International Journal of GEOMATE, Aug., 2020, Vol.19, Issue 72, pp.

1 - 7
ISSN: 2186-2982 (P), 2186-2990 (O), Japan, DOI: https://doi.org/10.21660/2020.72.5514
Special Issue on Science, Engineering and Environment

A STUDY ON HUMAN DAMAGE IN EVACUATION FROM

TSUNAMI CONSIDERING STREET-BLOCKADES CAUSED BY
DESTROY OF BUILDINGS

Ken-ichi Fujita1and Harumi Yashiro2

1
Faculty of Engineering, Nagasaki Institute of Applied Science, Japan;
2,3
School of Systems Engineering, National Defense Academy of Japan, Japan

*Corresponding Author, Received:14 Aug. 2019, Revised: 19 Nov. 2019, Accepted: 25 Feb. 2020

ABSTRACT: Damage estimation and disaster prevention plan for a Nankai megathrust earthquake which
occurrence is assumed in the future have been formulated cabinet office of Japan based on lesson learned from
the huge human damage caused by the tsunami of The Great East Japan Earthquake in 2011. Damage
estimation for earthquake and tsunami is usually evaluated independently. For the damage estimation by
tsunami, the human damage in evacuation from tsunami has been carried out. However, street-blockades have
not been considered in many studies and damage estimations. When evacuation routes are changed by the
street-blockades, configuration of the human damage in evacuation from tsunami seems to be different in no
street-blockades. To evaluate total damage in whole area, unified treatment of the human damage by the both
disasters seems to be significant. This study presents an evaluation method of human damage in evacuation
from tsunami considering street-blockades using area-wide mesh. The author’s method considering variances
of tsunami run-up speed and walking speed is used to the evaluation. The street-blockades of evacuation routes
are evaluated by using fragility curves of buildings for earthquake. Moreover, the difference in configuration
of the human damage by consider the street-blockades and no blockades is discussed.

Keywords: Tsunami evacuation, Human damage, Street-blockade, Human damage probability,

Area-wide mesh

1. INTRODUCTION of street walls and fire caused by large earthquakes

In disaster prevention planning for large scaled
natural disasters such as mega-class earthquake,
tsunami and eruption, estimation of buildings
damage and human damage is extremely important.
Damage estimation for the natural disaster in
cabinet office and local governments of Japan has
been usually evaluated independently for each
disaster [1], [2]. Using the damage estimation
results, disaster prevention planning for both
structural measures and non-structural measures are
projected.
The disaster prevention planning for tsunami are
as follows; construction of tide embankment,
relocation to higher ground of town, education on
evacuation and so on. Evacuation planning from
tsunami is especially important item for tsunami
disaster prevention. Evaluation of human damage
for tsunami is carried out for evacuation action to
safety area. In evaluation of the human damage,
difference between evacuation time and tsunami
arriving time to evacuation facilities is usually used.
In evaluating of human damage in evacuation from
tsunami, the effect of damage of buildings by
earthquake before arriving tsunami has not been
considered in many studies and damage estimation.
In evacuation planning from tsunami, to consider
street-blockade by destroy of buildings, fallen down
becomes important. The street-blockade has been
considered to lifesaving activities and fire-fighting
in the disaster estimation for large earthquakes [3].
When the street-blockades are caused a change of
evacuation routes, configuration of the human
damage in evacuation from a tsunami seems to be
different in no street-blockades.
This study presents an evaluation method of
human damage in evacuation from tsunami
considering the street-blockades by destroy of
buildings. Area-wide mesh is used to evaluate an
overview of the human damage in tsunami
evacuation under the street-blockades from a view
point of macro-perspective. The author’s method
[4]-[6] for the human damage taking account of
variances of tsunami run-up speed and walking
speed is also used to the evaluation considering the
street-blockades. Moreover, the difference in
configuration of the human damage between the
street-blockades and no street-blockades by using
the evaluation method.
2. EVALUATION FLOW OF HUMAN
DAMAGE
The area-wide mesh is used to the evaluation of
the human damage. Population composition, height
above sea level, tsunami inundation depth, seismic

