Adriano B. et al.
Paper:
Tsunami Inundation Mapping in Lima,
for Two Tsunami Source Scenarios
Bruno Adriano∗1, Erick Mas∗1, Shunichi Koshimura∗1, Yushiro Fujii∗2,
Sheila Yauri∗3, Cesar Jimenez∗4,∗5, and Hideaki Yanagisawa∗6
∗1Laboratory of Remote Sensing and Geoinformatics for Disaster Management,
International Research Institute of Disaster Science, Tohoku University
Aoba 6-6-03, Sendai 980-8579, Japan
E-mail: badrianoo@geoinfo.civil.tohoku.ac.jp
∗2International Institute of Seismology and Earthquake Engineering, Building Research Institute
1 Tachihara, Tsukuba, Ibaraki 305-0802, Japan
∗3Geophysical Institute of Peru (IGP)
Calle Badajoz 169, Mayorazgo IV Etapa, Ate Vitarte, Peru
∗4Fenlab, Universidad Nacional Mayor de San Marcos (UNMSM)
Av. Venezuela s/n, Lima, Peru
∗5Direccio´n de Hidrografı´a y Navegacio´n (DHN)
Calle Roca 116, Chucuito-Callao, Peru
∗6Department of Regional Management, Faculty of Liberal Arts, Tohoku Gakuin University
2-1-1 Tenjinzawa, Izumi-ku, Sendai, Miyagi 981-3193, Japan
[Received November 2, 2012; accepted December 19, 2012]
Within the framework of the project Enhancement of
Earthquake and Tsunami Disaster Mitigation Tech-
nology in Peru (JST-JICA SATREPS), this study de-
termines tsunami inundation mapping for the coastal
area of Lima city, based on numerical modeling and
two different tsunami seismic scenarios. Addition-
ally, remote sensing data and geographic information
system (GIS) analysis are incorporated in order to
improve the accuracy of numerical modeling results.
Moreover, tsunami impact is evaluated through appli-
cation of a tsunami casualty index (TCI) using tsunami
modeling results. Numerical results, in terms of max-
imum tsunami depth, show a maximum inundation
height of 6 m and 15.8 m for a potential scenario (first
source model) and for a past scenario (second source
model), respectively. In terms of inundation area, the
maximum extension is 1.3 km with a runup height of
5.3 m for the first scenario. The maximum extension
is 2.1 km with a runup height of 11.4 m for the second
scenario. The average TCI value obtained for the first
scenario is 0.36 for the whole inundation domain. The
second scenario gives a mean value of 0.64, where TCI
equal to 1.00 represents the highest condition of risk.
The results presented in this paper provide important
information about understanding tsunami inundation
features and, consequently, may be useful in designing
an adequate tsunami evacuation plan for Lima city.
Keywords: inundation modeling, inundation mapping
and casualty index
1. Introduction
Tsunami inundations have caused an unfortunate loss
of life and extensive property damage to coastal commu-
nities in seismic-prone countries. Tsunami events that oc-
curred in seismically similar areas such as the 2010 Chile
Tsunami, the 2012 Japan Tsunami and the December 26,
2004, Sumatra tsunami, for instance, are tragic reminders
of the devastation that these powerful events can gener-
ate in coastal areas. In the case of Peruvian seismic his-
tory, one of the most disastrous seismic events occurred
in 1746, whose epicenter was located in front of the cen-
tral area of Peruvian coast. According to historical testi-
mony, the capital city of Lima was completely destroyed
by ground shaking and the subsequent tsunami flood [1].
It is therefore important to develop an efficient tsunami
warning system in order to mitigate the catastrophic dam-
age due to tsunami events.
Detailed maps of potential tsunami inundation areas
are important for the delineation of evacuation routes and
long-term planning in vulnerable coastal communities.
The Direccio´n de Hidrografı´a y Naveagacio´n (DHN) [2]
is the Peruvian institution responsible for the Sistema
Nacional de Alerta de Tsunamis (SNAT The national
tsunami warning system). This institution publishes and
periodically updates inundation charts for main coastal
cities and ports along the whole Peruvian coast. These
inundation charts are mostly developed by using tsunami
refraction curves technique that estimates tsunami arrival
time and tsunami height along the coastline through pro-
gressive curves that are drawn from the epicenter and em-
pirical equations, respectively. This methodology is ex-
plained in [3], [4] and [5]. Additionally, the inundation
274 Journal of Disaster ResearchVol.8 No.2, 2013
Tsunami Inundation Mapping in Lima, for Two Tsunami Source Scenarios
Fig. 1. Inundation chart published by DHN for the La Punta district and Callao Province [2].
chart that covers the La Punta district and part of Callao
Province (see Fig. 1) was updated within the framework
of the project Sistema de Informacio´n Sobre Recursos
para Atencion de Desastres [6]. The inundation zone in-
dicated in this inundation chart is the result of the worst-
case scenario chosen from two different seismic sources
models and was calculated using the Tsunami Inundation
Modeling Exchange (TIME project) [7]. Fig. 1 shows the
inundation chart for La Punta and Callao.
The purpose of this study is to determine the tsunami
inundation area for the coastal region of Lima city by
adopting two different seismic source scenarios through
numerical modeling and a detailed topographymodel that
includes building height information and land use classi-
fication. Additionally, numerical modeling results such
as flow depths and current velocity are used to evaluate
tsunami impact by estimating a tsunami casualty index
within the inundation area.
2. Description of the Study Area
The study area essentially covers the central zone of the
Lima coast that corresponds to the constitutional province
of Callao (see Fig. 2). The geographic limits for the study
area are 11.95◦ – 12.09◦ south and 77.18◦ to 77.13◦ west.
