Show simple item record

dc.contributor.authorNúñez, Jorge
dc.contributor.authorCatalán, Patricio A.
dc.contributor.authorValle, Carlos
dc.contributor.authorZamora, Natalia
dc.contributor.authorValderrama, Alvaro
dc.contributor.otherBarcelona Supercomputing Center
dc.date.accessioned2022-06-28T13:53:41Z
dc.date.available2022-06-28T13:53:41Z
dc.date.issued2022-06
dc.identifier.citationNúñez, J. [et al.]. Discriminating the occurrence of inundation in tsunami early warning with one-dimensional convolutional neural networks. "Scientific Reports", Juny 2022, vol. 12, 10321.
dc.identifier.issn2045-2322
dc.identifier.urihttp://hdl.handle.net/2117/369225
dc.description.abstractTsunamis are natural phenomena that, although occasional, can have large impacts on coastal environments and settlements, especially in terms of loss of life. An accurate, detailed and timely assessment of the hazard is essential as input for mitigation strategies both in the long term and during emergencies. This goal is compounded by the high computational cost of simulating an adequate number of scenarios to make robust assessments. To reduce this handicap, alternative methods could be used. Here, an enhanced method for estimating tsunami time series using a one-dimensional convolutional neural network model (1D CNN) is considered. While the use of deep learning for this problem is not new, most of existing research has focused on assessing the capability of a network to reproduce inundation metrics extrema. However, for the context of Tsunami Early Warning, it is equally relevant to assess whether the networks can accurately predict whether inundation would occur or not, and its time series if it does. Hence, a set of 6776 scenarios with magnitudes in the range Mw 8.0–9.2 were used to design several 1D CNN models at two bays that have different hydrodynamic behavior, that would use as input inexpensive low-resolution numerical modeling of tsunami propagation to predict inundation time series at pinpoint locations. In addition, different configuration parameters were also analyzed to outline a methodology for model testing and design, that could be applied elsewhere. The results show that the network models are capable of reproducing inundation time series well, either for small or large flow depths, but also when no inundation was forecast, with minimal instances of false alarms or missed alarms. To further assess the performance, the model was tested with two past tsunamis and compared with actual inundation metrics. The results obtained are promising, and the proposed model could become a reliable alternative for the calculation of tsunami intensity measures in a faster than real time manner. This could complement existing early warning system, by means of an approximate and fast procedure that could allow simulating a larger number of scenarios within the always restricting time frame of tsunami emergencies.
dc.description.sponsorshipTide gauge data were obtained from the Sea Level Station Monitoring Facility of the Intergovernmental Oceanographic Commission (http://www.ioc-sealevelmonitoring.org/list.php). The coarser bathymetric and topographic data from the General Bathymetric Chart of the Ocean (https://www.gebco.net/data_and_products/gridded_bathymetry_data/). The authors acknowledge SHOA for providing nautical charts and coastal zone plans used to generate high resolution topo-bathymetric grids for research purposes. We are deeply grateful with A. Gubler that prepared a first version of the high resolution bathymetry grids. The authors acknowledge the computer resources at CTE-POWER (https://www.bsc.es/supportkc/docs/CTE-POWER/overview) and the technical support provided by BSC. We are greatly thankful the EDANYA Group at Málaga University for sharing the Tsunami-HySEA code. Most figures were generated with Python91,92,93 and Global Mapping Tools94. JN deeply thanks support of Mitiga Solutions during his secondment. PAC would like to thank funding by ANID, Chile Grants FONDEF ID19I10048, Centro de Investigación para la Gestión Integrada del Riesgo de Desastres (CIGIDEN) ANID/FONDAP/15110017, and Centro Científico Tecnológico de Valparaíso, ANID PIA/APOYO AFB180002. NZ has received funding from the Marie Skłodowska-Curie grant agreement H2020-MSCA-COFUND-2016-75443.
dc.format.extent20 p.
dc.language.isoeng
dc.publisherNature Research
dc.rightsAttribution 4.0 International
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subjectÀrees temàtiques de la UPC::Informàtica::Aplicacions de la informàtica::Aplicacions informàtiques a la física i l‘enginyeria
dc.subject.lcshTsunamis
dc.subject.lcshNeural networks (Computer science)
dc.subject.lcshFloods
dc.subject.otherTsunami
dc.subject.otherNeural networks
dc.subject.otherMathematical model
dc.titleDiscriminating the occurrence of inundation in tsunami early warning with one-dimensional convolutional neural networks
dc.typeArticle
dc.subject.lemacSimulació per ordinador
dc.identifier.doi10.1038/s41598-022-13788-9
dc.description.peerreviewedPeer Reviewed
dc.relation.publisherversionhttps://www.nature.com/articles/s41598-022-13788-9
dc.rights.accessOpen Access
dc.description.versionPostprint (published version)
local.citation.other10321
local.citation.publicationNameScientific Reports
local.citation.volume12


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record