1
 International Journal of GEOMATE, Aug., 2020, Vol.19, Issue 72, pp. 1 - 7
GIS Scenario earthquake
Area-wide mesh Tsunami hazard
Population and elevation Tsunami inundation area
Evacuation mesh
Conditions of evacuation
Evacuation time Tsunami arriving time
Damage probability
Survive
Evacuation completion Evacuation completion
Fig.1 Evaluation flow of human damage in evacuation from tsunami considered street-blockade
intensity, street-blockade ratio, and evacuation
awareness are set up for each mesh based on a real
area data. The population composition and the
height above sea level of GIS (Geographic
Information System) data are used.
Evacuation route from an evacuation mesh to an
evacuation facility is defined as the route along the
mesh. Evacuation distance including difference of
elevation between mesh is assumed to be 1.5 times
of length of plane distance. The human damage
probability is evaluated by using relation between
evacuation time and tsunami arriving time after
earthquake occurrence. The human damage ratio to
tsunami inundation depth in each mesh can be
obtained from the tsunami fragility curve [7]. The
street-blockade is evaluated by using the fragility of
buildings for earthquakes [8]. Decrement of the
walking speed is determined by effective sidewalk
width which passing is possible. From the above,
the evaluation flow on the human damage can be
shown in Fig.1.
The following assumptions are used in this
study.
1) The human damage is counted when a person is
caught up with tsunami.
2) The evacuation is only walk. The walking is only
permitted on sidewalk.
3) The evacuation facility is designated for each
2
Seismic hazard
Completely destroyed ratio and
partially destroyed ratio of buildings
Street-blockade ratio
Effective width of street
Decrement ratio of walking speed
Tsunami fragility on
human damage
Damage
Evacuation incompletion
evacuation mesh. The facilities are not damaged
to earthquakes.
4) The street-blockades are occurred by destroy of
buildings by large earthquakes
5) The evacuation direction is not toward to coast.
6) In emergency, to across any railroad crossing is
prohibited because crossing gates keeps down.
3. HUMAN DAMAGE PROBABILITY
3.1 Damage Probability of Human in Evacuation
The human damage probability is evaluated by
the Author’s method [5], [6]. The probability is
defined by a function of evacuation time and
tsunami arriving time after earthquake occurrence.
The evacuation time and the tsunami run-up time
from coast to evacuation facilities are assumed to
follow the normal distribution. The tsunami arriving
time is divided into the time of tsunami propagation
and run-up. The evacuation time and the tsunami
run-up time can be evaluated from walking speed
and tsunami run-up speed, respectively. The
variation of the tsunami run-up speed is considered
for evaluation of the tsunami arriving time. The
tsunami run-up time can be obtained by the relation
of the run-up speed and the minimum distance
between coast and evacuation facilities. The mean
 International Journal of GEOMATE, Aug., 2020, Vol.19, Issue 72, pp. 1 - 7

1
Wooden (Completely destroyed)
Wooden (Total destroyed)
0.8

Destruction ratio
0.6

0.4

0.2

0
5.5 6 6.5 7
Instrumental seismic intensity
(a) Wooden structure
Fig.2 Probability function of human damage 1 Non-wooden (Completely destroyed)
Non-wooden (Total destroyed)
0.8

Destruction ratio
value of the run-up speed can be determined by the
tsunami inundation depth [9]. Average tsunami
0.6
inundation depth from the coast to the evacuation
facilities is considered in this study. The standard 0.4
deviation of the run-up speed is evaluated by the
geometric standard deviation [10] which indicates 0.2
the adaptation between tsunami height marks on 0
land and tsunami simulation results. 5.5 6 6.5 7
In this study, the reliability evaluation method Instrumental seismic intensity
for structures [11] applied to the evaluation of the (b) Non-wooden structure
human damage. The evaluation equation of the
human damage can be provided the following Fig.3 Fragility curves of structures for earthquake [8]
equation.
partial destruction number of buildings. The street-
PR = 1 − Φ ( µ P / σ P ) (1) blockade ratio is evaluated for each mesh by using
the completely and partially destruction ratios.
where, Φ is the standard normal distribution Effective sidewalk width Wes by the street-
function with the mean value 0 and the standard blockade of each mesh can be obtained from the
deviation 1. The function of Φ is also shown in following equation;
Fig.2. In the figure, f e (t ) is the probability
function of evacuation time and f t (t ) is the Wes = (1 − Rd )Ws (4)
probability density function of tsunami arriving
time. µ P and σ P are provided by the following where, Ws is the sidewalk width in normal period,
equations, respectively; Rd is the street-blockade ratio which given for range
of Ws represented by the following equation;
µ P = µT − µ E (2)
 1.28Drb (Ws < 3m)