Morphologically, the coastal plain rises to approximately
15 m above sea level near the shoreline. While the coast-
line to the north forms a long sweeping bay, however, the
coastline to the south of the study area presents a small
peninsula that corresponds to the La Punta district. Ac-
cording to the Peruvian Institute of Statistic and Informat-
ics [8] (INEI), the population of Lima city is about 9 mil-
lion people. For Callao province, however, there are about
950,000 living in this province. Urban use in the study
area is divided mainly into three types (see Fig. 2c). The
first corresponds to residential type, with its area concen-
trated in the southern part of the inundation domain be-
low 12◦02′30′′S. This is the most populated area in Callao
province and as a result, the land surface is largely cov-
ered by houses and avenues. The second urban use is lo-
cated in the center of the study area between 12◦01′S and
12◦02′30′′S. This zone is covered by extensive vegetated
area that corresponds to the back of the International Air-
port of Lima and agricultural fields. There are also some
towns along the coastline, where housing types are prin-
cipally non engineering buildings [8]. In the third zone
type, the land surface along the coastline is mainly occu-
pied by factories.
3. Tsunami Source Scenario
Seismic events accompanied by large tsunamis along
coastal central Peru have been reported since historical
times. The 1746 earthquake, for instance, is considered
Journal of Disaster ResearchVol.8 No.2, 2013 275
Adriano B. et al.
Fig. 2. a) View of the first domain (405 m grid resolution). Boundaries for other domains are shown in dashed lines. b) View
of bathymetry and topography datasets for inundation modeling (5th domain). c) Study area that corresponds to the 5th domain,
background view of WorldView-2 satellite images in true composite.
one of the most catastrophic seismic events in Peruvian
history. The moment magnitude for this event has been
estimated between 8.8 and 9.0 and the reported tsunami
runup height was between 15 m and 24 m [1, 9]. Two
centuries later, only two event with considerable magni-
tude have occurred. The 1966 earthquake, whose epicen-
ter was located in the north central coastal area of Peru,
had a rupture length of 100 km and a local tsunami height
of 2.6 m. The last, the 1974 earthquake that occurred in
front of Lima city, had a rupture length of 140 km and a
local tsunami height of 1.6 m [1]. No large earthquakes
have been reported in the area since then. There is there-
fore a clear seismic gap of approximately 250 years. Fur-
thermore, considering the seismic history of this area, it is
urgent to take into account the high possibility of the oc-
currence of an enormous seismic event accompanied by a
catastrophic tsunami in front of Lima city.
In the evaluation of seismic models, the application of
global positioning system (GPS) observation and Interfer-
ometry Synthetic Aperture Radar (InSAR) makes it pos-
sible to measure strain/displacement associated with plate
convergence movement in high-seismicity zones. [10]
used a GPS array with about 1,000 permanent stations to
recognize coseismic deformations associated with large
earthquakes and ongoing deformation over long years in
Japan. [11] and [12] used pre- and post event SAR im-
agery datasets to estimate crustal movement from the
2008 Wenchuan, China, earthquake and the 2011 To-
hoku, Japan, earthquake, respectively. Historical infor-
mation such as earthquake intensity perception, runup
measurement and wave arrival times in past earthquake
events plays an important role in estimating seismic mod-
els for potential earthquake scenarios that can be used to
evaluate possible future impact [13]. In this study, the
source model is based on two different seismic scenar-
ios. The first is a megathrust earthquake that would likely
adversely affect the metropolitan Lima region and that is
appropriate for the simulation of long-period wave and
tsunami modeling [14]. The seismic source is based on
a model of interseismic coupling distribution in subduc-
tion areas for a period of 265 years since the 1746 earth-
quake. This data also includes sea floor deformation mea-
surements obtained offshore from Lima city by a combi-
nation of GPS receivers and acoustic transponders [15],
as well as information on historical earthquakes, to pro-
pose a slip distribution. The seismic source is divided
into 280 sub fault segments, each 20 km × 20 km, in a
700 km by 160 km rupture area with a moment magni-
tude of 9.0 Mw. The slip solution model shows two main
asperities, the largest approximately 70 km west of Lima
city with maximum slip of 15.4 m, and the second south
of Lima with slip up to 13.0 m. Fig. 3a shows the spatial
location for slip model distribution. The second seismic
scenario is a model of the tsunami source of the 1746 Peru
earthquake that was calculated through a direct compari-
son of tsunami modeling results and the interpretation of
276 Journal of Disaster ResearchVol.8 No.2, 2013
Tsunami Inundation Mapping in Lima, for Two Tsunami Source Scenarios
Fig. 3. a) Slip distribution proposed by [14] used for the
first seismic scenario. b) Slip distribution proposed by [16]
for the second scenario. The slip scale is shown below for
each model.
historical documents about tsunami flooding [16]. This
scenario estimates a 550 km × 140 km rupture area for
five sub fault models with a 110 km × 140 km area. This
model considers maximum slip of 17.5 m approximately
50 km southwest in front of Lima city. Fig. 3b shows slip
model distribution for this seismic scenario.
3.1. Initial Sea Floor Displacement
We use a rectangular dislocation model [17] to calcu-
late ocean bottom deformation due to our source scenar-
ios. Initial sea floor displacement is assumed from the
instant push up of seismic deformation on the ocean bot-
tom. Features of the bottom deformation thus reflect ini-
tial water surface displacement that is assumed as an ini-
tial condition of tsunami propagation modeling. Fig. 4a
shows displacement result for the first source scenario
and Fig. 4b results for the second source scenario. Areas
in red and blue represent uplift and subsidence displace-
ment, respectively. Contour lines are drawn at 0.25 m in-
tervals. In the case of the first scenario, there are two re-
gions of high uplift deformation that are consistent with
slip model distribution. The higher uplift region has a
maximum displacement of 4.6 m and is located about
150 km in front of Lima coastal area. In the case of the
second scenario, displacement distribution shows a homo-
geneous uplift area that is concentrated at the center of
the rupture fault with a maximum displacement of about
8.2 m.
