Articles | Volume 23, issue 3
https://doi.org/10.5194/nhess-23-1207-2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Special issue:
https://doi.org/10.5194/nhess-23-1207-2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Research article
| 
22 Mar 2023
Research article |
| 22 Mar 2023
| 22 Mar 2023
Development of a seismic loss prediction model for residential buildings using machine learning – Ōtautahi / Christchurch, New Zealand
Department of Civil and Environmental Engineering, Waipapa Taumata Rau / University of Auckland, Tāmaki Makaurau / Auckland, New Zealand
Quincy Ma
Department of Civil and Environmental Engineering, Waipapa Taumata Rau / University of Auckland, Tāmaki Makaurau / Auckland, New Zealand
Pavan Chigullapally
Department of Civil and Environmental Engineering, Waipapa Taumata Rau / University of Auckland, Tāmaki Makaurau / Auckland, New Zealand
Joerg Wicker
School of Computer Science, Waipapa Taumata Rau / University of Auckland, Tāmaki Makaurau / Auckland, New Zealand
Liam Wotherspoon
Department of Civil and Environmental Engineering, Waipapa Taumata Rau / University of Auckland, Tāmaki Makaurau / Auckland, New Zealand
Related authors
No articles found.
James H. Williams, Thomas M. Wilson, Nick Horspool, Ryan Paulik, Liam Wotherspoon, Emily M. Lane, and Matthew W. Hughes
Nat. Hazards Earth Syst. Sci., 20, 451–470, https://doi.org/10.5194/nhess-20-451-2020, https://doi.org/10.5194/nhess-20-451-2020, 2020
Short summary
Short summary
Post-event field survey data from two tsunami events, the 2011 Tōhoku tsunami, Japan, and the 2015 Illapel tsunami, Chile, are used in this study to develop fragility functions for roads and bridges. This study demonstrates the effectiveness of supplementing post-event field surveys with remotely sensed data. The resulting fragility functions address a substantial research gap in tsunami impacts on infrastructure and include a range of subtleties in asset and hazard characteristics.
Related subject area
Earthquake Hazards
Testing machine learning models for heuristic building damage assessment applied to the Italian Database of Observed Damage (DaDO)
The seismic hazard from the Lembang Fault, Indonesia, derived from InSAR and GNSS data
Rapid estimation of seismic intensities by analyzing early aftershock sequences using the robust locally weighted regression program (LOWESS)
Towards improving the spatial testability of aftershock forecast models
Accounting for path and site effects in spatial ground-motion correlation models using Bayesian inference
The footprint of an ancient forgotten earthquake: a VI Cent. A.D. event in the European Western Southern Alps
Seismogenic potential and tsunami threat of the strike-slip Carboneras fault in the western Mediterranean from physics-based earthquake simulations
Earthquake hazard characterization by using entropy: application to northern Chilean earthquakes
Seismic risk scenarios for the residential buildings in the Sabana Centro province in Colombia
Looking for undocumented earthquake effects: a probabilistic analysis of Italian macroseismic data
Spatiotemporal seismicity pattern of the Taiwan orogen
A web-based GIS (web-GIS) database of the scientific articles on earthquake-triggered landslides
Evaluation of liquefaction triggering potential in Italy: a seismic-hazard-based approach
Earthquake vulnerability assessment of the built environment in the city of Srinagar, Kashmir Himalaya, using a geographic information system
Earthquake-induced landslides in Norway
PERL: a dataset of geotechnical, geophysical, and hydrogeological parameters for earthquake-induced hazards assessment in Terre del Reno (Emilia-Romagna, Italy)
A non-extensive approach to probabilistic seismic hazard analysis
Inferring the depth and magnitude of pre-instrumental earthquakes from intensity attenuation curves
Tsunami scenario triggered by a submarine landslide offshore of northern Sumatra Island and its hazard assessment
Scrutinizing and rooting the multiple anomalies of Nepal earthquake sequence in 2015 with the deviation–time–space criterion and homologous lithosphere–coversphere–atmosphere–ionosphere coupling physics
On the calculation of smoothing kernels for seismic parameter spatial mapping: methodology and examples
Mass flows, turbidity currents and other hydrodynamic consequences of small and moderate earthquakes in the Sea of Marmara
Brief communication: The crucial assessment of possible significant vertical movements preceding the 28 December 1908, Mw = 7.1, Messina Straits earthquake
Seismic Background Noise Levels in Italian Strong Motion Network
Geologic and geodetic constraints on the magnitude and frequency of earthquakes along Malawi's active faults: the Malawi Seismogenic Source Model (MSSM)
Probabilistic fault displacement hazard analysis for the north Tabriz fault
Landslides triggered by the 2015 Mw 6.0 Sabah (Malaysia) earthquake: inventory and ESI-07 intensity assignment
Pseudo-prospective testing of 5-year earthquake forecasts for California using inlabru
An updated area-source seismogenic model (MA4) for seismic hazard of Italy
Identifying plausible historical scenarios for coupled lake level and seismicity rate changes: the case for the Dead Sea during the last 2 millennia
Analysis of seismic strain release related to the tidal stress preceding the 2008 Wenchuan earthquake
A morphotectonic approach to the study of earthquakes in Rome
Fault slip potential induced by fluid injection in the Matouying enhanced geothermal system (EGS) field, Tangshan seismic region, North China
Magnitude and source area estimations of severe prehistoric earthquakes in the western Austrian Alps
Hidden-state modeling of a cross-section of geoelectric time series data can provide reliable intermediate-term probabilistic earthquake forecasting in Taiwan
Sensitivity analysis of input ground motion on surface motion parameters in high seismic regions: a case of Bhutan Himalaya
Earthquake-induced landslide monitoring and survey by means of InSAR
Ground motion variability in Israel from 3-D simulations of M 6 and M 7 earthquakes
Ground motion prediction maps using seismic-microzonation data and machine learning
A sanity check for earthquake recurrence models used in PSHA of slowly deforming regions: the case of SW Iberia
Development of a seismic site-response zonation map for the Netherlands
Characterization of fault plane and coseismic slip for the 2 May 2020, Mw 6.6 Cretan Passage earthquake from tide gauge tsunami data and moment tensor solutions
Urban search and rescue (USAR) simulation system: spatial strategies for agent task allocation under uncertain conditions
Modelling earthquake rates and associated uncertainties in the Marmara Region, Turkey
Vulnerability and site effects in earthquake disasters in Armenia (Colombia) – Part 2 : Observed damage and vulnerability
Integrating macroseismic intensity distributions with a probabilistic approach: an application in Italy
Spatiotemporal heterogeneity of b values revealed by a data-driven approach for the 17 June 2019 MS 6.0 Changning earthquake sequence, Sichuan, China
A harmonised instrumental earthquake catalogue for Iceland and the northern Mid-Atlantic Ridge
A homogeneous earthquake catalogue for Turkey
Long-term magnetic anomalies and their possible relationship to the latest greater Chilean earthquakes in the context of the seismo-electromagnetic theory
Subash Ghimire, Philippe Guéguen, Adrien Pothon, and Danijel Schorlemmer
Nat. Hazards Earth Syst. Sci., 23, 3199–3218, https://doi.org/10.5194/nhess-23-3199-2023, https://doi.org/10.5194/nhess-23-3199-2023, 2023
Short summary
Short summary
This study explores the efficacy of several machine learning models for damage characterization, trained and tested on the Database of Observed Damage (DaDO) for Italian earthquakes. Reasonable damage prediction effectiveness (68 % accuracy) is observed, particularly when considering basic structural features and grouping the damage according to the traffic-light-based system used during the post-disaster period (green, yellow, and red), showing higher relevancy for rapid damage prediction.