Rd = 0.604 Drb (3m ≤ Ws < 5.5m) (5)
σ P = σ T2 + σ E2 (3)
0.194 Drb
 (5.5m < Ws ≤ 13m)
where, µ E and µT are the mean value of total
evacuation time and the tsunami arriving time to where, Drb is the destruction ratio represented by
evacuation facilities after earthquake, respectively, the following equation [3];
σ E and σ T are the standard deviation of µ E and
µT , respectively. Drb = Dc +D p / 2 (6)

4.2 Street-Blockade by Destroy of Buildings
In this study, the street-blockade is assumed to
be caused by completely destruction and partial
destruction of buildings. The destroy ratio can be
obtained by using fragility of buildings for wooden
and non-wooden structures shown in Fig.3 [8]. The
partial destruction ratio is also obtained by a
difference completely destruction number and
where, Dc and D p are the completely and
partially destruction ratio of buildings, respectively.
Deterioration of the walking speed is assumed
to be proportional to the effective width. The
deterioration ratio of the speed maybe expressed by
using consideration of the decrement ratio of
walking speed of crowded with people [12]. In this
study, the sidewalk width is assumed to be
proportional to the crowded with people per area. In

3
 International Journal of GEOMATE, Aug., 2020, Vol.19, Issue 72, pp. 1 - 7

evacuation, no deterioration of the walking speed is

considered for Wes more than 1.5m, the speed is
16
linearly deteriorated with proportional to Wes
223 274
within 0.5m to 1.5m and impassable for less than
127 490 64 597
0.5m. The Deterioration ratio Rw may be evaluated 518 645 438 326 168
by the following equation; 705 518 410 473 408 526 603 312 459
752 863 640 384 499 684 726 660 339 464 538
 1.0 (1.5m < Wes ) 542 347 730 491 652 617 506

Rw = 0.8Wes − 0.2 (0.5m ≤ Wes ≤ 1.5m) (7) 87 642 441 499 590
 (Wes < 0.5m) 350 516 428
 0
142 331
63 299 Number: Population
The walking speed in evacuation is evaluated by 0 239
multiplying Rw the walking speed in normal period. : Evacuation facility
: Evacuation upland
5. EVALUATION OF HUMAN DAMAGE Fig.4 Area-wide mesh and population distribution

