Tsunami inundation hazard across Japan
Introduction
Tsunamis triggered by earthquakes are perceived as infrequent compared to earthquake ground shaking hazard and therefore treated as tail risk by emergency planners and the insurance industry. Rare mega-thrust subduction earthquakes are the major contributors whereas ground shaking hazard and risk have a more diverse set of contributing earthquake sources down to moderate magnitudes. Despite some pioneering works in the early 1980s mainly on Japan [1,2], detailed large-scale Probabilistic Tsunami Hazard Assessments (PTHA) have only recently been developed [[3], [4], [5], [6], [7]]. This is primarily due to the scarcity of large tsunamis, the challenges of assessing tsunamogenic earthquake sources, difficulties in differentiating between ground shaking and tsunami impacts as well as the need for detailed inundation modelling. In this study, we perform a PTHA for the entire Japanese coastline for comparative hazard assessment.
Evaluating tsunami hazard has improved in recent years through the introduction of probabilistic tsunami hazard assessment [3,4,8]. Many local, regional and global applications have been published aiming to improve the understanding of inundation depths, tsunami wave heights, and flooding areas using different approaches [4]. In particular following the 2004 Andaman-Sumatra and the 2011 Tohoku earthquakes, many lessons have been learned about requirements for near-field tsunami assessments in comparison to far-field assessment and countermeasures for tsunami hazard (e.g. Ref. [9,10]). Geist et al. [11] initially pointed to the importance of rupture complexity for the near-field tsunami hazard assessment, the uncertainty of incoming tsunami wave heights and inundation depths due to the rupture complexity. This has been further quantified in detailed studies in several papers mostly following the 2011 Tohoku event [10,[12], [13], [14]].
Japan is prone to tsunami hazard and risk as large portions of the population, commercial and industrial exposure are located close to the coastline. More than 50% of the total population of Asia exposed to tsunami hazard lives in Japan according to Ref. [7]. This specific exposure distribution increases the ratio of tsunami-to-earthquake hazard and risk in comparison to many other countries. For Japan it is essential to understand the tsunami hazard as basis for all avenues of risk assessment, disaster mitigation planning and upscaling community resilience for earthquake and tsunami risk. Several studies for Japan targeted near-field tsunami hazard due to earthquakes [12,[14], [15], [16]], however, a PTHA on a national scale that considers a comprehensive set of tsunamogenic earthquakes is missing.
In this study, we present a PTHA with results for the entire Japanese coast. Our analysis is based on offshore earthquake sources as characterized in the 2017 release of the National Map of Earthquake Prediction of the Earthquake Research Committee (ERC) of the Headquarters for Earthquake Research Promotion (HERP) under the Japanese ministry of Education, Culture, Sport, Science, and Technology. We use the information about earthquake sources and rate models available from the Japanese Seismic Hazard Information Station [[17], [18], [19]]. (J-SHIS, http://www.j-shis.bosai.go.jp/en/). We include one additional source geometry for the Tohoku section to complement possible tsunamogenic sources and substitute time-dependent rates for the largest earthquake sources assuming time-independence following [20].
We differentiate tsunami hazard on a national scale in terms of inundation depth for coastal regions of Japan utilizing cities/city wards located within the first 10 km as a set of nationally recognized administrative boundaries. We focus the PTHA on local earthquake tsunami sources and characterize uncertainty in tsunami generation due to the variability of possible rupture scenarios on causative offshore faults. Sources include subduction interface and normal faulting outer-rise earthquakes with a moment magnitude MW ≥ 7.5. Each earthquake either is associated with one (MW < 8.0) or multiple (MW ≥ 8.0) non-uniform slip distributions on its source geometry. The non-uniform slip-distributions are generated with the method of [21,22]. With these specifications, more than 700 tsunami sources are created for which we model tsunami wave propagation at a 50 m2 resolution along the entire Japanese coastline with an inland extent of a few kilometers. The number of earthquake ruptures is limited as the numerical simulation of tsunami waves along the entire Japanese coast is time consuming. We limit the input for tsunami generation to the final slip distributions, though we acknowledge that the temporal evolution of slip on a rupture plane can play a strong role when the initial tsunami waves are generated [23,24].