Ekbal Hussain, Endra Gunawan, Nuraini Rahma Hanifa, and Qori'atu Zahro
Nat. Hazards Earth Syst. Sci., 23, 3185–3197, https://doi.org/10.5194/nhess-23-3185-2023, https://doi.org/10.5194/nhess-23-3185-2023, 2023
Short summary
Short summary
The earthquake potential of the Lembang Fault, located near the city of Bandung in West Java, Indonesia, is poorly understood. Bandung has a population of over 8 million people. We used satellite data to estimate the energy storage on the fault and calculate the likely size of potential future earthquakes. We use simulations to show that 1.9–2.7 million people would be exposed to high levels of ground shaking in the event of a major earthquake on the fault.
Huaiqun Zhao, Wenkai Chen, Can Zhang, and Dengjie Kang
Nat. Hazards Earth Syst. Sci., 23, 3031–3050, https://doi.org/10.5194/nhess-23-3031-2023, https://doi.org/10.5194/nhess-23-3031-2023, 2023
Short summary
Short summary
Early emergency response requires improving the utilization value of the data available in the early post-earthquake period. We proposed a method for assessing seismic intensities by analyzing early aftershock sequences using the robust locally weighted regression program. The seismic intensity map evaluated by the method can reflect the range of the hardest-hit areas and the spatial distribution of the possible property damage and casualties caused by the earthquake.
Asim M. Khawaja, Behnam Maleki Asayesh, Sebastian Hainzl, and Danijel Schorlemmer
Nat. Hazards Earth Syst. Sci., 23, 2683–2696, https://doi.org/10.5194/nhess-23-2683-2023, https://doi.org/10.5194/nhess-23-2683-2023, 2023
Short summary
Short summary
Testing of earthquake forecasts is important for model verification. Forecasts are usually spatially discretized with many equal-sized grid cells, but often few earthquakes are available for evaluation, leading to meaningless tests. Here, we propose solutions to improve the testability of earthquake forecasts and give a minimum ratio between the number of earthquakes and spatial cells for significant tests. We show applications of the proposed technique for synthetic and real case studies.
Lukas Bodenmann, Jack W. Baker, and Božidar Stojadinović
Nat. Hazards Earth Syst. Sci., 23, 2387–2402, https://doi.org/10.5194/nhess-23-2387-2023, https://doi.org/10.5194/nhess-23-2387-2023, 2023
Short summary
Short summary
Understanding spatial patterns in earthquake-induced ground motions is key for assessing the seismic risk of distributed infrastructure systems. To study such patterns, we propose a novel model that accounts for spatial proximity, as well as site and path effects, and estimate its parameters from past earthquake data by explicitly quantifying the inherent uncertainties.
Franz A. Livio, Maria F. Ferrario, Elisa Martinelli, Sahra Talamo, Alessandro M. Michetti, and Silvia Cercatillo
Nat. Hazards Earth Syst. Sci. Discuss., https://doi.org/10.5194/nhess-2023-71, https://doi.org/10.5194/nhess-2023-71, 2023
Revised manuscript accepted for NHESS
Short summary
Short summary
We here document the occurrence of an ancient earthquake occurred in the European Western Southern Alps in the VI Cent. A.D. The analysis of the effects due to earthquake shaking in Como City (N Italy) and the comparison with dated offshore landslides in the Alpine lakes allowed us to make an inference on the possible Magnitude and the location of the seismogenic source for this event.
José A. Álvarez-Gómez, Paula Herrero-Barbero, and José J. Martínez-Díaz
Nat. Hazards Earth Syst. Sci., 23, 2031–2052, https://doi.org/10.5194/nhess-23-2031-2023, https://doi.org/10.5194/nhess-23-2031-2023, 2023
Short summary
Short summary
The strike-slip Carboneras fault is one of the largest sources in the Alboran Sea, with it being one of the faster faults in the eastern Betics. The dimensions and location of the Carboneras fault imply a high seismic and tsunami threat. In this work, we present tsunami simulations from sources generated with physics-based earthquake simulators. We show that the Carboneras fault has the capacity to generate locally damaging tsunamis with inter-event times between 2000 and 6000 years.
Antonio Posadas, Denisse Pasten, Eugenio E. Vogel, and Gonzalo Saravia
Nat. Hazards Earth Syst. Sci., 23, 1911–1920, https://doi.org/10.5194/nhess-23-1911-2023, https://doi.org/10.5194/nhess-23-1911-2023, 2023
Short summary
Short summary
In this paper we understand an earthquake from a thermodynamics point of view as an irreversible transition; then it must suppose an increase in entropy. We use > 100 000 earthquakes in northern Chile to test the theory that Shannon entropy, H, is an indicator of the equilibrium state. Using variation in H, we were able to detect major earthquakes and their foreshocks and aftershocks, including the 2007 Mw 7.8 Tocopilla earthquake and 2014 Mw 8.1 Iquique earthquake.
Dirsa Feliciano, Orlando Arroyo, Tamara Cabrera, Diana Contreras, Jairo Andrés Valcárcel Torres, and Juan Camilo Gómez Zapata
Nat. Hazards Earth Syst. Sci., 23, 1863–1890, https://doi.org/10.5194/nhess-23-1863-2023, https://doi.org/10.5194/nhess-23-1863-2023, 2023
Short summary
Short summary
This article presents the number of damaged buildings and estimates the economic losses from a set of earthquakes in Sabana Centro, a region of 11 towns in Colombia.
Andrea Antonucci, Andrea Rovida, Vera D'Amico, and Dario Albarello
Nat. Hazards Earth Syst. Sci., 23, 1805–1816, https://doi.org/10.5194/nhess-23-1805-2023, https://doi.org/10.5194/nhess-23-1805-2023, 2023
Short summary
Short summary
The earthquake effects undocumented at 228 Italian localities were calculated through a probabilistic approach starting from the values obtained through the use of an intensity prediction equation, taking into account the intensity data documented at close localities for a given earthquake. The results showed some geographical dependencies and correlations with the intensity levels investigated.
Yi-Ying Wen, Chien-Chih Chen, Strong Wen, and Wei-Tsen Lu
Nat. Hazards Earth Syst. Sci., 23, 1835–1846, https://doi.org/10.5194/nhess-23-1835-2023, https://doi.org/10.5194/nhess-23-1835-2023, 2023
Short summary
Short summary
Knowing the spatiotemporal seismicity patterns prior to impending large earthquakes might help earthquake hazard assessment. Several recent moderate earthquakes occurred in the various regions of Taiwan, which help to further investigate the spatiotemporal seismic pattern related to the regional tectonic stress. We should pay attention when a seismicity decrease of 2.5 < M < 4.5 events around the southern Central Range or an accelerating seismicity of 3 < M < 5 events appears in central Taiwan.