5.1 Evaluation Conditions

0.7
The human damage in evacuation from tsunami 1.0 1.0
using the method in this study is discussed under 2.5 1.6 3.5 1.6
conditions of street-blockade and no street- 1.6 2.5 2.5 2.5 2.5
blockade. The area-wide mesh and population 3.5 3.5 3.5 3.5 2.5 2.5 0.7 0.0 0.0
distribution of target area [13] is shown in Fig.4. 4.5 4.5 3.5 3.5 4.5 3.5 2.5 2.5 0.7 0.0 0.0
The mesh size is 250m x 250m and the population 6.5 4.5 4.5 4.5 4.5 4.5 2.5
is 23,345. 7.5 7.5 5.5 4.5 4.5
Tsunami inundation depth and seismic intensity 7.5 6.5 5.5
used in this study are based on a Meio type scenario 7.5 6.5
earthquake (moment magnitude scale, Mw 8.4) and 5.5 5.5
a scenario earthquake along the Miura peninsula 6.5 6.5 Number: Tsunami inundation
faults (Mw7.0), respectively. To investigate the depth (m)
maximum case of the human damages considered
the effect of the street-blockade, two types scenario Fig.5 Tsunami inundation depth
earthquakes which are the maximum case of the
damage estimation in the target area [13] is selected. 6+
Tsunami inundation depth and seismic intensity 6- 6+
(JMA seismic intensity scale) are shown in Figs.5 6+ 6+ 6+ 6+
and 6, respectively. In addition, the maximum 6- 6- 6- 6- 6+
tsunami height is 8.94m. Though tsunami arriving 6- 6- 6+ 6+ 6+ 6+ 6+ 6+ 6+
time to coast is 59 minute [13], to clear the 6- 6- 6+ 6+ 6+ 6+ 6+ 6+ 6+ 6+ 6+
difference of evacuation awareness, tsunami 6- 6+ 6+ 6+ 6+ 6+ 6-
arriving time in this study is set to 20 minute. 6- 6+ 6+ 6+ 6+
Buildings and roads are uniformly allocated to 6- 6+ 6-
each mesh because of the evaluation to be 6- 6-
simplification. Total number of buildings in the area 6- 6-
is 9,256 (wooden: 6,664, non-wooden: 2,592). Road 6- 6-
area including sidewalk per mesh is 7,375m2 which Number: Seismic intensity
is 11.8% of one mesh area.
Population ratio of the evacuation activity is Fig.6 Seismic intensity
divided into four type; evacuation immediately,
evacuation after work, urgency evacuation and no walking speed is shown in Table3. The values are
evacuation [14]. The population ratio is set up by the obtained by the statistics of walking speed in
combining the following three types evacuation normal period [15]. In addition, the mean value is
awareness; strong awareness, average awareness deteriorated by the effect of the street-blockade and
and weak awareness. The evacuation awareness and the standard deviation is assumed to be not variable
the start time of evacuation after earthquake for the blockade in this study. Mean value of
occurrence are shown in Tables 1 and 2, tsunami run-up speed υ t is evaluated by the
respectively. following equation [9];
The mean value and standard deviation of

4
 International Journal of GEOMATE, Aug., 2020, Vol.19, Issue 72, pp. 1 - 7

Table 1 Population ratio for evacuation awareness

Evacuation immediately
Evacuation Urgency evacuation No evacuation
After earthquake After work
awareness (Weak awareness) (Weak awareness)
(Strong awareness) (Weak awareness)
Strong 80% 10% 5% 5%
Middle 50% 25% 15% 10%
Weak 15% 35% 30% 20%

Table 2 Evacuation start time after earthquake
occurrence
Evacuation Evacuation
immediately immediately Urgency
after after finished evacuation
earthquake the work
Tsunami
5min. 15min.
arriving time
Table 3 Mean value and standard deviation of
walking speed
Mean value Standard deviation
1.34m/s 0.167m/s
υ t = 1.1 gh
where, g is the gravitational acceleration and h is the
tsunami inundation depth. Standard deviation of the
tsunami run-up speed is evaluated by using the
Aida’s equation [10].
5.2 Evaluation Results without Street-Blockade
The human damage for several evacuation
awareness under no street-blockade condition is
shown in Fig.7. The large number of the damage is
also shown with deep color. The number of the
damage with the strong awareness is the minimum.
The damage near the evacuation area is less than
that of the other mesh. The damage on mesh of the
coast and the railroad crossing becomes large.
The number of the human damage is shown in
Table 4. The human damage with the strong
awareness decreased to 22% of the damage with the
weak awareness. The evacuation awareness is large
influence to decrease the human damage.
5.3 Evaluation Results with Street-Blockade
The street-blockages ratio by destroyed of
buildings is shown in Fig.8 The ratio becomes large
for strong seismic intensity. In addition, the
numbers of destroyed buildings simulated in this
study are 1,801 for completely destroyed and 1,114
for partially destroyed. The decrement ratio of the
walking speed by the blockade is shown in Fig.9
The number shown in the figure represents the
Table 4 Total number of human damage
Evacuation
No blockade Blockade
awareness
Strong 2,631 4,181
Middle 6,543 8,401
Weak 11,997 13,857
Unit: person
destined walking speed ratio for each evacuation
mesh population. The ratio is variable with the scale
of the blockades.
The human damage for several evacuation
awareness under street-blockade condition is shown
in Fig.10 The human damage becomes large in spite
of the evacuation awareness compare to that of the
no blockage. The number of the damage with the
(8)
strong awareness is the minimum.
The number of the human damage for the
blockade is also shown in Table 4. The damage is
increased to 1.58, 1.28 and 1.16 times for the strong,
middle and weak awareness, respectively. To
consider the decrement of walking speed by the
blockage is important for the human damage
estimation from tsunami. Enhancing of evacuation
awareness as well as seismic strengthening of
buildings become the one way to decrease the
human damage in evacuation.
6. CONCLUSIONS
An evaluation method of the human damage
considering the street-blockade by destroy of
buildings is proposed. The human damage in
evacuation from tsunami is discussed for the street-
blockade and no street blockade conditions. The
following conclusions can be drawn.
1) The human damage under the street-blockade
becomes large compared to that of no street-
blockade because of decrement of the walking
speed by the blockade.
2) To consider the effect of the street-blockade in
evacuation planning is important.
3) The human damage in evacuation from tsunami
can be decreased by enhancing evacuation
awareness for evacuation and education on
disaster prevention.

