

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 8120 8210 3.00 10.32 3.44

2 1

Figure 3.1.2-5: Debris #1 (Ofunato)

36

Debris #2 – 9510F to 9595F

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 9510 9595 2.83 10.32 3.64

2 1

Figure 3.1.2-6: Debris #2 (Ofunato)

Frame 9658 Debris #3 – 9658F-9737F

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 9658 9737 2.63 10.32 3.92

Frame 9737 2 1

Figure 3.1.2-7: Debris #3 (Ofunato)

37

The fourth debris was logged at about 4.75 minutes from the start of the tsunami. The depth of the water increased to about 3 meters, estimated by the story level of the brown building seen in Figure 3.1.2-8 and 3.1.2-11.

For the following debris measurements, the width of the brown building was used as the reference length, which measured approximately 15 meters. The fourth debris, shown in Figure 3.1.2-8, had the max speed at 6.82 m/s. At 6 minutes from the start of the tsunami, the fifth and sixth debris were recorded. The resulting speed and height for the fifth debris was 3.97 m/s and 5 meters, respectively. The resulting speed and height for the sixth debris was 4.61 m/s and 5 meters, respectively. These results indicate that the maximum velocity of the tsunami did not occur simultaneously with the maximum depth.

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) Debris #4 – Frame 10842 to 10875 10842 10875 1.10 7.5 6.82

2 1 Width of building = 15.2 m

Figure 3.1.2-8: Debris #4 (Ofunato)

38

Frame Frame Time Distance Velocity Debris #5 – 12714F to 12829F Start End (sec) (m) (m/s) 12714 12829 3.83 15.2 3.97

2 1 Width of building = 15.2 m

Figure 3.1.2-9: Debris #5 (Ofunato)

Frame Frame Time Distance Velocity Debris #6 – 12859F to 12958F Start End (sec) (m) (m/s) 12859 12958 3.30 15.2 4.61

2 1 Width of building = 15.2 m

Figure 3.1.2-10: Debris #6 (Ofunato)

39

Before the video ended, one last debris velocity was recorded at 4.11 m/s at the maximum height of 6 meters observed in the video. The video ended about 7 minutes from the start of the tsunami and did not show the drawdown. At the end, the tsunami was observed to still be increasing in depth but decreasing in flow velocity as the tsunami approaches the point of drawdown. The tsunami reached a maximum depth of approximately 6 meters as seen in Figure 3.1.2-12. From Mori et al. (2012), the survey point nearest this site had a maximum run-up depth of 9.4 meters, or an inundation depth of about 5.4 meters. This result agrees with the result of 6 meters from the video analysis. The result exceeds the reported maximum depth from JMA by almost 2 times. The difference may possibly be attributed to different locations for measurements.

Frame Frame Time Distance Velocity Debris #7 – 14400F to 14511F Start End (sec) (m) (m/s) 14400 14511 3.70 15.2 4.11

2 1 Width of building = 15.2 m

Figure 3.1.2-11: Debris #7 (Ofunato)

40

Max depth of tsunami observed in video (le )

Structural damage on building observed from Google Earth (right)

Figure 3.1.2-12: Maximum depth of flow was estimated at 6 meters

Ofunato Time History 8 160 Depth (m) Velocity (m/s) 7 140 Froude Number

Momentum Flux )

6 120 2

/s

3 (m

5 100 2

4 80

3 60

2 40

1 20

Specific Momentum Flux, hv Flux, Momentum Specific Depth (m) or Velocity (m/s) or Froutde Number or Froutde (m/s) orVelocity (m) Depth

0 0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Time (minutes)

41

Figure 3.1.2-13: Time history of tsunami at Ofunato Port

The tsunami at Ofunato Port followed the typical sequence for a tsunami. The tsunami started at 1.25 minutes into the video. Max velocity occurred at 4.75 minutes from the start of the tsunami at a speed of 6.82 m/s.

The max depth observed occurred at the end of the video, 7 minutes from the start of the tsunami, and reached a height of approximately 6 meters. The tsunami was still increasing in depth as the video ended, so it can be concluded that the tsunami reached a slightly higher depth than was observed in the video. Design parameters for tsunami forces based on these results could be set at a design velocity of 7 m/s with a depth of 3 meters or a design depth of 6 meters with a velocity at 5 m/s. The design momentum flux (Yeh et al., 2006) would be about 150 m3/s2 and the equivalent force would result to 225 kN per meter of object breadth (15.42 kip/ft).

Table 3.1.2-1: Summary of results for tsunami at Ofunato Port

Momentum Average Frame Depth Velocity Fr E Flux Pressure m m/s m3/s2 kN/m2 (k/ft2) ft 2230 0 0 0 0 (0) 0 Start of Tsunami 8120 0.25 3.44 2.96 11.8 (0.81) 2.197 0.853 Debris #1 9500 2 3.64 26.50 13.3 (0.906) 0.822 2.676 Debris #2 9658 2 3.92 30.73 15.4 (1.05) 0.885 2.783 Debris #3 10842 3 6.82 139.54 46.7 (3.19) 1.257 5.371 Debris #4 12714 5 3.97 78.80 15.8 (1.08) 0.567 5.804 Debris #5 12829 5 4.61 106.26 21.2 (1.45) 0.658 6.084 Debris #6 14511 6 4.11 101.35 16.8 (1.15) 0.536 6.861 Debris #7 14615 6 6 End of Video

Momentum flux reached a maximum of approximately 139.5 m3/s2, coinciding with the maximum velocity.

Momentum flux increased sharply along with flow velocity, while inundation depth rose steadily. The maximum for the momentum flux envelope (Yeh et al., 2006) for this site, based on Equation 3, was 134.9 m3/s2, which is very close to the result from the video analysis. Based on Equation 4, the maximum momentum flux was 104.6. Again, there is a disparity between the two results for the envelope, suggesting that the equations do not provide a reliable estimate for the maximum momentum flux. The equations may be far too general for use with real-life topography.

42

3.1.3 Haragama Soma, Fukushima Prefecture

Reference Video: Tsunami at Haragama

Reference Locations: Haragama Port/Fishery

Coordinates: 37°49’35” N 140°58’19” E

From Japan Meteorological Agency (JMA) report (March 13, 2011):

Tsunami Arrival Time: 14:55 JST

Maximum Tsunami Depth/ Time: 7.3 m or more/ 15:50 JST

From Mori and Takashi Survey:

Maximum Run-up Elevation: 4.85 m

Figure 3.1.3-1: Google Earth Satellite Image of Haragama Port

43

The cameraman that captured this footage was able to safely evacuate to a hill near the Haragama fishery, seen in Figure 3.1.3-3 and -4. This location is unique in that it is located on the coastline and is surrounded by water on three sides. The multi-story concrete building, seen in Figure 3.1.3-2, was used as a reference for estimating the depth of the tsunami. Using the Street View function in Google Earth, the top of the parapet on the low roof was estimated to be 5 meters high. This video was chosen for a time history analysis because the surviving building provided a convenient reference to estimate and there were multiple pieces of debris for velocity measurements.

Also, the cameraman was able to record a secondary wave that arrives while the site was still inundated from the first inflow.

Figure 3.1.3-2: Post-tsunami image from Google Street View of the reference building which performed well against the wave impact

44

Figure 3.1.3-3: Google Earth image of Haragama port on September 10, 2010.

Figure 3.1.3-4: Google Earth image of Haragama port on April 5, 2011.