4. Tsunami Numerical Modeling
Numerical simulation was conducted by using the To-
hoku University Numerical Analysis Model for Investi-
gation of Near-field tsunami No.2 (TUNAMI-N2) code
based on shallow water theory and a Cartesian coordinate
system that was developed by the Disaster Control Re-
search Center (DCRC), Tohoku University, Japan. The
set of nonlinear shallow water equations (Eqs. (1), (2) and
(3)) are discretized using a staggered leap-frog finite dif-
ference scheme [18],
∂η
∂ t
+
∂M
∂x
+
∂N
∂y
= 0 . . . . . . . . . . (1)
∂M
∂ t
+
∂
∂x
(
M2
D
)
+
∂
∂y
(
MN
D
)
=
−gD∂η
∂x
− gn
2
D7/3
M
√
M2+N2 . . . (2)
∂N
∂ t
+
∂
∂x
(
MN
D
)
+
∂
∂y
(
N2
D
)
=
−gD∂η
∂y
− gn
2
D7/3
N
√
M2+N2 . . . (3)
M =
∫ η
−h
udz . . . . . . . . . . . . . (4)
M =
∫ η
−h
vdz . . . . . . . . . . . . . . (5)
D= η +h . . . . . . . . . . . . . . (6)
where M and N are the discharge flux in the x- and y-
directions, respectively; η is the water level and h is the
water depth with respect to mean sea level.
To perform tsunami inundation modeling, the compu-
tational area is divided into five domains to construct a
nested grid system (see Fig. 2). Bathymetry/topography
data for the first and second domains are resampled from
General Bathymetry Chart of the Ocean (GEBCO) 30 arc-
Journal of Disaster ResearchVol.8 No.2, 2013 277
Adriano B. et al.
Fig. 4. a) Sea floor displacement calculated from the pro-
posed slip distribution in the first source scenario. b) Sea
floor displacement calculated from the estimated slip distri-
bution [16]. Areas in red represent the uplift. Areas in blue
represent subsidence displacement. Contour lines are drawn
at 0.25 m intervals for each model.
seconds grid data. Bathymetry data for the third to fifth
domains are constructed from a nautical chart provided
by DHN and topography data are merged from the Ther-
mal Emission and Reflection Radiometer (ASTER) 1 arc-
second resolution raster data and 5 m resolution contour
line data from the Callao regional government. The grid
size varies from 405 m to 5 m.
4.1. Roughness Coefficient for Inundation Zone
Generally, based on the relation between the scale of an
obstacle and the grid size of the computational domain,
there are basically two approaches to be applied in inun-
dation modeling [19]. The first is a topographymodel that
uses a constant roughness coefficient for the whole com-
putational domain and a very detailed topography dataset
including building height informations. This approach is
used when the obstacle or urban use, e.g., buildings, sea-
walls, roads or fields, is represented adequately within
cells in the inundation model. This model thus uses a
finer grid size where flow around and between obstacles
is well simulated. The second approach is known as the
equivalent roughness model, applied when obstacles are
much smaller than grid size. [20] introduced an appropri-
ate methodology for this case that is based on the calcula-
tion of a building/house occupation ratio within grid cells
in the inundation domain. This method has been utilized
and validated through a comparison with field survey data
in order to develop tsunami fragility functions [21, 22].
Furthermore, based on these methodologies, [23] pre-
sented a comparison of three runup models -a constant
roughness model, a topography model and an equivalent
roughness model- in highly populated areas. They con-
cluded that the topographic model is used to identify the
distribution of inundation parameters in residential areas.
Since high grid resolution and urban use influence the
estimation bottom roughness coefficient, it may change
the propagation of waves considerably due to obstacles
or roughness induced energy dissipation. Consequently,
the final input raster should include features with appro-
priate elevation heights and adequate roughness surface
mapping [24]. In this study, in order to improve the ac-
curacy of numerical results, we enhanced the topography
model approach by adding a specific bottom roughness
coefficient for each grid cell in the inundation domain.
These coefficient values are based on a roughness coeffi-
cient map (see Fig. 5a). [25] performed land use classi-
fication using satellite images from WorldView-2 sensor.
Coefficient values are assigned according to land use [26]
in order to present a roughness coefficient map for the
coastal area of Lima city. Additionally, topography data is
combined with a height building model (see Fig. 5b). For
the La Punta district, a GIS building shape file on a single
house scale is available that includes the number of sto-
ries, spatial location and construction material type fields.
In this case, the height building model is constructed by
multiplying the number of stories by 2.5 m, the standard
story height in Lima city. For the rest of the computa-
tional domain, however, only the spatial location for a
block scale is available. Nevertheless, in order to con-
struct the height building model, we assume a two-story
height for each block as a preliminary estimation. Only
the historical building known as “Fortaleza Real Felipe”
(location coordinates −77.15W, −12.06S) was manually
delineated, however, and its height is fixed equal to 8 m,
which is approximately the real height of the perimeter
wall. Fig. 5 shows a roughness coefficient map and the
final topography model use for inundation modeling.
278 Journal of Disaster ResearchVol.8 No.2, 2013
Tsunami Inundation Mapping in Lima, for Two Tsunami Source Scenarios
Fig. 5. a) Roughness coefficient map [25]. b) GIS building
shape file used to construct the building height model.
5. Tsunami Inundation Mapping
The computation time for tsunami modeling is 3 hours.