Luca Schilirò, Mauro Rossi, Federica Polpetta, Federica Fiorucci, Carolina Fortunato, and Paola Reichenbach
Nat. Hazards Earth Syst. Sci., 23, 1789–1804, https://doi.org/10.5194/nhess-23-1789-2023, https://doi.org/10.5194/nhess-23-1789-2023, 2023
Short summary
Short summary
We present a database of the main scientific articles published on earthquake-triggered landslides in the last 4 decades. To enhance data viewing, the articles were catalogued into a web-based GIS, which was specifically designed to show different types of information, such as bibliometric information, the relevant topic and sub-topic category (or categories), and earthquake(s) addressed. Such information can be useful to obtain a general overview of the topic, especially for a broad readership.
Simone Barani, Gabriele Ferretti, and Davide Scafidi
Nat. Hazards Earth Syst. Sci., 23, 1685–1698, https://doi.org/10.5194/nhess-23-1685-2023, https://doi.org/10.5194/nhess-23-1685-2023, 2023
Short summary
Short summary
In the present study, we analyze ground-motion hazard maps and hazard disaggregation in order to define areas in Italy where liquefaction triggering due to seismic activity can not be excluded. The final result is a screening map for all of Italy that classifies sites in terms of liquefaction triggering potential according to their seismic hazard level. The map and the associated data are freely accessible at the following web address: www.distav.unige.it/rsni/milq.php.
Midhat Fayaz, Shakil A. Romshoo, Irfan Rashid, and Rakesh Chandra
Nat. Hazards Earth Syst. Sci., 23, 1593–1611, https://doi.org/10.5194/nhess-23-1593-2023, https://doi.org/10.5194/nhess-23-1593-2023, 2023
Short summary
Short summary
Earthquakes cause immense loss of lives and damage to properties, particularly in major urban centres. The city of Srinagar, which houses around 1.5 million people, is susceptible to high seismic hazards due to its peculiar geological setting, urban setting, demographic profile, and tectonic setting. Keeping in view all of these factors, the present study investigates the earthquake vulnerability of buildings in Srinagar, an urban city in the northwestern Himalayas, India.
Mathilde B. Sørensen, Torbjørn Haga, and Atle Nesje
Nat. Hazards Earth Syst. Sci., 23, 1577–1592, https://doi.org/10.5194/nhess-23-1577-2023, https://doi.org/10.5194/nhess-23-1577-2023, 2023
Short summary
Short summary
Most Norwegian landslides are triggered by rain or snowmelt, and earthquakes have not been considered a relevant trigger mechanism even though some cases have been reported. Here we systematically search historical documents and databases and find 22 landslides induced by eight large Norwegian earthquakes. The Norwegian earthquakes induce landslides at distances and over areas that are much larger than those found for global datasets.
Chiara Varone, Gianluca Carbone, Anna Baris, Maria Chiara Caciolli, Stefania Fabozzi, Carolina Fortunato, Iolanda Gaudiosi, Silvia Giallini, Marco Mancini, Luca Paolella, Maurizio Simionato, Pietro Sirianni, Rose Line Spacagna, Francesco Stigliano, Daniel Tentori, Luca Martelli, Giuseppe Modoni, and Massimiliano Moscatelli
Nat. Hazards Earth Syst. Sci., 23, 1371–1382, https://doi.org/10.5194/nhess-23-1371-2023, https://doi.org/10.5194/nhess-23-1371-2023, 2023
Short summary
Short summary
In 2012, Italy was struck by a seismic crisis characterized by two main shocks and relevant liquefaction events. Terre del Reno is one of the municipalities that experienced the most extensive liquefaction effects; thus it was chosen as case study for a project devoted to defining a new methodology to assess the liquefaction susceptibility. In this framework, about 1800 geotechnical, geophysical, and hydrogeological investigations were collected and stored in the publicly available PERL dataset.
Sasan Motaghed, Mozhgan Khazaee, Nasrollah Eftekhari, and Mohammad Mohammadi
Nat. Hazards Earth Syst. Sci., 23, 1117–1124, https://doi.org/10.5194/nhess-23-1117-2023, https://doi.org/10.5194/nhess-23-1117-2023, 2023
Short summary
Short summary
We modify the probabilistic seismic hazard analysis (PSHA) formulation by replacing the Gutenberg–Richter power law with the SCP (Sotolongo-Costa and Posadas) non-extensive model for earthquake size distribution and call it NEPSHA. The proposed method (NEPSHA) is implemented in the Tehran region, and the results are compared with the classic PSHA method. The hazard curves show that NEPSHA gives a higher hazard, especially in the range of practical return periods.
Paola Sbarra, Pierfrancesco Burrato, Valerio De Rubeis, Patrizia Tosi, Gianluca Valensise, Roberto Vallone, and Paola Vannoli
Nat. Hazards Earth Syst. Sci., 23, 1007–1028, https://doi.org/10.5194/nhess-23-1007-2023, https://doi.org/10.5194/nhess-23-1007-2023, 2023
Short summary
Short summary
Earthquakes are fundamental for understanding how the earth works and for assessing seismic risk. We can easily measure the magnitude and depth of today's earthquakes, but can we also do it for pre-instrumental ones? We did it by analyzing the decay of earthquake effects (on buildings, people, and objects) with epicentral distance. Our results may help derive data that would be impossible to obtain otherwise, for any country where the earthquake history extends for centuries, such as Italy.
Haekal A. Haridhi, Bor Shouh Huang, Kuo Liang Wen, Arif Mirza, Syamsul Rizal, Syahrul Purnawan, Ilham Fajri, Frauke Klingelhoefer, Char Shine Liu, Chao Shing Lee, Crispen R. Wilson, Tso-Ren Wu, Ichsan Setiawan, and Van Bang Phung
Nat. Hazards Earth Syst. Sci., 23, 507–523, https://doi.org/10.5194/nhess-23-507-2023, https://doi.org/10.5194/nhess-23-507-2023, 2023
Short summary
Short summary
Near the northern end of Sumatra, the horizontal movement Sumatran fault zone extended to its northern offshore. The movement of offshore fault segments trigger submarine landslides and induce tsunamis. Scenarios of a significant tsunami caused by the combined effect of an earthquake and its triggered submarine landslide at the coast were proposed in this study. Based on our finding, the landslide tsunami hazard assessment and early warning systems in this region should be urgently considered.
Lixin Wu, Yuan Qi, Wenfei Mao, Jingchen Lu, Yifan Ding, Boqi Peng, and Busheng Xie
Nat. Hazards Earth Syst. Sci., 23, 231–249, https://doi.org/10.5194/nhess-23-231-2023, https://doi.org/10.5194/nhess-23-231-2023, 2023
Short summary
Short summary
Multiple seismic anomalies were reported to be related to the 2015 Nepal earthquake. By sufficiently investigating both the space–time features and the physical models of the seismic anomalies, the coupling mechanisms of these anomalies in 3D space were revealed and an integrated framework to strictly root the sources of various anomalies was proposed. This study provides a practical solution for scrutinizing reliable seismic anomalies from diversified earthquake observations.