Tsunami run-up began at frame 10075, about 5.6 minutes from the start of the video. This time will be used to base the tsunami events to the start of the tsunami. According to the JMA, the tsunami initiated at 14:55

JST. The initial run-up velocity was calculated by tracking the leading edge traveling over a small grass lot seen in

45

Figure 3.1.3-5. The average velocity resulted to 2.7 m/s and the depth of this flow is estimated at around 0.1 meters.

Because the ocean borders this site on three sites, the tsunami quickly inundates the area with surges from multiple directions. This made it difficult to capture the flow velocity of the tsunami as the flow was constantly changing direction.

Surge #1 Frame Frame Time Distance Speed Start End (s) (m) (m/s) 11175 11900 24.167 65.3 2.702

1 2

Figure 3.1.3-5: Surge #1

At about two minutes from the start of the tsunami (Frame 13890), the depth was at 1.5 meters and the first debris was logged at a speed of 4.46 m/s seen in Figure 3.1.3-6. This result was much higher than the overall flow velocity of the tsunami because its flow was greatly affected by its proximity to the adjacent building. Therefore, this velocity should rather be used to define the drag force on the pier building. This also suggests that when calculating drag forces on a structure, it would be a good measure to increase the estimated (or design) flow velocity due to flow concentration.

46

1 Debris #1

2 1 2

Frame Frame Time Distance Speed Start End (s) (m) (m/s)

13890 13987 3.23 14.42 4.460

Figure 3.1.3-6: Debris #1 (Haragama)

At about 2.6 minutes from the start of the tsunami, an oncoming wave can be seen in the distance. As the wave breaks on the seawall, it becomes a bore and continues towards the shore. From Figure 3.1.3-7, the wave is first observed at about 400 meters from the reference building. Some drawdown activity could be noticed but the overall depth of water did not decrease. The sea wall was used as the start point in order to track the speed of the oncoming wave. Using the building as the finish point, the approximate velocity of the bore was 9.12 m/s, seen in

Figure 3.1.3-8. The wave increased the depth of the tsunami from 3.5 meters to approximately 6 meters. The wave demolished most of the structures on the pier, but surprisingly left no visible damage on the reference building, seen in Figure 3.1.3-2.

47

Loca on of Seawall Frame 8130

Frame 15124

Figure 3.1.3-7: Haragama seawall location; the seawall location is not as apparent in Frame 15124 but by comparing it with an earlier frame, the seawall location can be established with some certainty

Frame Frame Time Distance Speed Start End (s) (m) (m/s)

15560 16810 41.67 380 9.120 Wave Impact

Figure 3.1.3-8: Wave impact and distance travelled by wave After the wave impact, the overall tsunami flow redirected to the same direction as the wave. Debris #2 was

48

logged at 2.78 m/s, seen in Figure 3.1.3-9. Debris #3 was logged with a speed much higher at 4.78 m/s, seen in

Figure 3.1.3-10. Debris #3 is a more accurate representation of the flow speed than debris #2 because, as a boat, it most likely did not have as much flow resistance.

Debris #2

Frame Frame Time Distance Speed Start End (s) (m) (m/s) 17270 17684 13.8 38.3 2.775

1

2

Figure 3.1.3-9: Debris #2

49

Debris #3 - Boat

Frame Frame Time Distance Speed Start End (s) (m) (m/s)

18750 19545 26.5 126.7 4.781

1

2

Figure 3.1.3-9: Debris #3 (boat) Haragama Time History

7 160

Depth (m) Velocity m/s

6 140

Froude Number ) 2

Momentum Flux 120 /s

5 3

(m

2 100 4 80 3 60

2 40

1 20

Specific Momentum Flux, hv Flux, Momentum Specific Depth (m) or Velocity (m/s) or Froude Number orFroude (m/s) orVelocity (m) Depth 0 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Time (Minutes)

Figure 3.1.3-10: Time history of tsunami at Haragama Soma

50

The video ended before drawdown occurred so the maximum depth was not observed in the video. The

time history of the tsunami at Haragama Soma is shown in Figure 3.1.3-10. This site differs from the previous sites

because the results were greatly affected by the secondary wave. The maximum depth observed in the video

reached approximately 6 m, compared to the maximum depth reported by JMA of 7.3 m. From the survey by Mori

et al. (2006), the maximum run-up elevation reported for the nearest survey point was 4.85 meters. The maximum

tsunami velocity observed was approximately 4.78 m/s. Other survey points near the site, reported maximum

inundation depths of 7.39 and 7.57 meters, which is closer to the result from the video analysis. The maximum flow

velocity observed in the video was 4.78 m/s which occurred at the end of video.

The tsunami is predicted to further increase in depth as the tsunami approaches maximum inundation.

Tsunami velocity is expected to increase further for a short time and then decrease and reverse as drawdown occurs.

Based on these results, design depth and velocity should be set at 7 m and 5 m/s, respectively. The design

momentum flux (Yeh et al., 2006) would be about 175 m3/s2 and the equivalent force would result to 262.5 kN per

meter of object breadth (17.99 kip/ft). The momentum flux envelope for this site could not be determined because

this site was located on a peninsula. The run-up distance and run-up elevation could not be narrowed down and the

topography of this site did not suggest a clear path for the flow of the tsunami.

Table 3.1.3-1: Summary of results for tsunami at Haragama Soma

Frame Depth Velocity Momentum Flux Pressure Fr E

m m/s m3/s2 kN/m (lb/ft) ft 11175 0.1 2.7 0.73 0.73 (0.050) 2.726 0.472 Start of Tsunami 13890 1.5 4.46 29.84 29.8 (2.04) 1.163 2.514 Debris #1 14800 3 0.000 3.000 Drawdown

16930 3.5 0.000 3.500 Wave impact

17270 6 2.775 46.20 46.2 (3.17) 0.362 6.393 Debris #2 18750 6 4.775 136.80 136.8 (9.38) 0.622 7.163 Debris Boat

51

3.1.4 Onagawa, Miyagi Prefecture

Reference Video: Onagawa Inflow and Outflow

Reference Locations: Marinparu Onagawa (See Figure xx)

Coordinates: 38°26’32” N 141°26’50” E

From Japan Meteorological Agency (JMA) report (March 13, 2011):

Tsunami Arrival Time: N/A

Maximum Tsunami Depth/ Time: N/A

From Mori and Takashi Survey:

Maximum Run-up Elevation: 15.57 m

Figure 3.1.4-1: Google Earth satellite image of Onagawa Port

52

The cameraman that captured the footage for the tsunami at Onagawa was located at a shopping plaza adjacent to the Onagawa bay. The shopping plaza was known as Marinparu Onagawa, but has since been demolished due to heavy amount of damage inflicted by the tsunami. The reference location can be seen in Figure

3.1.4-3.

Figure 3.1.4-2: Pre- and post-tsunami satellite images of Onagawa

Figure 3.1.4-3: Pre-tsunami photo of the reference location, Marinparu Onagawa (from Google Earth)

53

The video used for this analysis at Onagawa was not ideal for video analysis for two reasons. First, the video did not have continuous footage of the tsunami event. This causes difficulty in constructing a time history because the amount of time that the camera is not recording is unknown. The rate of change in flow or depth cannot be presented, which is one of the few advantages of video analysis. Second, the cameraman tried to capture as much content as he/she could by panning over many locations instead of focusing on a single area. A comprehensive time history cannot be constructed if the data collected is from different locations with different elevations and local geography.

The initial surge started at about 11 seconds into the video. The tsunami can be seen flowing across the parking lot adjacent to the port. The flow continued through the shopping plaza and reached the intersection on the other side of the shopping plaza at about 33 seconds. This resulted in an average velocity of 2.92 m/s. It should be noted that this result overestimates the surge flow because there were two gaps in the recording. During this flow period, the depth of the tsunami can be estimated at 3.35 meters above mean sea level. This result for depth is based on the average ground elevation obtained from Google Earth.