In order to satisfy the stability condition, the time step
for numerical computation is fixed at 0.2 s. Tsunami
inundation is calculated on the fifth domain using
5 m resolution of bathymetry/topography grid data with
1567 × 3346 grid points in the longitude and latitude
directions, respectively. Fig. 6 shows synthetic tsunami
height recorded at the Callao tide gauge station, which is
located at 77.16W, 12.06S (see Fig. 10), within 3 hours
of simulation. The tsunami arrival time registered at the
Callao station is about 20 min for both source scenarios,
22 min and 25 min for the first and second, respectively.
In addition, the time step when maximum tsunami height
occurred, which is approximately equal to 38 min on av-
erage, is considered alike for both models. Taking into
account, however, that the bathymetry dataset is constant
in both numerical simulations, it is certain that there is a
difference in tsunami height amplitude, while the maxi-
mum tsunami wave height is approximately 5 m for the
first scenario. In the case of the second scenario, it is
approximately 10 m. This notable difference is consis-
tent with the initial sea floor displacement due to each
source scenarios (Fig. 4). Hence, clear differences regard-
ing tsunami flooding features should be expected.
The local inundation depth is the result of measuring
water marks on structures above the ground. Inundation
Fig. 6. Synthetic tsunami height recoded at the Callao tide
gauge station. The solid line was calculated using the first
source scenario [14]. The dashed line was calculated using
the second source scenario [16].
depth and inundation area results for each source scenario
are shown in Fig. 7. As expected, there is a clear differ-
ence in terms of tsunami depth and inundation area be-
tween the two scenarios. In terms of inundation depth, its
maximum values reach 6m and 15.8 m in the first and sec-
ond scenarios, respectively. Considering the input of the
building height model, the inundation depth in the case
of the second scenario therefore completely over flows
approximately 95% of the buiding within the inundation
area (see Fig. 7b). In contrats, for the first scenario, only
one-storey buildings (approximately 5%) are completely
inundated (see Fig. 7a). In the northern part of the compu-
tational domain, inundation distantce reaches a maximum
extension of 1 km with a runup heights of 6 m for the first
scenario. In the case of the second scenario reaches about
1.4 km with a runup heights of 12.4 m. In the surrounding
area of the Rimac river estuary 12◦02′S latitude tsunami
inundation extends up to 1.3 km with a runup height of
5.3 m for the first scenario that extends up to 2.1 km with
a runup height of 11.4 m for the second scenario. Tsunami
depth and inundation area characteristics for the southern
part of the computational domain (Callao-La Punta area)
is discussed later in section 5.2.
5.1. Tsunami Impact Assessment
In order to analyze tsunami impact in the study area,
we estimated potential tsunami casualties due to tsunami
flooding. [27] introduced a practical methodology that
uses a simple human model based on cylinder members
to evaluate the effect of hydrodynamic force on it. It
is thus possible to determine locations and times within
the tsunami inundation zone where casualties are likely
to occur. [23] presented an improvement on the previous
method by considering that human model falls by two dif-
ferent mechanisms. The first considers that hydrodynamic
force exceeds friction force on the soles of human model
feet (Eq. (7)) and the second occurs when the moment
from the bottom back of the heel due to the hydrodynamic
force is bigger than the resistance moment due to human
model weight (Eq. (8)).
Journal of Disaster ResearchVol.8 No.2, 2013 279
Adriano B. et al.
Fig. 7. a) Tsunami inundation calculated using the first
source scenario. b) Tsunami inundation calculated using the
second source scenario. The inundation depth scale is shown
for each model.
μ(W0−W )≤
(∫ 1
2
ρCDu2dS +
∫
ρCM
∂u
∂ t
dV
)
. . . . . . . . . . . . . . . . . (7)
(W0−W )IG ≤
(∫ 1
2
ρCDu2dS +
∫
ρCM
∂u
∂ t
dV
)
hG
. . . . . . . . . . . . . . . . . (8)
In the above equations, μ is the friction coefficient,W0
is body weight,W is buoyancy, hG is the vertical distance
from the floor to the resultant force due to the flow depth,
IG is the distance between the center of gravity of the body
and the bottom back of the heel, ρ is sea water density, u
is velocity, S is the perpendicular projection area accord-
ing to flow depth, (∂u/∂ t) is local acceleration, V is the
volume of the submerged human body, andCD andCM are
drag and inertia coefficients, respectively. [27] introduced
a tsunami casualty index (TCI) that is defined by Eq. (9)
where TC is the duration of the tsunami inundation flow
that satisfies Eqs. (7) and (8) and Ti is the total duration of
the tsunami inundation flow. TCI is used to illustrate the
spatial distribution of potential tsunami casualties.
TCI =
TC
Ti
. . . . . . . . . . . . . . (9)
In this study, the human body model is shown in Fig. 8.
This model is constructed by examining technical reports
Fig. 8. Schematic explanation and measurements of human
model adopted in this study.
elaborated by the Peruvian Health Institute [28], and its
measurements are a good physical representation of the
average Peruvian adult in Lima. Additionally, the friction
coefficient in foot soles is assumed as 0.7 [29] andCD and
CM are assumed equal to 1.0 and 0.5 [27], respectively.
TCI results for both models are shown in Fig. 9. Basi-
cally, the TCI for each model follows consistent behavior
similar to inundation depth and inundation area. Based
on the impact assessment model, howevaer, in the case
of the first scenario, there are areas with inundation depth
up to 4 m for which the TCI shows values of 0.5 (see
Fig. 9a), which might indicate that tsunami flow veloc-
ity is not strong enough to produce hydrodynamic force
able to adversely affect the balance of the human model
within 3 hours of tsunami simulation [23]. It is important
to note that some of these areas are concentrated in the La
Punta district, where the maximum TCI value is 0.52 for
most of the streets. The average TCI value obtained for
the first scenario is 0.36 throughout the whole inundation
domain. In the case of the second scenario, however, ap-
proximately 85% of the inundation area shows TCI values
over 0.8 (see Fig. 9b). Specifically, for the La Punta dis-
tric, practically 90% of sreets shows a TCI value of 0.93.