This PTHA uses only earthquake sources that also cause earthquake shaking hazard, i.e., we disregard tsunamis from other sources around the Pacific rim as well as all other possible sources of tsunamis. The total amount of inundation footprints per location partially limits robust calculations of statistics as not each individual grid cell is sampled with the same amount of inundation values. We treat this shortcoming by using the city wards as boundaries for our analysis so that the data used for our inferences is based on a large enough number of modelled solutions.
Our PTHA aims to model tsunamis that are prone to cause a substantial amount of damage. Thus we dismiss solutions that do not provide inundations depths with a maximum inundation depth of Imax = 0.5 m over larger coastal areas. We introduce this threshold based on engineering considerations and fragility/vulnerability functions developed with data of the 2011 Tohoku event [25,26]. This a priori threshold definition is not ideal, however, necessary to limit the immense computational demand for detailed tsunami-wave propagation and inundation modelling.
Our interest is to quantify recurrence for different thresholds of inundation depth hazard within city wards enabling to differentiate between them as information for potential stakeholders. We identify the major contributors to the inundation hazard and provide insight into which sources are relevant for a set of selected city wards across Japan and in the Tokyo bay area. Our analysis shows that the largest mega-thrust events are mostly, but not always, the key drivers of inundation hazard. In particular for the Tokyo region, the key drivers are events in the Sagami trough, a particularly complicated tectonic triple junction off-shore Tokyo and the source region of large historic events such as the 1703 Genroku and the 1923 Great Kanto earthquake [17,18,27,28].
Section snippets
Data
Japan has a long history of seismic hazard assessment and the Headquarters for Earthquake Research Promotion (HERP) provides regular updates on the earthquake sources, their geometries and recurrence rates on their website and the Japan Seismic Hazard Information Station (http://www.j-shis.bosai.go.jp). We utilize all offshore earthquake sources with a moment magnitude of MW ≥ 7.5 as possible tsunami sources. This set of events is a subset of the sources used for the 2017 National Seismic
Method
PTHA combines a set of individual components to estimate exceedance probabilities of an intensity parameter within a given time frame, such as exceedance probabilities of inundation heights or inundation levels within a given period. We utilize an event-based method to generate a probabilistic model that consists of five major components (Fig. 2). Similar event-based approaches have been proposed e.g. by [30], however, often with the simplifying assumptions of uniform slip. Here, we add the
Results
We illustrate tsunami inundation depth hazard and recurrence statistics for the administrative regions of city wards that locate entirely or partly within the first 10 km from any coastline. We disaggregate inundation depth hazard to understand the contributing sources in terms of rate, location and magnitude. In this way we shed light on the importance of the contributing sources. As metrics, we use the inundation depths for which annual rates and annual probability of exceedance curves are
Discussion & summary
We presented a probabilistic tsunami hazard assessment from local earthquake sources for Japan based on the 2017 Japanese Seismic hazard model [18] sources quantitatively differentiating mean and maximum tsunami inundation depth hazard at the city wards level. The analysis is based on 769 tsunami generating sources. The sources are characterized by time-independent rates based on and/or derived from the 2017 Japanese Seismic hazard model. These data are available from the Japanese Seismic
Data availability
Characterization of the earthquake sources is publically available from the Japanese Seismic Hazard Information Station (http://www.j-shis.bosai.go.jp/en/). We last access the data in October 2017. Bathymetry and elevation data were provided by the Cabinet Office of the Japanese Government.
We thank Prof. Diego Melgar for providing the MudPy package that we used as basis for our slip distribution modelling with modifications described in the manuscript (Table 1). The base codes are available on
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors acknowledge insightful discussion with members of the Model Development Department at Risk Management Solutions. We particulary thank D. Fitzenz, N. Shome, N. Porto and M. Nyst for critical internal reviews and discussions during development and manuscript preparation. We thank Prof. D. Melgar for sharing his code on github and initial guidance on the usage of several of the codes in the MudPy package.
References (63)
- et al.
Global tsunami hazard and exposure due to large co-seismic slip
Int J Disaster Risk Reduct
(2014) - et al.
Rapid tsunami loss estimation using regional inundation hazard metrics derived from stochastic tsunami simulation
Int J Disaster Risk Reduct
(2019) - et al.