5
 International Journal of GEOMATE, Aug., 2020, Vol.19, Issue 72, pp. 1 - 7

1 0.60

16 17 0.05 0.60
0.60 0.60 0.60 0.60
11 40 5 34
0.05 0.05 0.05 0.05 0.60
37 62 36 25 9
0.05 0.05 0.60 0.60 0.60 0.60 0.60 0.60 0.60
66 79 34 36 34 76 76 0 0
0.05 0.05 0.60 0.60 0.60 0.60 0.60 0.60 0.60 0.60 0.60
114 158 125 49 49 128 119 98 30 0 0
0.05 0.60 0.60 0.60 0.60 0.60 0.05
59 35 140 49 64 60 48
0.05 0.60 0.60 0.60 0.60
18 69 45 61 61
0.05 0.60 0.05
70 95 46 0.05 0.05
27 35 0.05 0.05
12 29 0.00 0.05
0 44 Number: Street-blockade ratio
Damaged population: 2,631
Fig.8 Street-blockade ratio by destroy of buildings
(a) Strong awareness 0.60
1.00 0.60
2 0.60 0.60 0.60 0.60
39 41 1.00 0.87 0.80 1.00 0.60
26 99 11 84 1.00 1.00 0.60 0.60 0.80 0.60 0.60 0.60 0.60

92 154 89 62 22 1.00 1.00 0.60 0.60 0.60 0.60 0.60 0.60 0.60 0.60 0.60
0.80 0.60 0.60 0.60 0.60 0.60 1.00
164 197 84 89 84 190 188 0 0
0.73 0.60 0.60 0.60 0.60
284 395 312 123 123 319 296 245 74 0 0
0.87 0.80 1.00
147 87 349 121 160 148 120
1.00 1.00
44 173 111 153 153
1.00 1.00
173 237 114 1.00 1.00 Number: Decrement ratio of
67 87 walking speed
29 72
0 110
Fig.9 Decrement ratio of walking speed by street-
blockade
Damaged population: 6,543
1
(b) Middle awareness 16 17
11 42 9 35
4 37 77 38 25 11
77 82 66 79 57 48 44 195 156 0 0
51 197 21 167 114 158 391 75 81 295 323 206 36 0 0
175 292 177 123 44 90 53 325 80 71 97 48
326 339 164 178 163 334 342 0 0 25 123 67 95 95
500 672 530 228 245 545 509 428 148 0 0 75 101 46
287 174 596 241 320 296 234 27 35
74 337 221 285 285 12 29
294 408 223 0 44
114 171 Damaged population: 4,181
49 141
(a) Strong awareness
0 186
2
Damaged population: 11,997
39 43
(c) Weak awareness 28 104 18 87
92 192 95 62 26
Fig.7 Human damage distribution without street- 164 197 141 120 110 291 290 0 0
blockade 284 395 479 185 202 429 438 342 89 0 0
224 131 470 198 177 242 120
4) Seismic strengthening of buildings is the one way 48 307 166 233 233
to decrease the street-blockade for large 177 252 114
earthquakes. 67 87
29 72
[1] Kanagawa Prefecture, Manual of Tsunami 0 110
Inundation Forecast, 2012. (in Japanese) Damaged population: 8,401
[2] Cabinet Office, Report of Damage Estimation of
Nankai Trough Earthquake (2st. Report), 2013. (b) Middle awareness