In the case of the second scenario, the mean value for the
TCI is 0.64. Additionally, based on our numerical calcula-
tion, the balance instability of the human model for each
cell grid in almost the whole inundation domain occurs
between 2 and 5 minutes after the tsunami wave arrives.
5.2. Tsunami Mapping for Callao-La Punta
A comparison of tsunami inundation for the Callao-La
Punta area is presented in Fig. 10. Based on numerical
results, the second scenario is almost twice that of the
first scenario in terms of inundation depth and inundation
area. This fact is due to the slip model of each scenario.
In the case of the first scenario, inundation depth values
reach 6 m, which may inundate one-storey buildings com-
280 Journal of Disaster ResearchVol.8 No.2, 2013
Tsunami Inundation Mapping in Lima, for Two Tsunami Source Scenarios
Fig. 9. a) TCI estimated using tsunami feature results from
the first source scenario. b) TCI estimated using tsunami
feature results from the second source scenario. The TCI
scale is shown for each model.
pletely. The inundation depth for most of the streets in
the La Punta district reaches values up to approximately
4 m. Based on the time series shows in the previous
Fig. 6, however, this state occurs three times with a time
period of about 50 minutes in 3 hours of tsunami simu-
lation. Inundation area extends several blocks on land in
Callao province (about 1 km of inundation) with a max-
imum runup height of 4 m (see Fig. 10a). In the case
of the second scenario, inundation depth reaches 15.8 m,
which is why almost 90% of the buildings are completely
inundated. The inundation depth for all of the streets in
the La punta district reaches a maximum value of 10.6 m.
There are therefore some buildings six stories high that
may not be inundated. The inundation area also extends
several blocks on land in Callao province (about 2.15 km
inundation) with a maximum runup height of 11 m (see
Fig. 10b). There is a clear inundation zone, however,
that extends 1.7 km and that corresponds to the inunda-
tion zone when buildings are completely inundated.
6. Conclusions
Tsunami inundation modeling using two different
tsunami source scenarios and a detailed bathymetry raster
dataset was carried out in order to determine the tsunami
Fig. 10. Comparison of tsunami inundation from both
source models for the Callao-La Punta area. a) Inundation
mapping using the first source scenario. b) Inundation map-
ping using the second source scenario. The green circle
shows the location of the Callao tide gauge station.
inundation area for the coastal region of Lima city. Ad-
ditionally, flow depth and flow velocity results were used
to evaluate tsunami impact by estimating a tsunami casu-
alty index within the inundation area. Inundation maps
developed using both tsunami scenarios show two highly
different tsunami risks for the study area. The inunda-
tion depth and inundation area for the second scenario
(past source scenario) shows most of the tsunami features
to be twice the results using the first scenario (potential
source scenario). If we therefore consider Peruvian seis-
mic history and lessons learned from past tsunami events
in other regions, such as the 2011 Tohoku tsunami, both
mapping results should be considered in order to propose
a tsunami warning system. Additionally, TCI maps esti-
mated using both scenarios show consistent behavior sim-
ilar to that shown by the inundation maps. Mean values of
Journal of Disaster ResearchVol.8 No.2, 2013 281
Adriano B. et al.
the TCI in the computation domain are 0.36 and 0.64 for
the first and second scenarios, respectively. These TCI
maps may be useful for developing tsunami evacuation
plans, especially in the case of the La Punta district where
the TCI value is greater than 0.5 for both models. Finally,
mapping results presented in this paper give important in-
formation for understanding tsunami inundation features,
and they therefore may be useful in designing an adequate
tsunami warning system that can be included in tsunami
evacuation plans for Lima city.
Acknowledgements
The authors acknowledge financial and other support from the
Japan Ministry of Education, Culture, Sports, Science and Tech-
nology (MEXT) and the project of Science and Technology Re-
search Partnership for Sustainable Development (SATREPS) sup-
ported by JST-JICA throughout the study. Such support made pos-
sible the forthcoming international conference, as well as interna-
tional joint research projects and exchange programs with over-
seas institutions.
References:
[1] L. Dorbath, A. Cisternas, and C. Dorbath, “Assessment of t h e
size of large and great historical earthquakes in Peru,” Bulletin of
the Seismological Society of America, Vol.80, No.3, pp. 551-576,
1990.
[2] Direccio´n de Hidrografı´a y Naveagacio´n, “Cartas de Inundacio´n,”
2012.
[3] H. Godoy and J. Monge, “Metodologı´a para la evaluacio´n del riesgo
de tsunami. Santiago, Chile,” tech. rep., Publicacio´n SES I 3-75,
1975.
[4] A. Delgado and C. Garcı´a, “Plan de Evacuacio´n de Ciudades Afec-
tadas por Tsunamis, Zona La Punta-Pucusana. Lima,” B.s. eng. the-
sis, National University of Engineering, Peru, 1982.
[5] E. Mas and V. Jacome, “Estudios de tsunamis de origen cercano en
el Callao centro-norte, planes de evacuacion y uso de suelo,” B.s.
eng. thesis, National University of Engineering, Peru, 2008.
[6] PNUD/SDP-052/2009, “Investigacio´n sobre el PELIGRO DE
TSUNAMI en el A´rea Metropolitana de Lima y Callao,”
in SISTEMA DE INFORMACIO´N SOBRE RECURSOS
PARA ATENCIO´N DE DESASTRES, COOPERAZIONE INTER-
NAZIONALE COOPI, 2010.