David Montiel-López, Sergio Molina, Juan José Galiana-Merino, and Igor Gómez
Nat. Hazards Earth Syst. Sci., 23, 91–106, https://doi.org/10.5194/nhess-23-91-2023, https://doi.org/10.5194/nhess-23-91-2023, 2023
Short summary
Short summary
One of the most effective ways to describe the seismicity of a region is to map the b-value parameter of the Gutenberg-Richter law. This research proposes the study of the spatial cell-event distance distribution to define the smoothing kernel that controls the influence of the data. The results of this methodology depict tectonic stress changes before and after intense earthquakes happen, so it could enable operational earthquake forecasting (OEF) and tectonic source profiling.
Pierre Henry, M. Sinan Özeren, Nurettin Yakupoğlu, Ziyadin Çakir, Emmanuel de Saint-Léger, Olivier Desprez de Gésincourt, Anders Tengberg, Cristele Chevalier, Christos Papoutsellis, Nazmi Postacıoğlu, Uğur Dogan, Hayrullah Karabulut, Gülsen Uçarkuş, and M. Namık Çağatay
Nat. Hazards Earth Syst. Sci., 22, 3939–3956, https://doi.org/10.5194/nhess-22-3939-2022, https://doi.org/10.5194/nhess-22-3939-2022, 2022
Short summary
Short summary
Seafloor instruments at the bottom of the Sea of Marmara recorded disturbances caused by earthquakes, addressing the minimum magnitude that may be recorded in the sediment. A magnitude 4.7 earthquake caused turbidity but little current. A magnitude 5.8 earthquake caused a mudflow and strong currents that spread sediment on the seafloor over several kilometers. However, most known earthquake deposits in the Sea of Marmara spread over larger zones and should correspond to larger earthquakes.
Nicola Alessandro Pino
Nat. Hazards Earth Syst. Sci., 22, 3787–3792, https://doi.org/10.5194/nhess-22-3787-2022, https://doi.org/10.5194/nhess-22-3787-2022, 2022
Short summary
Short summary
The 1908 Messina Straits earthquake is one of the most severe seismic catastrophes in human history and is periodically back in the public discussion because of a project of building a bridge across the Straits. Some models proposed for the fault assume precursory subsidence preceding the quake, resulting in a structure significantly different from the previously debated ones and important hazard implications. The analysis of the historical sea level data allows the rejection of this hypothesis.
Simone Francesco Fornasari, Deniz Ertuncay, and Giovanni Costa
Nat. Hazards Earth Syst. Sci. Discuss., https://doi.org/10.5194/nhess-2022-258, https://doi.org/10.5194/nhess-2022-258, 2022
Revised manuscript accepted for NHESS
Short summary
Short summary
We analyzed the background seismic noise for the Italian strong motion network. Several stations located near urban areas are affected by human activities and have high noise levels in the low periods. As such, noise levels show clear daily and weekly patterns and reductions during the COVID-19 lockdown. Moreover, we found slight seasonal variations. Our results provide an overview of the background noise of the network and can be used as a station selection criterion for future research.
Jack N. Williams, Luke N. J. Wedmore, Åke Fagereng, Maximilian J. Werner, Hassan Mdala, Donna J. Shillington, Christopher A. Scholz, Folarin Kolawole, Lachlan J. M. Wright, Juliet Biggs, Zuze Dulanya, Felix Mphepo, and Patrick Chindandali
Nat. Hazards Earth Syst. Sci., 22, 3607–3639, https://doi.org/10.5194/nhess-22-3607-2022, https://doi.org/10.5194/nhess-22-3607-2022, 2022
Short summary
Short summary
We use geologic and GPS data to constrain the magnitude and frequency of earthquakes that occur along active faults in Malawi. These faults slip in earthquakes as the tectonic plates on either side of the East African Rift in Malawi diverge. Low divergence rates (0.5–1.5 mm yr) and long faults (5–200 km) imply that earthquakes along these faults are rare (once every 1000–10 000 years) but could have high magnitudes (M 7–8). These data can be used to assess seismic risk in Malawi.
Mohamadreza Hosseini and Habib Rahimi
Nat. Hazards Earth Syst. Sci., 22, 3571–3583, https://doi.org/10.5194/nhess-22-3571-2022, https://doi.org/10.5194/nhess-22-3571-2022, 2022
Short summary
Short summary
Earthquakes, not only because of earth-shaking but also because of surface ruptures, are a serious threat to many human activities. Reducing earthquake losses and damage requires predicting the amplitude and location of ground movements and possible surface displacements in the future. Using the probabilistic approach and earthquake method, the surface displacement of the north Tabriz fault has been investigated, and the possible displacement in different scenarios has been estimated.
Maria Francesca Ferrario
Nat. Hazards Earth Syst. Sci., 22, 3527–3542, https://doi.org/10.5194/nhess-22-3527-2022, https://doi.org/10.5194/nhess-22-3527-2022, 2022
Short summary
Short summary
I mapped over 5000 landslides triggered by a moment magnitude 6.0 earthquake that occurred in 2015 in the Sabah region (Malaysia). I analyzed their number, dimension and spatial distribution by dividing the territory into 1 km2 cells. I applied the Environmental Seismic Intensity (ESI-07) scale, which allows the categorization of earthquake damage due to environmental effects. The presented approach promotes the collaboration among the experts in landslide mapping and in ESI-07 assignment.
Kirsty Bayliss, Mark Naylor, Farnaz Kamranzad, and Ian Main
Nat. Hazards Earth Syst. Sci., 22, 3231–3246, https://doi.org/10.5194/nhess-22-3231-2022, https://doi.org/10.5194/nhess-22-3231-2022, 2022
Short summary
Short summary
We develop probabilistic earthquake forecasts that include different spatial information (e.g. fault locations, strain rate) using a point process method. The performance of these models is tested over three different periods and compared with existing forecasts. We find that our models perform well, with those using simulated catalogues that make use of uncertainty in model parameters performing better, demonstrating potential to improve earthquake forecasting using Bayesian approaches.
Francesco Visini, Carlo Meletti, Andrea Rovida, Vera D'Amico, Bruno Pace, and Silvia Pondrelli
Nat. Hazards Earth Syst. Sci., 22, 2807–2827, https://doi.org/10.5194/nhess-22-2807-2022, https://doi.org/10.5194/nhess-22-2807-2022, 2022
Short summary
Short summary
As new data are collected, seismic hazard models can be updated and improved. In the framework of a project aimed to update the Italian seismic hazard model, we proposed a model based on the definition and parametrization of area sources. Using geological data, seismicity and other geophysical constraints, we delineated three-dimensional boundaries and activity rates of a seismotectonic zoning and explored the epistemic uncertainty by means of a logic-tree approach.
Mariana Belferman, Amotz Agnon, Regina Katsman, and Zvi Ben-Avraham
Nat. Hazards Earth Syst. Sci., 22, 2553–2565, https://doi.org/10.5194/nhess-22-2553-2022, https://doi.org/10.5194/nhess-22-2553-2022, 2022
Short summary
Short summary
Internal fluid pressure in pores leads to breaking. With this mechanical principle and a correlation between historical water level changes and seismicity, we explore possible variants for water level reconstruction in the Dead Sea basin. Using the best-correlated variant, an additional indication is established regarding the location of historical earthquakes. This leads us to propose a certain forecast for the next earthquake in view of the fast and persistent dropping level of the Dead Sea.