Surge #1

Google Earth image showing distance traveled as 67 meters. Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 308 995 22.9 67 2.92

Video frame rate: 30 frames / second

Figure 3.1.4-4: Onagawa Surge #1

54

For Debris A and B, seen in Figure 3.1.4-5 and -6, debris flow was measured by using a nearby building as a distance marker. The debris was determined to be moving at about 3.52 and 3.91 m/s, respectively. These results reflect expected tsunami behavior during the initial stages of inflow. The depth of the flow during this period was estimated at 7.88 meters above mean sea level. Debris A

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 2456 2505 1.63 5.75 3.52

Figure 3.1.4-5: Onagawa Debris A Debris B

11.5 m

11.5 m

Google Earth image showing distance traveled as half of 11.5 meters.

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 2531 2575 1.47 5.75 3.91 Video frame rate: 30 frames / second

Figure 3.1.4-6: Onagawa Debris B

55

Debris No. 1 through 4, seen in Figures 3.1.4-7 and -8, were measured during the outflow of the tsunami as a heavy stream of debris flowed through the shopping plaza. From Figures 3.1.4-9, the results for velocity ranged from 8.19 m/s to 7.43 m/s, showing an extreme velocity for drawdown. The major factor contributing to this high flow velocity is the position of the two buildings causing the flow of water to converge through the plaza. This result highlights the importance of breakaway walls.

Figure 3.1.4-7: Onagawa Debris No.1 and No.2

Figure 3.1.4-8: Debris No. 3 and No. 4

56

Figure 3.1.4-9: Results for Debris No. 1 through No. 4

The tsunami at Onagawa was selected for a time history analysis because the same video was analyzed by researchers from Japan. In a report coordinated by Mori and Takahashi, the research group produced a time history for the tsunami at Onagawa, shown in Figure 3.1.4-10. By performing an independent analysis, comparisons could be made between the results and the procedure could be tested to see whether it could produce consistent results.

The results between the two independent analyses showed only a slight correlation. The results for depth between the two analyses showed a similar shape but the magnitudes seemed to have been based off different values for mean sea level. This could be attributed to Google Earth providing only estimated ground elevation. Velocity results during inflow had a high variance, but overall, velocity fell in the same range of 6 m/s during inflow to 8 m/s during outflow.

Although the results between the two analyses did not show a strong correlation, the comparison shows that: (1) results for depth and velocity can vary greatly during the actual tsunami event, caused by local factors

(geographic, building locations, etc.), (2) excessive camera movement and gaps in recording can cause difficulty in obtaining accurate results. It is very important that videos meet a certain criteria in order to execute a video analysis for time history data. Otherwise, analysis should be limited to obtaining discrete results.

57

Figure 3.1.4-10: Time history of the tsunami at Onagawa by Japanese researchers

Figure 3.1.4-11: Time history of the tsunami at Onagawa from video analysis

58

Table 3.1.4-1 Summary of Results for Onagawa

Time in Elevation Interpolated Momentum Froude Frame End Video above MSL Elevation Velocity Flux Number min m m m/s m3/s2 0 0 0 0 0.00 218 0.12 3.35 308 0.17 0.144 2.93 1.24 2.464 980 0.54 4.57 995 0.55 1.362 2.93 11.69 0.802 1033 0.57 5.07 1595 0.89 5.57 1777 0.99 7.57 1870 1.04 7.57 2456 1.36 4.634 3.52 57.42 0.522 2531 1.41 4.687 3.92 72.03 0.578 4091 2.27 9.14 4616 2.56 7.57 4743 2.64 4.22 -8.18 282.37 -1.271 4808 2.67 4.22 -8 270.08 -1.243 4879 2.71 4.22 -7.42 232.34 -1.153 4968 2.76 4.22 -7.5 237.38 -1.166 5043 2.80 7.57 5388 2.99 7.57

The tsunami at Onagawa had very high magnitudes for momentum flux, most likely attributed to the heavy concentrated outflow through the Onagawa shopping plaza. This also caused the outflow velocity to be a lot greater than the inflow velocity. Depth measurements were taken across several locations, accounting for the local elevation at each point. Although the results may not be highly accurate, they still reflect that Onagawa was one of the hardest-hit cities of the 2011 Tohoku Tsunami. Most of the buildings had to be demolished due to the extreme structural damage caused by the tsunami, as seen in Figure 3.1.4-2.

59

3.1.5 Summary on Momentum Flux Envelopes

The equations presented by Yeh et al. (2006) for the envelope momentum flux provide an easy and convenient procedure for determining design forces for structures in tsunami zones with only a few required variables. However, the results from this project do not indicate a strong correlation between the momentum flux envelope and real tsunami behavior. The differences stem from the generalization of the actual geographical terrain.

This can be seen in Figure 3.1.4-1 and -2, which show the elevation profile of the site at Kamaishi and the site at Ofunato, respectively. Three profiles are shown for each site. The first profile represents the cross-section intersecting the site of analysis and intersecting the point of max run-up presented by Mori et al. (2012). The other two profiles represent cross sections positioned 100 meters offset parallel to the initial cross-section. It is evident that there are many differences between the cross sections and that these cross-sections do not conform to model elevation profiles used by many tsunami researchers. Therefore, the results and these elevation profiles suggest that the estimation of momentum flux by these equations do not provide valid design criteria for tsunami-prone structures. A procedure that is capable of quantifying topographical features is needed in order to modify Yeh’s equations and make it a valid estimate for tsunami forces.

60

Figure 3.1.5-1: Elevation profiles of Kamaishi from Google Earth

61

Figure 3.1.5-2: Elevation profiles of Ofunato from Google Earth

62

3.2 Discrete Event Analysis

3.2.1 Minamisanriku, Miyagi Prefecture

Reference Video: Shizugawa HS

Reference Locations: Shizugawa High School

Coordinates: 38°40’51” N 141°26’25” E

From Japan Meteorological Agency (JMA) report (March 13, 2011):

Tsunami Arrival Time: N/A

Maximum Tsunami Depth/ Time: N/A

63

Figure 3.2.1-1: Satellite image of Minamisanriku on June 24, 2010

Figure 3.2.1-2: Satellite image of Minamisanriku on April 5, 2011

64

The cameraman for this footage was located on a road leading up to Shizugawa High School. The area of the footage was located about 750 meters from the ocean so the initial start time of the tsunami could not be determined. The high school is located on top of a very high hill and served as an evacuation center for the town.

The tsunami came into view as it crossed the elevated railroad track. The farm crop lines were used to gauge the distance as the tsunami made its way across the valley.

Two measurements were taken of the leading edge of the surge. The tsunami flowed over the elevated railroad track and traveled 76.5 meters at a speed of 4.41 m/s. The surge then traveled another 47.5 meters at a speed of 2.97 m/s. This result confirms that the surge slowed as it made its way across the field.

Figure 3.2.1-3: Surge #1

65

Figure 3.2.1-4: Surge #2

Debris velocity was difficult to measure in the foreground because the tsunami uprooted most of the structures, leaving few reference landmarks. However in the background, a large bunch of driftwood flowed north behind the railroad track for a distance of 45 meters at a speed of 11.25 m/s. This location was approximately 700 meters inland.

66

Figure 3.2.1-5: Debris #1

Because the location of the footage was in a small valley, the tsunami slowed and continued to increase in depth. The tsunami filled the valley up to at least a depth of 3 meters, estimated by the observation that the uprooted houses were floating with their first story below water. This would be the maximum depth observed in this video as the video ended shortly after.