[7] C. Goto and Y. Ogawa, “Numerical Method of Tsunami Simulation
with the Leap-frog Scheme. Translated for the TIME project by N.
Shuto,” 1992.
[8] Instituto Nacional de Estadstica e Informatica (Institute of Statistic
and Informatics), “Censos Nacionales 2007,” 2007.
[9] S. L. Beck and L. J. Ruff, “Great earthquakes and subduction along
the Peru trench,” Physics of the Earth and Planetary Interiors,
Vol.57, pp. 199-224, Nov. 1989.
[10] T. Sagiya, S. Miyazaki, and T. Tada, “Continuous GPS array and
present-day crustal deformation of Japan,” Pure and Applied Geo-
physics, Vol.157, pp. 2303-2322, 2000.
[11] M. Hashimoto, M. Enomoto, and Y. Fukushima, “Coseismic De-
formation from the 2008 Wenchuan, China, Earthquake Derived
from ALOS/PALSAR Images,” Tectonophysics, Vol.491, pp. 59-
71, Aug. 2010.
[12] W. Liu and F. Yamazaki, “Detection of Crustal Movement From
TerraSAR-X Intensity Images for the 2011 Tohoku, Japan Earth-
quake,” IEEE GEOSCIENCE AND REMOTE SENSING LET-
TERS, Vol.10, No.1, pp. 199-203, 2013.
[13] J. Kuroiwa, “Disaster Reduction, Living in harmony with nature,”
Lima: Editorial NSG S.A.C, first edit ed., 2004.
[14] N. Pulido, H. Tavera, H. Perfettini, M. Chlieh, Z. Aguilar, S. Aoi,
S. Nakai, and F. Yamazaki, “Estimation of Slip Scenarios for
Megathrust Earthquakes: A Case Study for Peru,” in Effects of
Surface Geology on Seismic Motion, pp. 1-6, 2011.
[15] K. Gagnon, C. D. Chadwell, and E. Norabuena, “Measuring the
onset of locking in the Peru-Chile trench with GPS and acoustic
measurements.,” Nature, Vol.434, pp. 205-8, Mar. 2005.
[16] C. Jimenez, N. Moggiano, E. Mas, B. Adriano, S. Koshimura, Y.
Fujii, and H. Yanagisawa, “Seismic Source of 1746 Callao Earth-
quake from Tsunami Numerical Modeling,” Journal of Disaster
Research, Vol.8, No.2, pp. 266-273, 2013 (this number).
[17] Y. Okada, “Surface deformation due to shear and tensile faults in
a half-space,” Bulletin of the Seismological Society of America,
Vol.75, No.4, pp. 1135-1154, 1985.
[18] F. Imamura, “Review of the tsunami simulation with a finite differ-
ence method, Long Wave Run-up Models,” World Scientifi, pp. 25-
42, 1995.
[19] S. J. Hong, “Study on the Two and Three Dimensional Numerical
Analysis of Tsunamis near a coastal Area,” Ph.D thesis, Tohoku
University, Japan, 2004.
[20] T. Aburaya and F. Imamura, “The proposal of a tsunami runup
simulation using combined equivalent roughness,” Annual Journal
of Coastal Engineering, Japan Society of Civil Engineers, Vol.49,
pp. 276-280, 2002.
[21] S. Koshimura and T. Oie, “Developing fragility functions for
tsunami damage estimation using numerical model and post-
tsunami data from Banda Aceh, Indonesia,” Coastal Engineering
Journal, Vol.51, No.3, pp. 243-273, 2009.
[22] A. Suppasri, S. Koshimura, and F. Imamura, “Developing tsunami
fragility curves based on the satellite remote sensing and the nu-
merical modeling of the 2004 Indian Ocean tsunami in Thailand,”
Natural Hazards and Earth System Science, Vol.11, pp. 173-189,
Jan. 2011.
[23] A. Muhari, F. Imamura, S. Koshimura, and J. Post, “Examination
of three practical run-up models for assessing tsunami impact on
highly populated areas,” Natural Hazards and Earth System Sci-
ence, Vol.11, pp. 3107-3123, Dec. 2011.
[24] G. Gayer, S. Leschka, I. No¨hren, O. Larsen, and H. Gu¨nther,
“Tsunami inundation modelling based on detailed roughness maps
of densely populated areas,” Natural Hazards and Earth System
Science, Vol.10, pp. 1679-1687, Aug. 2010.
[25] B. Adriano, E. Mas, S. Koshimura, and Y. Fujii, “Remote Sensing-
based Assessment of Tsunami Vulnerability in the coastal area of
Lima , Peru,” in The 10th International Workshop on Remote Sens-
ing for Disaster Management, 2012.
[26] M. Kotani, F. Imamura, and N. Shuto, “Tsunami run-up simulation
and damage estimation by using GIS,” in Proceedings of coastal
engineering, JSCE 45, pp. 356-360, 1998.
[27] S. Koshimura, T. Katada, H. O. Mofjeld, and Y. Kawata, “A method
for estimating casualties due to the tsunami inundation flow,” Nat-
ural Hazards, Vol.39, pp. 265-274, Oct. 2006.
[28] Ministerio de Salud del Peru, “Instituto Nacional de la Salud,” 2012.
[29] H. Yeh, “Gender and Age Factors in Tsunami Casualties,” Natural
Hazards Review, Vol.11, pp. 29-34, Feb. 2010.