Xuezhong Chen, Yane Li, and Lijuan Chen
Nat. Hazards Earth Syst. Sci., 22, 2543–2551, https://doi.org/10.5194/nhess-22-2543-2022, https://doi.org/10.5194/nhess-22-2543-2022, 2022
Short summary
Short summary
When the tectonic stress in the crust increases, the b value will decrease, meaning the effects of tidal stresses are enhanced gradually. Increase in the tidal Coulomb failure stress might promote the occurrence of earthquakes, whereas its decrease could have an opposite effect. This observation may provide an insight into the processes leading to the Wenchuan earthquake and its precursors.
Fabrizio Marra, Alberto Frepoli, Dario Gioia, Marcello Schiattarella, Andrea Tertulliani, Monica Bini, Gaetano De Luca, and Marco Luppichini
Nat. Hazards Earth Syst. Sci., 22, 2445–2457, https://doi.org/10.5194/nhess-22-2445-2022, https://doi.org/10.5194/nhess-22-2445-2022, 2022
Short summary
Short summary
Through the analysis of the morphostructural setting in which the seismicity of Rome is framed, we explain why the city should not expect to suffer damage from a big earthquake.
Chengjun Feng, Guangliang Gao, Shihuai Zhang, Dongsheng Sun, Siyu Zhu, Chengxuan Tan, and Xiaodong Ma
Nat. Hazards Earth Syst. Sci., 22, 2257–2287, https://doi.org/10.5194/nhess-22-2257-2022, https://doi.org/10.5194/nhess-22-2257-2022, 2022
Short summary
Short summary
We show how FSP (Fault Slip Potential) software can be used in quantitative screening to estimate the fault slip potential in a region with some uncertainties in the ambient stress field and to assess the reactivation potential on these faults of presumably higher criticality in response to fluid injection. The case study of the Matouying enhanced geothermal system (EGS) field has important implications for deep geothermal exploitation in China, especially for the Gonghe EGS in Qinghai Province.
Patrick Oswald, Michael Strasser, Jens Skapski, and Jasper Moernaut
Nat. Hazards Earth Syst. Sci., 22, 2057–2079, https://doi.org/10.5194/nhess-22-2057-2022, https://doi.org/10.5194/nhess-22-2057-2022, 2022
Short summary
Short summary
This study provides the first regional earthquake catalogue of the eastern Alps spanning 16 000 years by using three lake paleoseismic records. Recurrence statistics reveal that earthquakes recur every 1000–2000 years in an aperiodic pattern. The magnitudes of paleo-earthquakes exceed the historically documented values. This study estimates magnitude and source areas for severe paleo-earthquakes, and their shaking effects are explored in the broader study area.
Haoyu Wen, Hong-Jia Chen, Chien-Chih Chen, Massimo Pica Ciamarra, and Siew Ann Cheong
Nat. Hazards Earth Syst. Sci., 22, 1931–1954, https://doi.org/10.5194/nhess-22-1931-2022, https://doi.org/10.5194/nhess-22-1931-2022, 2022
Short summary
Short summary
Recently, there has been growing interest from earth scientists to use the electric field deep underground to forecast earthquakes. We go one step further by using the electric fields, which can be directly measured, to separate/classify time periods with two labels only according to the statistical properties of the electric fields. By checking against historical earthquake records, we found time periods covered by one of the two labels to have significantly more frequent earthquakes.
Karma Tempa, Komal Raj Aryal, Nimesh Chettri, Giovanni Forte, and Dipendra Gautam
Nat. Hazards Earth Syst. Sci., 22, 1893–1909, https://doi.org/10.5194/nhess-22-1893-2022, https://doi.org/10.5194/nhess-22-1893-2022, 2022
Short summary
Short summary
This paper performs site response analysis and studies soil amplification for Bhutan Himalaya. A sensitivity study is performed to assess the effect of variation in strong ground motion.
Tayeb Smail, Mohamed Abed, Ahmed Mebarki, and Milan Lazecky
Nat. Hazards Earth Syst. Sci., 22, 1609–1625, https://doi.org/10.5194/nhess-22-1609-2022, https://doi.org/10.5194/nhess-22-1609-2022, 2022
Short summary
Short summary
The Sentinel-1 SAR datasets and Sentinel-2 data are used in this study to investigate the impact of natural hazards (earthquakes and landslides) on struck areas. In InSAR processing, the use of DInSAR, CCD methods, and the LiCSBAS tool permit generation of time-series analysis of ground changes. Three land failures were detected in the study area. CCD is suitable to map landslides that may remain undetected using DInSAR. In Grarem, the failure rim is clear in coherence and phase maps.
Jonatan Glehman and Michael Tsesarsky
Nat. Hazards Earth Syst. Sci., 22, 1451–1467, https://doi.org/10.5194/nhess-22-1451-2022, https://doi.org/10.5194/nhess-22-1451-2022, 2022
Short summary
Short summary
Due to an insufficient number of recorded moderate–strong earthquakes in Israel, estimating the ground motions and the subsequent seismic hazard mitigation becomes a challenge. To fill this gap, we performed a series of 3-D numerical simulations of moderate and moderate–strong earthquakes. We examined the ground motions and their variability through a self-developed statistical model. However, the model cannot fully capture the ground motion variability due to the local seismotectonic setting.
Federico Mori, Amerigo Mendicelli, Gaetano Falcone, Gianluca Acunzo, Rose Line Spacagna, Giuseppe Naso, and Massimiliano Moscatelli
Nat. Hazards Earth Syst. Sci., 22, 947–966, https://doi.org/10.5194/nhess-22-947-2022, https://doi.org/10.5194/nhess-22-947-2022, 2022
Short summary
Short summary
This work addresses the problem of the ground motion estimation over large areas as an important tool for seismic-risk reduction policies. In detail, the near-real-time estimation of ground motion is a key issue for emergency system management. Starting from this consideration, the present work proposes the application of a machine learning approach to produce ground motion maps, using nine input proxies. Such proxies consider seismological, geophysical, and morphological parameters.
Margarida Ramalho, Luis Matias, Marta Neres, Michele M. C. Carafa, Alexandra Carvalho, and Paula Teves-Costa
Nat. Hazards Earth Syst. Sci., 22, 117–138, https://doi.org/10.5194/nhess-22-117-2022, https://doi.org/10.5194/nhess-22-117-2022, 2022
Short summary
Short summary
Probabilistic seismic hazard assessment (PSHA) is the most common tool used to decide on the acceptable seismic risk by society and mitigation measures. In slowly deforming regions, such Iberia, the earthquake generation models (EGMs) for PSHA suffer from great uncertainty. In this work we propose two sanity tests to be applied to EGMs, comparing the EGM moment release with constrains derived from GNSS observations or neotectonic modelling. Similar tests should be part of other region studies.