The viewpoint of the cameraman did not provide a good point of observation for estimating momentum flux. The cameraman was safely located inland, at a high elevation, and far from where the major destruction was occurring. The velocity measurements for the surge travelling across the farmland had a very low flow depth, so the resultant momentum flux is relatively insignificant. For the measurement of the flow that occurred behind the railroad track, the momentum flux can be estimated at approximately 380 m3/s2. This is the highest result for momentum flux obtained for this project. This result may be very inaccurate because the measurement was recorded for debris flowing far in the background. Nevertheless, Minami-sanriku was one of the most severely affected cities and the max momentum flux observed was far greater than any other result from this project.

67

3.2.2 Sendai, Miyagi Prefecture

Figure 3.2.2-1: Google Earth Image of Sendai Harbor and coastline before tsunami

Reference Video: Sendai – Mitsui Mall

Reference Locations: Mitsui Mall, Ken Taku Real Estate Office

Coordinates: 38°16’36” N 140°59’11” E

From Japan Meteorological Agency (JMA) report (March 13, 2011):

Tsunami Arrival Time: N/A

Maximum Tsunami Depth/ Time: N/A

68

Figure 3.2.2-2: Satellite image of cameraman location on April 3, 2010

Figure 3.2.2-3: Satellite image of cameraman location on April 5, 2011

69

The cameraman was located on top of a nine-story concrete building. The site is approximately located

1080 meters inland. The surge velocity was measured across 10.1 meters at a speed of 2.16 m/s. The debris logged was a white unidentified object traveling 46.3 meters at a speed of 2.13 m/s. These results were fairly low due to the site’s location. In addition to being further inland than other sites, the industrial site had many wide roads for the tsunami flow to spread around and stay flat. The video ends before the drawdown period and the maximum tsunami height observed was less than 0.5 meters. The maximum momentum flux observed was approximately 2.33 m3/s2.

This is the lowest result for momentum flux obtained for this project.

Figure 3.2.2-4: Surge #1 (Mitsui Mall)

70

Figure 3.2.2-5: Debris #1 (Mitsui Mall)

71

3.2.3 Shiogama, Miyagi Prefecture

Figure 3.2.3-1: Google Earth image of Shiogama showing camera location

Reference Video: Shiogama

Reference Locations: Shiogama mall

Coordinates: 38°19’07”N, 141°01’24”E

From Japan Meteorological Agency (JMA) report (March 13, 2011):

Tsunami Arrival Time: N/A

Maximum Tsunami Depth/ Time: N/A

72

Figure 3.2.3-2: Satellite image of cameraman location on March 30, 2009

Figure 3.2.3-3: Satellite image of cameraman location on April 5, 2011

73

The cameraman who shot this video was located on top of a 12-story apartment building adjacent to the port. The video begins just as the tsunami enters the city. In Figure 3.2.3-4, the tsunami’s initial flow traveled at approximately 4.2 m/s over the portside park. From the park, the tsunami continued inland at a speed of 5.5 m/s, estimated by Debris #1, seen in Figure 3.2.3-5. The associated flow depth could be estimated at no more than 0.5 meters, based on the story height of the adjacent buildings. The inflow continued for most of the video, lasting about 5.0 minutes. The maximum depth that was reached was about 2 meters, estimated against the story height of the buildings in the video. The video ends during the transition phase between inflow and drawdown. The maximum observed momentum flux for this video was 15.1 m3/s2.

Frame Frame Speed (m/ Surge #1 Start End Time (s) Distance (m) s) 335 811 15.87 66.93 4.218

Figure 3.2.3-4: Surge #1 (Shiogama)

74

1 2

Debris #1

1 2

Frame Frame Time Distance Speed Start End (s) (m) (m/s) 4268 4368 3.33 18.3 5.490

Figure 3.2.3-5: Debris #1 (Shiogama)

75

3.2.4 Tagajo, Miyagi Prefecture

Figure 3.2.4-1: Google Earth image of Sendai Harbor showing camera location in Tagajo

Reference Video: Tagajo

Reference Locations: AEON shopping mall

Coordinates: 38°16’57”N, 141°00’15”E

From Japan Meteorological Agency (JMA) report (March 13, 2011):

Tsunami Arrival Time: N/A

Maximum Tsunami Depth/ Time: N/A

76

Figure 3.2.4-2: Satellite image of cameraman location on March 30, 2009

Figure 3.2.4-3: Satellite image of cameraman location on April 5, 2011

The site is located about 1 kilometer from the ocean. The reference video did not capture the time of the initial tsunami, but it did have surge velocity and debris velocity. The initial surge traveled a total of 54.5 meters at

77

an average velocity of 3.7 m/s and a max velocity of 4.5 m/s. The debris (Debris #1 and #2) that followed the leading edge traveled at a much higher velocity of 4.7 and 6.8 m/s, respectively.

Surge #1 3 18.21 m 2

1 36.24 m

1

3 2

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 848 1090 8.07 36.24 4.49 1090 1290 6.67 18.21 2.73

Figure 3.2.4-4: Surge #1 (Tagajo)

78

Debris #1

2

38.01 m

1

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 1140 1385 8.17 38.01 4.65

Figure 3.2.4-5: Debris #1 (Tagajo)

Debris #2 (debris stream)

2

38.01 m

1

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 1312 1480 5.60 38.01 6.79

Figure 3.2.4-6: Debris #2 (Tagajo)

79

By about 2 minutes into the video, the tsunami flow gathered into a uniform stream and continued northwest inland at about 5.0 m/s, measured from Debris #3, seen in Figure 3.2.4-7.

Debris #3

38.01 m 1 2

Frame Frame Time Distance Velocity Start End (sec) (m) (m/s) 3825 4035 7.00 35.26 5.04

Figure 3.2.4-7: Debris #3 (Tagajo)

The inflow continued for the rest of the video, lasting a total of 8.8 minutes for this site. At the end of the video, the tsunami approached its drawdown phase. The maximum depth remained fairly constant at about 2 meters

(door height) throughout the video, due to all the open area at this commercial area. This resulted in a maximum momentum flux of approximately 50.8 m3/s2.

80

4 Summary and Conclusion

Video analysis of tsunamis, although somewhat crude, proved sufficient enough to produce time history data for some locations. Many of footages were shaky and chaotic, but by analyzing video frame-by-frame the flow velocity and inundation depth were estimated with just a few seconds of video. The resulting time history graphs followed the expected behavior for tsunamis. Although flow velocity and depth differ for each structure, location, and tsunami magnitude, conservative design loads can be estimated through video analysis. In addition, results from video analysis may be valuable as a check for computer models for tsunamis.

Table 4-1: Summary of results

Site Maximum Inundation Depth Maximum Flow Velocity Maximum Momentum Flux

(m) (m/s) (m3/s2)

Kamaishi ~ 4 3.88 56.0

Ofunato > 6 6.82 139.5

Haragama Soma > 6 > 4.78 56.0

Minami-sanriku ~ 3 ~ 11.5 379.7

Sendai < 0.5 2.16 2.33

Shiogama < 0.5 5.5 15.1

Tagajo 2 6.8 50.8

From this project, the factors that affected the magnitude of tsunami forces were inundation depth and flow velocity. The parameters that affect the maximum inundation depth were the elevation of site, the proximity to the coast, and the relative slope. Maximum flow velocity is dependent on the same parameters, but is also highly dependent on the inundation depth. Another factor that was observed to affect flow velocity was the type of area

(commercial, industrial, farmland, etc.). Locations that were more densely populated resulted with higher flow velocities. Results for momentum flux were shown to be highly dependent on the flow velocity of the tsunami. By determining the momentum flux, the maximum tsunami design forces could be obtained. However, the envelope did not show a strong correlation to the results from the video analysis. Until a better procedure is found, the author

81

suggests using historical data to determine design forces from tsunamis. This procedure would be far more prudent.