282 Journal of Disaster ResearchVol.8 No.2, 2013
Tsunami Inundation Mapping in Lima, for Two Tsunami Source Scenarios
Name:
Bruno Adriano Ortega
Affiliation:
Research Student, Laboratory of Remote Sens-
ing and Geoinformatics for Disaster Manage-
ment (ReGiD), International Research Institute
of Disaster Science (IRIDeS), Tohoku University
Address:
Aoba 6-6-3, Sendai 980-8579, Japan
Brief Career:
2009-2010 Master of Disaster Management, National Graduate Institute
for Policy Studies, Japan
2010-2012 Adjunct Professor, Faculty of Civil Engineering, National
University of Engineering, Peru
2012-present Research Student, ReGiD, IRIDeS, Tohoku University, Japan
Selected Publications:
• B. Adriano, E. Mas, S. Koshimura, and Y. Fujii, “Remote Sensing-based
Assessment of Tsunami Vulnerability in the coastal area of Lima, Peru,” in
The 10th International Workshop on Remote Sensing for Disaster
Management, Japan, 2012.
• E. Mas, B. Adriano, S. Koshimura, F. Imamura, J. Kuroiwa, F.
Yamazaki, C. Zavala, and M. Estrada, “Evaluation of Tsunami Evacuation
Building Demand through the Multi-Agent System Simulation of
Residents’ Behavior,” in Proceedings of International Sessions in Coastal
Engineering, JSCE, Vol.3, 2012.
• B. Adriano, S. Koshimura, and Y. Fujii, “Tsunami Source and Inundation
Modeling of the June 2001 Peru Earthquake,” in Joint Conference
Proceedings 9CUEE/4ACEE, Tokyo, pp. 2061-2065, 2012.
Academic Societies & Scientific Organizations:
• Peruvian Engineering College
• Japan Geoscience Union (JpGU)
Name:
Erick Mas Samanez
Affiliation:
Assistant Professor, Laboratory of Remote Sens-
ing and Geoinformatics for Disaster Manage-
ment (ReGiD), International Research Institute
of Disaster Science (IRIDeS), Tohoku University
Address:
Aoba 6-6-3, Sendai 980-8579, Japan
Brief Career:
1999-2004 B.S. Civil Engineering, National University of Engineering,
Peru
2006-2009 M.Sc. Disaster Risk Management, National University of
Engineering, Peru
2009-2012 PhD Civil Engineering, Tsunami Engineering, Tohoku
University, Japan
2012- Assistant Professor, ReGiD, IRIDeS, Tohoku University, Japan
Selected Publications:
• E. Mas, A. Suppasri, F. Imamura, and S. Koshimura, “Agent Based
simulation of the 2011 Great East Japan Earthquake Tsunami evacuation.
An integrated model of tsunami inundation and evacuation,” Journal of
Natural Disaster Science, Vol.34, Iss.1, pp. 41-57, 2012.
• E. Mas, S. Koshimura, A. Suppasri, M. Matsuoka, M. Matsuyama, T.
Yoshii, C. Jimenez, F. Yamazaki, and F. Imamura, “Developing Tsunami
fragility curves using remote sensing and survey data of the 2010 Chilean
Tsunami in Dichato,” Natural Hazards and Earth System Science, Vol.12,
pp. 2689-2697, 2012, doi:10.5194/nhess-12-2689-2012.
Academic Societies & Scientific Organizations:
• Japan Geoscience Union (JpGU)
• European Geosciences Union (EGU)
• American Geophysical Union (AGU)
Name:
Shunichi Koshimura
Affiliation:
Professor, Laboratory of Remote Sensing
and Geoinformatics for Disaster Management
(ReGiD), International Research Institute of
Disaster Science (IRIDeS), Tohoku University
Address:
Aoba 6-6-3, Sendai 980-8579, Japan
Brief Career:
2000-2002 JSPS Research Fellow, National Oceanic and Atmospheric
Administration
2002-2005 Research Scientist, Disaster Reduction and Human Renovation
Institute
2005-2012 Associate Professor, Tohoku University
2012- Professor, ReGiD, IRIDeS, Tohoku University, Japan
Selected Publications:
• S. Koshimura, T. Oie, H. Yanagisawa, and F. Imamura, “Developing
fragility functions for tsunami damage estimation using numerical model
and post-tsunami data from Banda Aceh, Indonesia,” Coastal Engineering
Journal, No.3, pp. 243-273, 2009.
• S. Koshimura, Y. Hayashi, K. Munemoto, and F. Imamura, “Effect of the
Emperor seamounts on trans-oceanic propagation of the 2006 Kuril Island
earthquake tsunami,” Geophysical Research letters, Vol.35, L02611,
doi:10.1029/2007GL032129, 24, 2008.
• S. Koshimura, T. Katada, H. O.Mofjeld, and Y. Kawata, “A method for
estimating casualties due to the tsunami inundation flow,” Natural Hazard
Vol.39, pp. 265-274, 2006.
Academic Societies & Scientific Organizations:
• Japan Society of Civil Engineers (JSCE)
• Institute of Social Safety Science
• Japan Association for Earthquake Engineering (JAEE)
• Japan Society for Computational Engineering and Science (JSCES)
• American Geophysical Union (AGU)
Name:
Yushiro Fujii
Affiliation:
Senior Research Scientist, International Insti-
tute of Seismology and Earthquake Engineering,
Building Research Institute
Address:
1 Tachihara, Tsukuba, Ibaraki 305-0802, Japan
Brief Career:
2004 Ph.D. Earth and Planetary Sciences, Faculty of Sciences, Kyushu
University, Japan
2005 Researcher, Research Institute for Information Technology, Kyushu
University, Japan
2006 Researcher, AFRC, Advanced Industrial Science and Technology
2006- Research Scientist, IISEE, Building Research Institute, Japan
Selected Publications:
• Y. Fujii and K. Satake, “Slip Distribution and Seismic Moment of the
2010 and 1960 Chilean Earthquakes Inferred from Tsunami Waveforms
and Coastal Geodetic Data,” Pure and Applied Geophysics, 2012.