Janneke van Ginkel, Elmer Ruigrok, Jan Stafleu, and Rien Herber
Nat. Hazards Earth Syst. Sci., 22, 41–63, https://doi.org/10.5194/nhess-22-41-2022, https://doi.org/10.5194/nhess-22-41-2022, 2022
Short summary
Short summary
A soft, shallow subsurface composition has the tendency to amplify earthquake waves, resulting in increased ground shaking. Therefore, this paper presents a workflow in order to obtain a map classifying the response of the subsurface based on local geology, earthquake signals, and background noise recordings for the Netherlands. The resulting map can be used as a first assessment in regions with earthquake hazard potential by mining or geothermal energy activities, for example.
Enrico Baglione, Stefano Lorito, Alessio Piatanesi, Fabrizio Romano, Roberto Basili, Beatriz Brizuela, Roberto Tonini, Manuela Volpe, Hafize Basak Bayraktar, and Alessandro Amato
Nat. Hazards Earth Syst. Sci., 21, 3713–3730, https://doi.org/10.5194/nhess-21-3713-2021, https://doi.org/10.5194/nhess-21-3713-2021, 2021
Short summary
Short summary
We investigated the seismic fault structure and the rupture characteristics of the MW 6.6, 2 May 2020, Cretan Passage earthquake through tsunami data inverse modelling. Our results suggest a shallow crustal event with a reverse mechanism within the accretionary wedge rather than on the Hellenic Arc subduction interface. The study identifies two possible ruptures: a steeply sloping reverse splay fault and a back-thrust rupture dipping south, with a more prominent dip angle.
Navid Hooshangi, Ali Asghar Alesheikh, Mahdi Panahi, and Saro Lee
Nat. Hazards Earth Syst. Sci., 21, 3449–3463, https://doi.org/10.5194/nhess-21-3449-2021, https://doi.org/10.5194/nhess-21-3449-2021, 2021
Short summary
Short summary
Task allocation under uncertain conditions is a key problem for agents attempting to achieve harmony in disaster environments. This paper presents an agent-based simulation to investigate task allocation considering appropriate spatial strategies to manage uncertainty in urban search and rescue (USAR) operations.
Thomas Chartier, Oona Scotti, Hélène Lyon-Caen, Keith Richard-Dinger, James H. Dieterich, and Bruce E. Shaw
Nat. Hazards Earth Syst. Sci., 21, 2733–2751, https://doi.org/10.5194/nhess-21-2733-2021, https://doi.org/10.5194/nhess-21-2733-2021, 2021
Short summary
Short summary
In order to evaluate the seismic risk, we first model the annual rate of occurrence of earthquakes on the faults near Istanbul. By using a novel modelling approach, we consider the fault system as a whole rather than each fault individually. We explore the hypotheses that are discussed in the scientific community concerning this fault system and compare the modelled results with local recorded data and a physics-based model, gaining new insights in particular on the largest possible earthquake.
Francisco J. Chávez-García, Hugo Monsalve-Jaramillo, and Joaquín Vila-Ortega
Nat. Hazards Earth Syst. Sci., 21, 2345–2354, https://doi.org/10.5194/nhess-21-2345-2021, https://doi.org/10.5194/nhess-21-2345-2021, 2021
Short summary
Short summary
We analyze earthquake damage observed in Armenia, Colombia, during the 1999 event. We investigate the reasons behind the damage and the possibility of predicting it using vulnerability studies. We show that vulnerability was a major factor and that observed damage was predicted by a vulnerability study made in 1993, which sadly had no societal impact. The comparison between two vulnerability studies, in 1993 and 2004, suggests that Armenia may still be highly vulnerable to future earthquakes.
Andrea Antonucci, Andrea Rovida, Vera D'Amico, and Dario Albarello
Nat. Hazards Earth Syst. Sci., 21, 2299–2311, https://doi.org/10.5194/nhess-21-2299-2021, https://doi.org/10.5194/nhess-21-2299-2021, 2021
Short summary
Short summary
We present a probabilistic approach for integrating incomplete intensity distributions by means of the Bayesian combination of estimates provided by intensity prediction equations (IPEs) and data documented at nearby localities, accounting for the relevant uncertainties. The performance of the proposed methodology is tested at 28 Italian localities with long and rich seismic histories and for the strong 1980 and 2009 earthquakes in Italy. An application of this approach is also illustrated.
Changsheng Jiang, Libo Han, Feng Long, Guijuan Lai, Fengling Yin, Jinmeng Bi, and Zhengya Si
Nat. Hazards Earth Syst. Sci., 21, 2233–2244, https://doi.org/10.5194/nhess-21-2233-2021, https://doi.org/10.5194/nhess-21-2233-2021, 2021
Short summary
Short summary
The b value is a controversial parameter that has the potential to identify the location of an upcoming strong earthquake. We conducted a case study using a newly developed algorithm that can overcome the subjectivity of calculation. The results confirmed the scientific significance of the b value for seismic hazard analysis and revealed that fluid intrusion may have been the cause of the overactive aftershocks of the studied earthquake.
Kristján Jónasson, Bjarni Bessason, Ásdís Helgadóttir, Páll Einarsson, Gunnar B. Guðmundsson, Bryndís Brandsdóttir, Kristín S. Vogfjörd, and Kristín Jónsdóttir
Nat. Hazards Earth Syst. Sci., 21, 2197–2214, https://doi.org/10.5194/nhess-21-2197-2021, https://doi.org/10.5194/nhess-21-2197-2021, 2021
Short summary
Short summary
Local information on epicentres and Mw magnitudes from international catalogues have been combined to compile a catalogue of earthquakes in and near Iceland in the years 1900–2019. The magnitudes are either moment-tensor modelled or proxy values obtained with regression on Ms or exceptionally on mb. The catalogue also covers the northern Mid-Atlantic Ridge with less accurate locations but similarly harmonised magnitudes.
Onur Tan
Nat. Hazards Earth Syst. Sci., 21, 2059–2073, https://doi.org/10.5194/nhess-21-2059-2021, https://doi.org/10.5194/nhess-21-2059-2021, 2021
Short summary
Short summary
Turkey is one of the most seismically active regions. In this study, an extended and homogenized earthquake catalogue, which is essential for seismic hazard studies, is presented in an easily manageable format for a wide range of researchers in earth sciences. It is the most comprehensive catalogue for Turkey and contains approximately ~ 378 000 events between 1900 and 2018.
Enrique Guillermo Cordaro, Patricio Venegas-Aravena, and David Laroze
Nat. Hazards Earth Syst. Sci., 21, 1785–1806, https://doi.org/10.5194/nhess-21-1785-2021, https://doi.org/10.5194/nhess-21-1785-2021, 2021
Short summary
Short summary
We developed a methodology that generates free externally disturbed magnetic variations in ground magnetometers close to the Chilean convergent margin. Spectral analysis (~ mHz) and magnetic anomalies increased prior to large Chilean earthquakes (Maule 2010, Mw 8.8; Iquique 2014, Mw 8.2; Illapel 2015, Mw 8.3). These findings relate to microcracks within the lithosphere due to stress state changes. This physical evidence should be thought of as a last stage of the earthquake preparation process.