With conservative engineering judgment, the designer would produce a more structurally sound design.

82

References

Department of Planning and Permitting of Honolulu Hawai'i. 2012. City and County of Honolulu Building Code.

Chapter 16 Article 11.

FEMA P-55 (2011). Coastal Construction Manual. Prepared by U.S. Department of Homeland Security.

Fritz, H. M., J. C. Borrero, C.E. Synolakis, and J. Yoo. 2006. 2004 Indian Ocean tsunami flow velocity

measurements from survivor videos. Geophysical Research Letters, 33, L24605,

doi:10.1029/2006GL026784.

Fritz, H., D. Phillips, A. Okayasu, T. Shimozono, H. Liu, F. Mohammed, V. Skanavis, C. Synolakis, T. Takahashi.

2012. The 2011 Japan tsunami current velocity measurements from survivor videos at Kesennuma Bay

using LIDAR. Geophysical Research Letters, 39, L00G23, DOI: 10.1029/2011GL050686.

Mori, N., T. Takahashi. 2012. Nationwide Post Event Survey and Analysis of the 2011 Tohoku Earthquake

Tsunami. Coastal Engineering Journal, Vol. 54, No. 1, 1250001, DOI: 10.1142/s0578563412500015.

Yeh, H., I. Robertson, and J. Preuss. 2005. Development of Design Guidelines for Structures that Serve as

Tsunami Vertical Evacuation Sites. Prepared for Washington Department of Natural Resources.

Yeh, H. 2006. Maximum fluid forces in the tsunami runup zone. Journal of Waterway, Port, Coastal, and

Ocean Engineering, November-December, 496-500.

83

APPENDIX

A. Report from the Japan Meteorological Agency (JMA)

Copied and Pasted from: http://www.jma.go.jp/en/tsunami/observation_04_20110313180559.html on May 26, 2012.

Occurred at 14:46 JST 11 Mar 2011 Region name Sanriku Oki Depth about 20 km Magnitude 9.0

Initial Tsunami Observation Click the map to zoom in

84

Maximum Tsunami Observation

85

Tsunami Information NUMBER 64 (Tsunami Observations)

Issued at 18:05 JST 13 Mar 2011

Tsunami Observations As of 18:00 JST In some coastal regions, it is inferred that tsunamis higher than those recorded at observation sites have arrived.

Kushiro Initial Tsunami 15:34 JST 11 Mar (+) 2.0 m Maximum Tsunami 23:39 JST 11 Mar 2.1 m

Nemuro-shi Hanasaki Initial Tsunami 15:34 JST 11 Mar (-) Slight Maximum Tsunami 15:57 JST 11 Mar 2.8 m

Nemuro-ko Initial Tsunami 16:08 JST 11 Mar (+) 0.3 m Maximum Tsunami 00:03 JST 12 Mar 0.7 m

Hamanaka-cho Kiritappu-ko Initial Tsunami 15:27 JST 11 Mar (-) Slight Maximum Tsunami 22:19 JST 11 Mar 2.6 m

Urakawa Initial Tsunami 15:19 JST 11 Mar (-) 0.2 m Maximum Tsunami 16:42 JST 11 Mar 2.7 m

Tokachi-ko Initial Tsunami 15:26 JST 11 Mar (-) 0.2 m Maximum Tsunami 15:57 JST 11 Mar 2.8 m or more

Erimo-cho Shoya Initial Tsunami 15:18 JST 11 Mar (-) 0.1 m

86

Maximum Tsunami 15:44 JST 11 Mar 3.5 m

Hakodate Initial Tsunami 16:11 JST 11 Mar (+) 1.9 m Maximum Tsunami 23:35 JST 11 Mar 2.4 m

Tomakomai-nishiko Initial Tsunami 15:37 JST 11 Mar (-) 0.2 m Maximum Tsunami 17:30 JST 11 Mar 2.1 m

Tomakomai-higashiko Initial Tsunami 15:40 JST 11 Mar (-) 0.2 m Maximum Tsunami 16:17 JST 11 Mar 2.5 m or more

Shiraoi-ko Initial Tsunami 15:40 JST 11 Mar (-) 0.1 m Maximum Tsunami 16:01 JST 11 Mar 1.8 m

Oshima Mori-ko Initial Tsunami 15:56 JST 11 Mar (-) Slight Maximum Tsunami 19:37 JST 11 Mar 1.8 m

Muroran-ko Initial Tsunami 15:56 JST 11 Mar (-) Slight Maximum Tsunami 20:06 JST 11 Mar 1.0 m

Wakkanai Initial Tsunami 18:43 JST 11 Mar (+) 0.1 m Maximum Tsunami 02:23 JST 12 Mar 0.4 m

Rumoi Initial Tsunami (*3)

Maximum Tsunami 05:33 JST 12 Mar 0.3 m

Otaru Initial Tsunami (*3)

Maximum Tsunami 03:25 JST 12 Mar 0.3 m

Ishikariwan-shinko Initial Tsunami (*3)

Maximum Tsunami 01:07 JST 12 Mar 0.4 m

Iwanai-ko Initial Tsunami (*3)

Maximum Tsunami 02:23 JST 12 Mar 0.3 m

Abashiri Initial Tsunami 17:02 JST 11 Mar (+) 0.1 m Maximum Tsunami 02:32 JST 12 Mar 0.3 m

Esashi-ko Initial Tsunami 17:45 JST 11 Mar (+) 0.2 m Maximum Tsunami 23:36 JST 11 Mar 0.4 m

Tappi Initial Tsunami 16:09 JST 11 Mar (+) 0.5 m Maximum Tsunami 16:32 JST 11 Mar 0.5 m

Hachinohe Initial Tsunami 15:22 JST 11 Mar (-) 0.8 m Maximum Tsunami 16:51 JST 11 Mar 2.7 m or more

Mutsu-shi Sekinehama Initial Tsunami 15:20 JST 11 Mar (-) 0.1 m Maximum Tsunami 18:16 JST 11 Mar 2.9 m

Miyako Initial Tsunami 14:48 JST 11 Mar (+) 0.2 m Maximum Tsunami 15:21 JST 11 Mar 4.0 m or more

Ofunato Initial Tsunami 14:46 JST 11 Mar (-) 0.2 m Maximum Tsunami 15:15 JST 11 Mar 3.2 m or more

Kamaishi Initial Tsunami 14:45 JST 11 Mar (-) 0.1 m Maximum Tsunami 15:21 JST 11 Mar 4.1 m or more

Iwate Kamaishi-oki* Initial Tsunami 14:50 JST 11 Mar Unknown

Iwate Miyako-oki* Initial Tsunami 14:50 JST 11 Mar Unknown

Iwate Kuji-oki* Initial Tsunami 14:56 JST 11 Mar Unknown

Ishinomaki-shi Ayukawa Initial Tsunami 14:46 JST 11 Mar (+) 0.1 m Maximum Tsunami 15:20 JST 11 Mar 3.3 m or more

87

Kesennuma Hirotawan-oki* Initial Tsunami 14:54 JST 11 Mar Unknown

Sakata Initial Tsunami (*3)

Maximum Tsunami 00:55 JST 12 Mar 0.4 m

Soma Initial Tsunami 14:55 JST 11 Mar (+) 0.3 m Maximum Tsunami 15:50 JST 11 Mar 7.3 m or more