• Y. Fujii, K. Satake, S. Sakai, M. Shinohara, and T. Kanazawa, “Tsunami
source of the 2011 off the Pacific coast of Tohoku Earthquake,” Earth,
Planets and Space, Vol.63, pp. 815-20, 2011.
• Y. Fujii and K. Satake, “Tsunami Source of the 2004 Sumatra-Andaman
Earthquake Inferred from Tide Gauge and Satellite Data,” Bulletin of the
Seismological Society of America, Vol.97, pp. S192-S207, 2007.
Academic Societies & Scientific Organizations:
• Seismological Society of Japan (SSJ)
• Japan Geoscience Union (JpGU)
• American Geophysical Union (AGU)
Journal of Disaster ResearchVol.8 No.2, 2013 283
Adriano B. et al.
Name:
Sheila Yauri Condo
Affiliation:
Assistant Research, Instituto Geofisico del Peru´
(IGP)
Address:
Calle Badajoz 169, Mayorazgo IV etapa, Ate Vitarte, Peru´
Brief Career:
2007-2010 Assistant Researcher, Seccio´n de Seismologı´a, Instituto
Geofisico del Peru´
2012- Assistant Researcher, Seccio´n de Geofı´sica y Sociedad, Instituto
Geofisico del Peru´
Selected Publications:
•M. Ioualalen, H. Perfettini, S. Yauri, C. Jimenez, H. Tavera, “Tsunami
Modeling to Validate Slip Models of the 2007 Mw8.0 Pisco Earthquake,
Central Peru,” Pure and Applied Geophysics. 2012.
• S. Yauri, F. Fujii, B. Shibazaki, “Tsunami hazard assessment for the
central coast of Peru using numerical simulations for the 1974, 1966 and
1746 earthquakes,” Master thesis, National Graduate Institute for Policy
Studies, Japan, 2012.
• S. Yauri, H. Tavera,“Caractersticas Generales del Tsunami del 15 de
Agosto de 2007. En Tavera, H. (Ed.) El terremoto de Pisco (Peru) del 15
de Agosto de 2007 (7.5 Mw),” Direccin de Sismologs - CNDG / Instituto
Geofsico del Per. Volumen Especial, pp. 371-386, 2007.
Academic Societies & Scientific Organizations:
• Society Geological of Peru (SGP)
Name:
Cesar Jimenez Tintaya
Affiliation:
Fenlab, Universidad Nacional Mayor de San
Marcos (UNMSM)
Direccio´n de Hidrografı´a y Naveagacı´on (DHN)
Address:
Av. Venezuela s/n, Lima, Peru´
Calle Roca 116, Chucuito-Callao, Peru´
Brief Career:
2000-2007 Assistant Research, Instituto Geofisico del Peru´ IGP
2008-2012 Research Scientist, Direccio´n de Hidrografı´a y Naveagacı´on
2009-2012 Assistant Professor, UNMSM, Peru´
2012- Graduate Studies in Geophysics, UNMSM, Peru´
Selected Publications:
•M. Ioualalen, H. Perfettini, S. Yauri, C. Jimenez and H. Tavera,
“Tsunami Modeling to Validate Slip Models of the 2007 Mw8.0 Pisco
Earthquake, Central Peru,” Pure and Applied Geophysics. 2012.
• E. Mas, S. Koshimura, A. Suppasri, M. Matsuoka, M. Matsuyama, T.
Yoshii, C. Jimenez, et al. (2012). “Developing Tsunami fragility curves
using remote sensing and survey data of the 2010 Chilean Tsunami in
Dichato,” Natural Hazards and Earth System Science, 12, 2689-2697.
doi:10.5194/nhess-12-2689-2012.
• T. Yoshii, M. Imamura, M. Matsuyama, S. Koshimura, M. Matsuoka, E.
Mas and C. Jimenez. “Salinity in Soils and Tsunami Deposits in Areas
Affected by the 2010 Chile and 2011 Japan Tsunamis,” Pure and Applied
Geophysics, 2012.
Academic Societies & Scientific Organizations:
• Peruvian Society of Physics (SOPERFI)
Name:
Hideaki Yanagisawa
Affiliation:
Department of Regional Management, Faculty
of Liberal Arts, Tohoku Gakuin University
Address:
2-1-1 Tenjinzawa, Izumi-ku, Sendai, Miyagi 981-3193, Japan
Brief Career:
2008-2009 Post-Doctoral Research Fellow, Graduate School of
Engineering, Tohoku University
2009- Company Member, Tokyo Electric Power Services Company
Limited
2012- Lecturer, Department of Regional Management, Faculty of Liberal
Arts, Tohoku Gakuin University
Selected Publications:
• H. Yanagisawa, S. Koshimura, K. Goto, T. Miyagi, F. Imamura, A.
Ruangrassamee, and C. Tanavud, “Damage of mangrove forest by the
2004 Indian Ocean tsunami at Pakarang Cape and Namkem, Thailand,”
Polish Journal of Environmental Studies, Vol.18, No.1, pp. 35-42, 2009.
• H. Yanagisawa, S. Koshimura, K. Goto, T. Miyagi, F. Imamura, A.
Ruangrassamee, and C. Tanavud, “The reduction effects of mangrove
forest on a tsunami based on field surveys at Pakarang Cape, Thailand and
numerical analysis, Estuarine,” Coastal and Shelf Science, Vol.81, pp.
27-37, 2009.
Academic Societies & Scientific Organizations:
• Japan Society of Civil Engineers (JSCE)
• Japan Society for Mangroves
284 Journal of Disaster ResearchVol.8 No.2, 2013