Cited articles
Atkinson, J., Salmond, C., and Crampton, P.: NZDep2018 Index of Deprivation, Final Research Report, Final Research Report, University of Otago, Wellington, New Zealand, 1–65, https://www.otago.ac.nz/wellington/otago823833.pdf (last access: 4 March 2023), 2020. a
Bellagamba, X., Lee, R., and Bradley, B. A.: A neural network for automated quality screening of ground motion records from small magnitude earthquakes, Earthq. Spectra, 35, 1637–1661, https://doi.org/10.1193/122118EQS292M, 2019. a
Burkov, A.: Machine Learning Engineering, Vol. 1, True Positive Inc., ISBN 10: 1999579577/ISBN 13: 9781999579579, 2020. a
Cousins, J. and McVerry, G. H.: Overview of strong-motion data from the Darfield earthquake, Bulletin of the New Zealand Society for Earthquake Engineering, 43, 222–227, https://doi.org/10.5459/bnzsee.43.4.222-227, 2010. a
Cubrinovski, M., Green, R. A., Allen, J., Ashford, S., Bowman, E., Bradley, B., Cox, B., Hutchinson, T., Kavazanjian, E., Orense, R., Pender, M., Quigley, M., and Wotherspoon, L.: Geotechnical reconnaissance of the 2010 Darfield (Canterbury) earthquake, Bulletin of the New Zealand Society for Earthquake Engineering, 43, 243–320, https://doi.org/10.5459/bnzsee.43.4.243-320, 2010. a
Cubrinovski, M., Bradley, B., Wotherspoon, L., Green, R., Bray, J., Wood, C., Pender, M., Allen, J., Bradshaw, A., Rix, G., Taylor, M., Robinson, K., Henderson, D., Giorgini, S., Ma, K., Winkley, A., Zupan, J., O'Rourke, T., DePascale, G., and Wells, D.: Geotechnical aspects of the 22 February 2011 Christchurch earthquake, Bulletin of the New Zealand Society for Earthquake Engineering, 44, 205–226, https://doi.org/10.5459/bnzsee.44.4.205-226, 2011. a
Du, M., Liu, N., and Hu, X.: Techniques for interpretable machine learning, Commun. ACM, 63, 68–77, https://doi.org/10.1145/3359786, 2020. a
Earthquake Commission (EQC): Briefing to the Public Inquiry into the Earthquake Commission: Canterbury Home Repair Programme, Tech. Rep., PIES_010.1, 121 pp., EQC, Wellington, New Zealand, https://www.eqc.govt.nz/assets/Publications-Resources/7-v3.-Canterbury-Home-Repair-Programme-Briefing-rs.pdf (last access: 15 November 2022), 2019. a
Earthquake Commission (EQC), Ministry of Business Innovation and Employment
(MBIE), and New Zealand Government: New Zealand Geotechnical Database (NZGD), https://www.nzgd.org.nz/Default.aspx (last access: 8 November 2022), 2012. a
Feltham, C.: Insurance and reinsurance issues after the Canterbury earthquakes, Parliamentary Library Reasearch Paper, 1–2, https://www.parliament.nz/resource/en-NZ/00PlibCIP161/e9e4aba0e5a8b0eccf517d16183735b2f3c871a5 (last access: 4 March 2023), 2011. a
GeoNet: GeoNet Strong Motion Data Tool, https://strongmotion.geonet.org.nz/ (last access: 20 June 2021), 2012. a
Géron, A.: Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, O'Reilly Media, 3rd edn., ISBN: 9781098122461, 1098122461, 2022. a
Ghimire, S., Guéguen, P., Giffard-Roisin, S., and Schorlemmer, D.: Testing machine learning models for seismic damage prediction at a regional scale using building-damage dataset compiled after the 2015 Gorkha Nepal earthquake, Earthq. Spectra, 38, 2970–2993, https://doi.org/10.1177/87552930221106495, 2022. a
Harirchian, E., Kumari, V., Jadhav, K., Rasulzade, S., Lahmer, T., and Das, R. R.: A synthesized study based on machine learning approaches for rapid classifying earthquake damage grades to rc buildings, Appl. Sci.-Basel, 11, 7540, https://doi.org/10.3390/app11167540, 2021. a
Hastie, T., Tibshirani, R., and Friedman, J.: The Elements of Statistical Learning, Springer Series in Statistics, Springer, New York, https://doi.org/10.1007/978-0-387-84858-7, 2009. a
Honegger, M.: Shedding Light on Black Box Machine Learning Algorithms, PhD
thesis, Karlsruhe Institue of Technology, Germany, arXiv [preprint], https://doi.org/10.48550/arXiv.1808.05054, 15 August 2018. a
Kaiser, A., Van Houtte, C., Perrin, N., Wotherspoon, L., and Mcverry, G.: Site Characterisation of GeoNet Stations for the New Zealand Strong Motion Database, Bulletin of the New Zealand Society for Earthquake Engineering, 50, 39–49, https://doi.org/10.5459/bnzsee.50.1.39-49, 2017. a
Kalakonas, P. and Silva, V.: Earthquake scenarios for building portfolios using artificial neural networks: part II—damage and loss assessment, B. Earthq. Eng., https://doi.org/10.1007/s10518-022-01599-2, 2022a a
Kalakonas, P. and Silva, V.: Seismic vulnerability modelling of building portfolios using artificial neural networks, Earthq. Eng. Struct. D., 51, 310–327, https://doi.org/10.1002/eqe.3567, 2022b. a
Kiani, J., Camp, C., and Pezeshk, S.: On the application of machine learning techniques to derive seismic fragility curves, Comput. Struct., 218, 108–122, https://doi.org/10.1016/j.compstruc.2019.03.004, 2019. a
Kim, T., Song, J., and Kwon, O. S.: Pre- and post-earthquake regional loss assessment using deep learning, Earthq. Eng. Struct. D., 49, 657–678, https://doi.org/10.1002/eqe.3258, 2020. a
King, A., Middleton, D., Brown, C., Johnston, D., and Johal, S.: Insurance: Its Role in Recovery from the 2010–2011 Canterbury Earthquake Sequence, Earthq. Spectra, 30, 475–491, https://doi.org/10.1193/022813EQS058M, 2014. a
Land Information New Zealand (LINZ): NZ Property Titles, LINZ Data Service [data set], https://data.linz.govt.nz/layer/50804-nz-property-titles/ (last access: 5 June 2022), 2020a. a
Land Information New Zealand (LINZ): NZ Street Address, LINZ Data Service [data set], https://data.linz.govt.nz/layer/53353-nz-street-address/ (last access: 5 June 2022), 2020b. a
Land Resource Information Systems (LRIS): Soil map for the Upper Plains and Downs of Canterbury, LRIS [data set], https://doi.org/10.26060/1PGJ-JE57, 2010. a
Lemaître, G., Nogueira, F., and Aridas, C. K.: Imbalanced-learn: A Python Toolbox to Tackle the Curse of Imbalanced Datasets in Machine Learning, J. Mach. Learn. Res., 18, 1–5, http://jmlr.org/papers/v18/16-365.html (last access: 16 October 2021), 2017. a
Lundberg, S. M. and Lee, S.-I.: A Unified Approach to Interpreting Model Predictions, in: Advances in Neural Information Processing Systems 30 (NIPS 2017), edited by: Guyon, I., Von Luxburg, U., Bengio, S., Wallach, H., Fergus, R., Vishwanathan, S., and Garnett, R., Curran Associates, Inc., https://proceedings.neurips.cc/paper/2017/file/8a20a8621978632d76c43dfd28b67767-Paper.pdf (last access: 20 October 2020), 2017. a