Fukushima Onahama-oki* Initial Tsunami 14:52 JST 11 Mar Unknown

Oarai Initial Tsunami 15:15 JST 11 Mar (+) 1.8 m Maximum Tsunami 16:52 JST 11 Mar 4.2 m

Choshi Initial Tsunami 15:13 JST 11 Mar (+) 0.5 m Maximum Tsunami 17:22 JST 11 Mar 2.4 m

Tateyama-shi Mera Initial Tsunami 15:29 JST 11 Mar (+) 1.3 m Maximum Tsunami 17:05 JST 11 Mar 1.6 m

Tokyo Harumi Initial Tsunami 16:37 JST 11 Mar (+) 0.8 m Maximum Tsunami 19:15 JST 11 Mar 1.3 m

Yokosuka Initial Tsunami 15:52 JST 11 Mar (+) 0.9 m Maximum Tsunami 17:16 JST 11 Mar 1.6 m

Chiba Initial Tsunami 16:38 JST 11 Mar (+) 0.7 m Maximum Tsunami 18:18 JST 11 Mar 0.9 m

Yokohama Initial Tsunami 16:09 JST 11 Mar (+) 0.8 m Maximum Tsunami 17:37 JST 11 Mar 1.6 m

Izu-oshima Okada Initial Tsunami 14:52 JST 11 Mar (-) 0.3 m Maximum Tsunami 15:49 JST 11 Mar 0.7 m

Miyakejima Tsubota Initial Tsunami 15:26 JST 11 Mar Unknown

Maximum Tsunami 15:34 JST 11 Mar 0.8 m

Hachijojima Yaene Initial Tsunami 15:40 JST 11 Mar (+) 1.4 m Maximum Tsunami 15:48 JST 11 Mar 1.4 m

Kozushima Kozushima-ko Initial Tsunami 15:40 JST 11 Mar (+) 0.4 m Maximum Tsunami 00:30 JST 12 Mar 0.8 m

Miyakejima Ako Initial Tsunami 15:27 JST 11 Mar (+) 0.6 m Maximum Tsunami 15:39 JST 11 Mar 0.6 m

Hachijojima Kaminato Initial Tsunami 15:35 JST 11 Mar (+) 1.2 m Maximum Tsunami 15:45 JST 11 Mar 1.2 m

Chichijima Futami Initial Tsunami 16:14 JST 11 Mar (+) 1.0 m Maximum Tsunami 16:46 JST 11 Mar 1.8 m

Minami-torishima Initial Tsunami 16:51 JST 11 Mar (+) 0.5 m Maximum Tsunami 16:55 JST 11 Mar 0.5 m

Odawara Initial Tsunami 15:33 JST 11 Mar (+) 0.9 m Maximum Tsunami 15:48 JST 11 Mar 0.9 m

Niigata Initial Tsunami 18:02 JST 11 Mar (+) 0.1 m Maximum Tsunami 04:54 JST 12 Mar 0.2 m

Awashima Initial Tsunami (*3)

Maximum Tsunami 21:05 JST 11 Mar 0.1 m

Kashiwazaki-shi Kujiranami Initial Tsunami (*3)

Maximum Tsunami 22:19 JST 11 Mar 0.1 m

88

Sado-shi Washizaki Initial Tsunami (*3)

Maximum Tsunami 00:56 JST 12 Mar Slight

Toyama Initial Tsunami 20:55 JST 11 Mar (+) Slight Maximum Tsunami 22:37 JST 11 Mar 0.1 m

Fushiki Toyama-ko Shinminato Initial Tsunami 21:02 JST 11 Mar (+) 0.1 m Maximum Tsunami 22:39 JST 11 Mar 0.1 m

Suzu-shi Nagahashi Initial Tsunami (*3)

Maximum Tsunami 23:25 JST 11 Mar 0.1 m

Kanazawa Initial Tsunami 21:56 JST 11 Mar (+) 0.1 m Maximum Tsunami 12:54 JST 12 Mar 0.2 m

Tsuruga-ko Initial Tsunami (*3)

Maximum Tsunami 13:58 JST 13 Mar 0.3 m

Numazu-shi Uchiura Initial Tsunami 16:01 JST 11 Mar (+) 1.4 m Maximum Tsunami 16:16 JST 11 Mar 1.4 m

Shimizu Initial Tsunami 15:58 JST 11 Mar (+) 0.9 m Maximum Tsunami 16:17 JST 11 Mar 0.9 m

Minami-izu-cho Irozaki Initial Tsunami 15:43 JST 11 Mar (+) 0.2 m Maximum Tsunami 15:56 JST 11 Mar 0.8 m

Omaezaki Initial Tsunami 16:03 JST 11 Mar (+) 1.0 m Maximum Tsunami 17:18 JST 11 Mar 1.4 m

Maisaka Initial Tsunami 16:12 JST 11 Mar (+) 0.7 m Maximum Tsunami 17:37 JST 11 Mar 0.8 m

Shimoda-ko Initial Tsunami 15:41 JST 11 Mar (+) 0.7 m Maximum Tsunami 22:55 JST 11 Mar 0.8 m

Ito Initial Tsunami 15:30 JST 11 Mar (+) 0.1 m Maximum Tsunami 15:52 JST 11 Mar 0.7 m

Nishiizu-cho Tago Initial Tsunami 15:55 JST 11 Mar (+) 0.4 m Maximum Tsunami 16:18 JST 11 Mar 0.4 m

Yaizu Initial Tsunami 15:58 JST 11 Mar (+) 0.8 m Maximum Tsunami 16:10 JST 11 Mar 0.8 m

Tahara-shi Akabane Initial Tsunami 16:21 JST 11 Mar (+) 1.1 m Maximum Tsunami 17:32 JST 11 Mar 1.6 m

Nagoya Initial Tsunami 17:43 JST 11 Mar (+) 0.7 m Maximum Tsunami 19:36 JST 11 Mar 1.0 m

Handa-shi Kinuura Initial Tsunami 17:18 JST 11 Mar (+) 0.5 m Maximum Tsunami 21:35 JST 11 Mar 0.7 m

Yokkaichi Initial Tsunami 17:20 JST 11 Mar (+) 0.4 m Maximum Tsunami 20:13 JST 11 Mar 0.5 m

Toyohashi-shi Mikawa-ko Initial Tsunami 17:17 JST 11 Mar (+) 0.4 m Maximum Tsunami 20:15 JST 11 Mar 0.6 m

Toba Initial Tsunami 16:34 JST 11 Mar (+) 0.5 m Maximum Tsunami 19:13 JST 11 Mar 1.8 m

Owase Initial Tsunami 16:17 JST 11 Mar (+) 1.0 m Maximum Tsunami 17:12 JST 11 Mar 1.7 m

89

Kumano-shi Yuki Initial Tsunami 16:12 JST 11 Mar (+) 0.7 m Maximum Tsunami 16:28 JST 11 Mar 0.7 m

Mie Owase-oki* Initial Tsunami 16:07 JST 11 Mar Unknown

Maizuru Initial Tsunami (*3)

Maximum Tsunami 11:40 JST 13 Mar 0.3 m

Osaka Tempozan Initial Tsunami 18:20 JST 11 Mar (+) 0.5 m Maximum Tsunami 21:02 JST 11 Mar 0.6 m

Misaki-cho Tannowa Initial Tsunami 17:30 JST 11 Mar (+) 0.2 m Maximum Tsunami 17:59 JST 11 Mar 0.2 m

Toyooka-shi Tsuiyama Initial Tsunami (*3)

Maximum Tsunami 01:03 JST 12 Mar Slight

Kobe Initial Tsunami 17:56 JST 11 Mar (+) 0.3 m Maximum Tsunami 03:19 JST 12 Mar 0.3 m