6
 International Journal of GEOMATE, Aug., 2020, Vol.19, Issue 72, pp. 1 - 7
4 [7] Koshimura S., Namegawa Y., Yanagisawa H.,
77 85
54 204 28 172
175 346 185 123 49
326 339 244 221 199 405 447
500 672 582 314 355 587 572 502 169
394 235 641 350 343 427 234
75 525 299 395 395
295 429 223
114 171
49 141
0 186
Damaged population: 13,857
(c) Weak awareness
Fig.10 Human damage distribution with street-
blockade
7. REFERENCES
[3] Cabinet Office, Report of Damage Estimation of
Nankai Trough Earthquake (1st. Report), 2012.
[4] Yashiro H., Fujita K., Evaluation Method of
Evacuation Safety Performance for Tsunami
using Area-Wide Mesh, Second European
Conference on Earthquake Engineering and
Seismology, CD-ROM, 2014, pp.1-9.
[5] Fujita K., Yashiro H., A Study on the Number
of Installed and Installation Location of
Tsunami Evacuation Facilities Using Area-wide
Mesh, Journal of Japan Society of Civil
Engineers, Ser. F6 (Safety Problem), Vol. 72,
No.2, 2016, pp.I_151-I_156.
[6] Fujita K., Yashiro H., Evaluation of Human
Damage in Tsunami Evacuation Considered the
Variance of Walking and Tsunami Run-up
Speeds, Internet Journal of Social Management
Systems, Issue 11 Vol.1, sms17-8200, 2017,
pp.213-223.
Fragility Functions for Tsunami Damage
Estimation, Journal of Japan Society of Civil
Engineers, ser. B, Vol.65 ， No.4 ， 2009,
0 0 pp.320-331.
0 0 [8] Midorikawa S., Ito Y., Miura H., Vulnerability
Functions of Buildings Based on Damage
Survey Data of Earthquake after the 1995 Kobe
Earthquake, Journal of Japan Association for
Earthquake Engineering, Vol.11, No.4, 2011,
pp.34-43.
[9] Shuto N., Imamura F., Koshimura S., Satake K.,
Matsutomi H., Encyclopedia of Tsunami,
Asakura Publishing Co., Ltd., 2007.
[10] Aida I., Simulations of Large Tsunamis
Occurring in the Past off the Coast of the
Sanriku District, Bulletin of the Earthquake
Research Institute, Univ. of Tokyo, Vol.52,
No.1, 2009, pp.71-101.
[11] Hoshiya M., Ishii K., Reliability Based Design
for Structures, Kajima Institute Publishing Co.,
Ltd., 1986.
[12] Cabinet Office, Report of Simulation results
on return home action, 2007.
[13] Zushi City Homepage, http://www.city.zushi.
kanagawa.jp/
[14] Ministry of Land, Infrastructure, Transport
and Tourism, Report of Present State Survey
in Tsunami Affected Area of the Great East
Japan Earthquake (3rd. Report), Results of
Actual Survey on Tsunami Evacuation
(Preliminary), 2011.
[15] Matsumoto N., Kiyota S., Ito M., Relationship
between the Characteristics of Streetscape and
Walking Speed, Journal of Architecture and
Planning, Vol.74, No.640, 2009, pp.1371-
1377.
Copyright © Int. J. of GEOMATE. All rights reserved,
including the making of copies unless permission is
obtained from the copyright proprietors.