Lundberg, S. M., Erion, G. G., and Lee, S.-I.: Consistent Individualized Feature Attribution for Tree Ensembles, 2017 ICML Workshop, arXiv [preprint], https://doi.org/10.48550/arXiv.1802.03888, 7 March 2019. a, b
Mangalathu, S. and Burton, H. V.: Deep learning-based classification of earthquake-impacted buildings using textual damage descriptions, Int. J. Disast. Risk Re., 36, 101111, https://doi.org/10.1016/j.ijdrr.2019.101111, 2019. a
Mangalathu, S. and Jeon, J.-S.: Machine Learning–Based Failure Mode Recognition of Circular Reinforced Concrete Bridge Columns: Comparative Study, J. Struct. Eng., 145, 04019104, https://doi.org/10.1061/(ASCE)ST.1943-541X.0002402, 2019. a
Mangalathu, S., Hwang, S. H., Choi, E., and Jeon, J. S.: Rapid seismic damage evaluation of bridge portfolios using machine learning techniques, Eng. Struct., 201, 109785, https://doi.org/10.1016/j.engstruct.2019.109785, 2019. a
Mangalathu, S., Sun, H., Nweke, C. C., Yi, Z., and Burton, H. V.: Classifying earthquake damage to buildings using machine learning, Earthq. Spectra, 36, 183–208, https://doi.org/10.1177/8755293019878137, 2020. a
NIWA and GNS Science: RiskScape - Asset Module Metadata, GitHub [data set], https://bit.ly/RiskScapeAssetModuleMetadata20151204 (last access: 25 November 2022), 2015. a
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E.: Scikit-learn: Machine Learning in Python, J. Mach. Learn. Res., 12, 2825–2830, https://www.jmlr.org/papers/volume12/pedregosa11a/pedregosa11a.pdf (last access: 15 March 2022), 2011. a, b, c, d
Ribeiro, M. T., Singh, S., and Guestrin, C.: “Why Should I Trust You?”: Explaining the Predictions of Any Classifier, in: Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations, San Diego, California, 1135–1144, https://doi.org/10.18653/v1/N16-3020, 2016a. a
Ribeiro, M. T., Singh, S., and Guestrin, C.: Model-Agnostic Interpretability of Machine Learning, in: 2016 ICML Workshop on Human Interpretability in Machine Learning (WHI 2016), New York, NY, arXiv [preprint], https://doi.org/10.48550/arXiv.1606.05386, 16 June 2016b. a
Ribeiro, M. T., Singh, S., and Guestrin, C.: Anchors: High-Precision Model-Agnostic Explanations, in: Proceedings of the AAAI Conference on Artificial Intelligence (AAAI'18), 32, 1527–1535, https://doi.org/10.1609/aaai.v32i1.11491, 2018. a
Roeslin, S.: Predicting Seismic Damage and Loss for Residential Buildings using Data Science, PhD thesis, University of Auckland, Auckland, New Zealand, https://hdl.handle.net/2292/57074 (last access: 18 December 2022), 2021. a
Roeslin, S., Ma, Q., Wicker, J., and Wotherspoon, L.: Data Integration for the Development of a Seismic Loss Prediction Model for Residential Buildings in New Zealand, in: Machine Learning and Knowledge Discovery in Databases, edited by: Cellier, P. and Driessens, K., Springer, Cham, Switzerland, Comm. Com. Inf. Sc., 1168, 88–100, https://doi.org/10.1007/978-3-030-43887-6_8, 2020.
a
Rogers, N., van Ballegooy, S., Williams, K., and Johnson, L.: Considering Post-Disaster Damage to Residential Building Construction - Is Our Modern Building Construction Resilient?, in: 6th International Conference on Earthquake Geotechnical Engineering (6ICEGE), Christchurch, New Zealand, 1–4 November 2015, https://www.issmge.org/publications/publication/considering-postdisaster-damage-to-residential-building-construction-is-
our-modern-building-construction-resilient
(last access: 4 March 2023), 2015. a
Russell, J. and van Ballegooy, S.: Canterbury Earthquake Sequence: Increased Liquefaction Vulnerability assessment methodology, Tonkin & Taylor Ltd, Auckland, New Zealand, Tech. rep., 0028-1-R-JICR-2015, 204 pp., https://www.eqc.govt.nz/assets/Publications-Resources/CES-
Increased-Liquefaction-Vulnerability-Assessment-Methodology
-T+T-Report.pdf
(last access: 23 April 2022), 2015. a, b
Sarker, I. H.: Machine Learning: Algorithms, Real-World Applications and Research Directions, SN Computer Science, 2, 160, https://doi.org/10.1007/s42979-021-00592-x, 2021. a
Stojadinović, Z., Kovačević, M., Marinković, D., and Stojadinović, B.: Rapid earthquake loss assessment based on machine learning and representative sampling, Earthq. Spectra, 38, 152–177, https://doi.org/10.1177/87552930211042393, 2022. a
Sun, H., Burton, H. V., and Huang, H.: Machine Learning Applications for Building Structural Design and Performance Assessment: State-of-the-Art Review, Journal of Building Engineering, 33, 101816, https://doi.org/10.1016/j.jobe.2020.101816, 2020. a
Van Houtte, C., Bannister, S., Holden, C., Bourguignon, S., and Mcverry, G.: The New Zealand strong motion database, Bulletin of the New Zealand Society for Earthquake Engineering, 50, 1–20, https://doi.org/10.5459/bnzsee.50.1.1-20, 2017. a
Wood, C. M., Cox, B. R., Wotherspoon, L. M., and Green, R. A.: Dynamic site
characterization of Christchurch strong motion stations, Bulletin of the New
Zealand Society for Earthquake Engineering, 44, 195–204,
https://doi.org/10.5459/bnzsee.44.4.195-204, 2011. a
Xie, Y., Ebad Sichani, M., Padgett, J. E., and DesRoches, R.: The promise of implementing machine learning in earthquake engineering: A state-of-the-art review, Earthq. Spectra, 36, 1769–1801, https://doi.org/10.1177/8755293020919419, 2020. a
Zhang, Y., Burton, H. V., Sun, H., and Shokrabadi, M.: A machine learning framework for assessing post-earthquake structural safety, Struct. Saf., 72, 1–16, https://doi.org/10.1016/j.strusafe.2017.12.001, 2018. a
Short summary
This paper presents a new framework for the rapid seismic loss prediction for residential buildings in Christchurch, New Zealand. The initial model was trained on insurance claims from the Canterbury earthquake sequence. Data science techniques, geospatial tools, and machine learning were used to develop the prediction model, which also delivered useful insights. The model can rapidly be updated with data from new earthquakes. It can then be applied to predict building loss in Christchurch.
This paper presents a new framework for the rapid seismic loss prediction for residential...
Special issue
Altmetrics
Final-revised paper
Preprint