Himeji Initial Tsunami 18:30 JST 11 Mar (+) 0.1 m Maximum Tsunami 20:46 JST 11 Mar 0.2 m

Sumoto Initial Tsunami 17:20 JST 11 Mar (+) 0.2 m Maximum Tsunami 17:37 JST 11 Mar 0.2 m

Nachi-katsuura-cho Uragami Initial Tsunami 16:14 JST 11 Mar (+) 0.9 m Maximum Tsunami 18:06 JST 11 Mar 1.3 m

Kushimoto-cho Fukuro-ko Initial Tsunami 16:16 JST 11 Mar (+) 0.7 m Maximum Tsunami 01:32 JST 12 Mar 1.4 m

Wakayama Initial Tsunami 17:13 JST 11 Mar (+) 0.6 m Maximum Tsunami 19:35 JST 11 Mar 0.7 m

Gobo-shi Haraido Initial Tsunami 16:35 JST 11 Mar (+) 0.7 m Maximum Tsunami 17:57 JST 11 Mar 1.1 m

Shirahama-cho Katata Initial Tsunami 16:34 JST 11 Mar (+) 0.9 m Maximum Tsunami 00:35 JST 12 Mar 1.0 m

Wakayama Shirahama-oki* Initial Tsunami 16:25 JST 11 Mar Unknown

Sakaiminato-shi Sakai Initial Tsunami (*3)

Maximum Tsunami 05:03 JST 12 Mar 0.3 m

Hamada Initial Tsunami (*3)

Maximum Tsunami 07:53 JST 12 Mar 0.1 m

Oki Saigo Initial Tsunami (*3)

Maximum Tsunami 04:06 JST 12 Mar 0.1 m

Tamano-shi Uno Initial Tsunami 18:29 JST 11 Mar (+) Slight Maximum Tsunami 20:01 JST 11 Mar 0.1 m

Hiroshima Initial Tsunami 19:28 JST 11 Mar (+) 0.3 m Maximum Tsunami 20:11 JST 11 Mar 0.3 m

Kure Initial Tsunami 19:43 JST 11 Mar (+) 0.3 m Maximum Tsunami 20:37 JST 11 Mar 0.3 m

Komatsushima Initial Tsunami 17:03 JST 11 Mar (+) 0.7 m Maximum Tsunami 19:50 JST 11 Mar 0.7 m

Tokushima Yuki Initial Tsunami 16:37 JST 11 Mar (+) 1.0 m Maximum Tsunami 20:27 JST 11 Mar 1.1 m

90

Takamatsu Initial Tsunami 18:20 JST 11 Mar (+) Slight Maximum Tsunami 19:41 JST 11 Mar 0.2 m

Sakaide-shi Yoshima-ko Initial Tsunami 19:38 JST 11 Mar (+) 0.1 m Maximum Tsunami 13:33 JST 12 Mar 0.2 m

Tadotsu-ko Initial Tsunami 19:56 JST 11 Mar (+) Slight Maximum Tsunami 11:25 JST 12 Mar 0.2 m

Uwajima Initial Tsunami 17:42 JST 11 Mar (+) 0.5 m Maximum Tsunami 07:10 JST 12 Mar 0.6 m

Matsuyama Initial Tsunami 18:43 JST 11 Mar (+) 0.2 m Maximum Tsunami 23:30 JST 11 Mar 0.2 m

Imabari-shi Oshima Initial Tsunami 19:28 JST 11 Mar (+) 0.1 m Maximum Tsunami 11:40 JST 12 Mar 0.1 m

Muroto-shi Muroto-misaki Initial Tsunami 16:39 JST 11 Mar (+) 0.4 m Maximum Tsunami 04:42 JST 12 Mar 0.6 m

Kochi Initial Tsunami 16:56 JST 11 Mar (+) 0.6 m Maximum Tsunami 21:26 JST 11 Mar 0.6 m

Tosa-shimizu Initial Tsunami 16:56 JST 11 Mar (+) 0.9 m Maximum Tsunami 01:58 JST 12 Mar 1.3 m

Susaki-ko Initial Tsunami 17:00 JST 11 Mar (+) 1.4 m Maximum Tsunami 20:59 JST 11 Mar 2.6 m

Shimonoseki-shi Haedomari-ko Initial Tsunami 20:28 JST 11 Mar (+) Slight Maximum Tsunami 03:25 JST 12 Mar 0.1 m

Shimonoseki-shi Hikoshima-deshimatsu Initial Tsunami 19:22 JST 11 Mar (+) 0.2 m Maximum Tsunami 23:09 JST 11 Mar 0.4 m

Tokuyama Initial Tsunami 18:41 JST 11 Mar (+) 0.3 m Maximum Tsunami 08:06 JST 12 Mar 0.4 m

Ube-ko Initial Tsunami 19:08 JST 11 Mar (+) 0.2 m Maximum Tsunami 09:14 JST 12 Mar 0.3 m

Mitajiri Nakanoseki-ko Initial Tsunami 18:42 JST 11 Mar (+) 0.2 m Maximum Tsunami 23:39 JST 11 Mar 0.2 m

Shimonoseki-ko Chofu Initial Tsunami 19:33 JST 11 Mar (+) 0.3 m Maximum Tsunami 23:00 JST 11 Mar 0.4 m

Kitakyushu-shi Moji Initial Tsunami 19:46 JST 11 Mar (+) 0.2 m Maximum Tsunami 23:10 JST 11 Mar 0.4 m

Kanda-ko Initial Tsunami 19:26 JST 11 Mar (+) 0.3 m Maximum Tsunami 20:16 JST 11 Mar 0.3 m

Kitakyushu-ko Aohama Initial Tsunami 19:21 JST 11 Mar (+) 0.3 m Maximum Tsunami 23:04 JST 11 Mar 0.4 m

Fukuoka-shi Hakata Initial Tsunami 21:06 JST 11 Mar (+) 0.2 m Maximum Tsunami 02:12 JST 12 Mar 0.3 m

Kitakyushu-ko Hiagari Initial Tsunami 20:06 JST 11 Mar (+) Slight Maximum Tsunami 23:10 JST 11 Mar 0.2 m

Tara-cho Ouranozaki Initial Tsunami 19:32 JST 11 Mar (+) 0.2 m Maximum Tsunami 08:00 JST 12 Mar 0.2 m

91

Misumi Initial Tsunami 19:24 JST 11 Mar (+) 0.1 m Maximum Tsunami 06:47 JST 12 Mar 0.1 m

Yatsushiro-ko Initial Tsunami 19:49 JST 11 Mar (+) 0.3 m Maximum Tsunami 20:17 JST 12 Mar 0.3 m

Amakusa-shi Hondo-ko Initial Tsunami 19:08 JST 11 Mar (+) 0.3 m Maximum Tsunami 21:05 JST 11 Mar 0.8 m

Kumamoto-ko Initial Tsunami 19:41 JST 11 Mar (+) 0.1 m Maximum Tsunami 22:30 JST 11 Mar 0.1 m

Karatsu-ko Initial Tsunami 20:43 JST 11 Mar (+) 0.1 m Maximum Tsunami 05:15 JST 12 Mar 0.2 m

Genkai-cho Kariya Initial Tsunami (*3)

Maximum Tsunami 05:31 JST 12 Mar 0.3 m

Nagasaki Initial Tsunami 18:55 JST 11 Mar (+) 0.5 m Maximum Tsunami 21:20 JST 11 Mar 0.8 m

Kuchinotsu Initial Tsunami 18:52 JST 11 Mar (+) 0.1 m Maximum Tsunami 10:07 JST 12 Mar 0.2 m