The 2016 Fukushima earthquake and tsunami: local tsunami behavior and recommendations for tsunami disaster risk reduction
Int J Disaster Risk Reduct
(2017) - et al.
A new approach to probabilistic earthquake-induced tsunami risk assessment
Ocean Coast Manag.
(2016) Algorithms for the calculation of exact displacements, strains, and stresses for triangular dislocation elements in a uniform elastic half space
Comput. Geosci.
(2007)- et al.
History of the great Kanto earthquakes inferred from the ages of Holocene marine terraces revealed by a comprehensive drilling survey
Earth Planet Sci. Lett.
(2017) - et al.
Recent occurrences of serious tsunami damage and the future challenges of tsunami disaster risk reduction
Prog Disaster Sci
(2019) - et al.
A preliminary investigation of tsunami hazard
Bull. Seismol. Soc. Am.
(1982) - et al.
Tsunami hazard probability in Japan
Bull. Seismol. Soc. Am.
(1988) - et al.
Probabilistic analysis of tsunami hazards*
Nat. Hazards
(2006)
Probabilistic tsunami hazard analysis: multiple sources and global applications
Rev. Geophys.
A probabilistic tsunami hazard assessment for Indonesia
Nat. Hazards Earth Syst. Sci.
A probabilistic tsunami hazard assessment for western Australia
Pure Appl. Geophys.
Recent process in probabilistic tsunami HAzard analysis (PTHA) for mega thrust subduction earthquakes
Improvement of tsunami countermeasures based on lessons from the 2011 great east Japan earthquake and Tsunami ? Situation after five years
Coast Eng. J.
Complex earthquake rupture and local tsunamis
J. Geophys. Res.
Sensitivity of tsunami wave profiles and inundation simulations to earthquake slip and fault geometry for the 2011 Tohoku earthquake
Earth Planets Space
Quantification of source uncertainties in seismic probabilistic tsunami hazard analysis (SPTHA)
Geophys J Intell
Simulation-based probabilistic tsunami hazard analysis: empirical and robust hazard predictions
Pure Appl. Geophys.
Stochastic analysis and uncertainty assessment of tsunami wave height using a random source parameter model that targets a Tohoku-type earthquake fault
Stoch. Environ. Res. Risk Assess.
Long-Term Evaluation of Seismicity along the Nankai Trough (2nd Version - in Japanese)
National Map of Earthquake Prediction 2016
Technical Reports on the National Seismic Hazard Maps for Japan
Conditional probability of what? Example of the Nankai interface in Japan
Bull. Seismol. Soc. Am.
Kinematic rupture scenarios and synthetic displacement data: an example application to the Cascadia subduction zone
J Geophys Res Solid Earth
Generating random earthquake events for probabilistic tsunami hazard assessment
Pure Appl. Geophys.
Tsunami simulations of mega-thrust earthquakes in the Nankai–Tonankai Trough (Japan) based on stochastic rupture scenarios
Geol Soc London Spec Publ
Time and space distribution of coseismic slip of the 2011 Tohoku earthquake as inferred from tsunami waveform data
Bull. Seismol. Soc. Am.
A multivariate generalized linear tsunami fragility model for Kesenuma City based on maximum flow depths, velocities and debris impact, with evaluation of predictive accuracy
Nat. Hazards
Quantitative assessment of epistemic uncertainties in tsunami hazard effects on building risk assessments
Geosciences
Source model of the 1703 Genroku Kanto earthquake tsunami based on historical documents and numerical simulations: modeling of an offshore fault along the Sagami Trough
Earth Planets Space
Cited by (6)
Towards an integrated framework for the risk assessment of coastal structures exposed to earthquake and tsunami hazards
2022, Resilient Cities and StructuresProbabilistic tsunami hazard assessment for the makran subduction zone using logic tree and stochastic rupture sources
2024, Coastal Engineering JournalA new tsunami hazard assessment for eastern Makran subduction zone by considering splay faults and applying stochastic modeling
2023, Coastal Engineering JournalTsunami inundation characteristics along the Japan Sea coastline: effect of dunes, breakwaters, and rivers
2022, Earth, Planets and SpaceStochastic analysis of tsunami hazard of the 1945 Makran subduction zone MW 8.1–8.3 earthquakes
2020, Geosciences (Switzerland)