Fukuejima Fukue-ko Initial Tsunami 18:45 JST 11 Mar (+) 0.2 m Maximum Tsunami 04:58 JST 12 Mar 0.3 m

Sasebo Initial Tsunami 19:32 JST 11 Mar (+) 0.4 m Maximum Tsunami 21:52 JST 11 Mar 0.7 m

Nagasaki-ko Kougo Initial Tsunami 18:49 JST 11 Mar (+) 0.4 m Maximum Tsunami 21:22 JST 11 Mar 0.6 m

Hirado-shi Tabira-ko Initial Tsunami 19:36 JST 11 Mar (+) 0.2 m Maximum Tsunami 21:56 JST 11 Mar 0.3 m

Reihoku-machi Tororo Initial Tsunami 18:43 JST 11 Mar (+) 0.2 m Maximum Tsunami 03:12 JST 12 Mar 0.3 m

Oita Initial Tsunami 17:49 JST 11 Mar (+) 0.2 m Maximum Tsunami 22:29 JST 11 Mar 0.4 m

Beppu-ko Initial Tsunami 17:59 JST 11 Mar (+) 0.3 m Maximum Tsunami 22:17 JST 11 Mar 0.5 m

Saiki-shi Matsuura Initial Tsunami 17:22 JST 11 Mar (+) 0.4 m Maximum Tsunami 17:41 JST 11 Mar 0.4 m

Hyuga-shi Hososhima Initial Tsunami 16:54 JST 11 Mar (+) 0.8 m Maximum Tsunami 21:47 JST 11 Mar 0.9 m

Nichinan-shi Aburatsu Initial Tsunami 17:05 JST 11 Mar (+) 1.0 m Maximum Tsunami 00:12 JST 12 Mar 1.1 m

Miyazaki-ko Initial Tsunami 17:06 JST 11 Mar (+) 1.4 m Maximum Tsunami 03:33 JST 12 Mar 1.6 m

Minami-osumi-cho Odomari Initial Tsunami 17:29 JST 11 Mar (+) 0.5 m Maximum Tsunami 06:51 JST 12 Mar 0.9 m

Shibushi-ko Initial Tsunami 17:12 JST 11 Mar (+) 1.1 m Maximum Tsunami 17:39 JST 11 Mar 1.1 m

Tanegashima Nishino-omote Initial Tsunami 17:20 JST 11 Mar (+) 0.2 m Maximum Tsunami 23:45 JST 11 Mar 0.8 m

92

Tanegashima Kumano Initial Tsunami 16:52 JST 11 Mar (+) 0.8 m Maximum Tsunami 03:23 JST 12 Mar 1.5 m

Nakanoshima Initial Tsunami (*3)

Maximum Tsunami 05:29 JST 12 Mar 0.7 m

Amami-shi Kominato Initial Tsunami 17:01 JST 11 Mar (-) 0.1 m Maximum Tsunami 01:48 JST 12 Mar 1.2 m

Amami-shi Naze Initial Tsunami 17:10 JST 11 Mar (+) 0.4 m Maximum Tsunami 01:21 JST 12 Mar 0.5 m

Kagoshima Initial Tsunami 18:19 JST 11 Mar (+) 0.1 m Maximum Tsunami 04:47 JST 12 Mar 0.1 m

Makurazaki Initial Tsunami 17:55 JST 11 Mar (+) 0.3 m Maximum Tsunami 02:28 JST 12 Mar 0.8 m

Akune Initial Tsunami 18:38 JST 11 Mar (+) 0.5 m Maximum Tsunami 07:18 JST 12 Mar 0.5 m

Naha Initial Tsunami 18:03 JST 11 Mar (+) 0.2 m Maximum Tsunami 21:12 JST 11 Mar 0.6 m

Nanjo-shi Azama Initial Tsunami 17:49 JST 11 Mar (+) 0.3 m Maximum Tsunami 02:20 JST 12 Mar 0.4 m

Minami-daito-gyoko Initial Tsunami 17:11 JST 11 Mar (+) 0.2 m Maximum Tsunami 17:24 JST 11 Mar 0.2 m

Yonagunijima Kubura Initial Tsunami 18:34 JST 11 Mar (+) 0.1 m Maximum Tsunami 04:46 JST 12 Mar 0.2 m

Ishigakijima Ishigaki-ko Initial Tsunami 18:25 JST 11 Mar (+) Slight Maximum Tsunami 07:01 JST 12 Mar 0.3 m

Miyakojima Hirara Initial Tsunami 18:37 JST 11 Mar (+) 0.5 m Maximum Tsunami 19:34 JST 11 Mar 0.7 m

*3 mark: Arrival of initial tsunami unconfirmed.

No Tsunami Warnings and Advisories are currently in effect. Pay attention when fishing, swimming or engaging in other activities in the above coastal regions, as there may be still sea-level changes for the time being.

Earthquake Information Occurred at 14:46 JST 11 Mar 2011 Region name SANRIKU OKI Latitude 38.1N Longitude 142.9E Depth about 20 km Magnitude 9.0

*** [Tsunami Observed by offshore GPS buoys(marked with *)] ***

Please note that heights of all tsunamis observed by offshore GPS buoys are described as Unknown in the above text in order to avoid confusion with those observed by tidal stations in coastal areas. Below are the details of tsunamis observed by offshore GPS buoys.

93

Iwate Kamaishi-oki* Initial Tsunami 14:50 JST 11 Mar (-) 0.3 m Maximum 15:12 JST 11 Mar 6.8 m Tsunami Iwate Miyako-oki* Initial Tsunami 14:50 JST 11 Mar (-) 0.4 m Maximum 15:12 JST 11 Mar 6.3 m Tsunami Iwate Kuji-oki* Initial Tsunami 14:56 JST 11 Mar (-) 0.4 m Maximum Arrival of maximum tsunami

Tsunami expected. Kesennuma Hirotawan- Initial Tsunami 14:54 JST 11 Mar (+) 6.0 m oki* Maximum 15:14 JST 11 Mar 6.0 m Tsunami Fukushima Onahama-oki* Initial Tsunami 14:52 JST 11 Mar (+) 1.0 m Maximum 15:04 JST 11 Mar 1.0 m Tsunami Mie Owase-oki* Initial Tsunami 16:07 JST 11 Mar (+) 0.5 m Maximum 16:26 JST 11 Mar 0.5 m Tsunami Wakayama Shirahama- Initial Tsunami 16:25 JST 11 Mar (+) 0.3 m oki* Maximum 16:38 JST 11 Mar 0.3 m Tsunami

All tsunamis described here were observed offshore, and are therefore expected to increase in size by the time they reach coastal areas.

** [Estimated tsunami heights and arrivals times at coastal areas near offshore GPS buoys] ** Below are the estimated tsunami heights and arrival times at coastal areas near offshore GPS buoys.

Estimated Tsunami Arrival Time Estimated Tsunami Height

Iwate Kamaishi 14:55 - 15:10 JST 10 Mar 6 m - 10 m or more Iwate Miyako 14:55 - 15:10 JST 10 Mar 6 m - 10 m or more Iwate Kuji 15:01 - 15:16 JST 11 Mar 1 m - 2 m Kesennuma Hirotawan 14:59 - 15:14 JST 10 Mar 6 m - 10 m or more Fukushima Onahama 14:57 - 15:12 JST 10 Mar 1 m - 4 m Mie Owase 16:12 - 16:27 JST 11 Mar 1 m - 2 m Wakayama Shirahama 16:30 - 16:45 JST 11 Mar 0.5 m

Tsunamis may have already arrived in some areas.

94