Paper Info
Title
Development of optimization-based probabilistic earthquake scenarios for the city of Tehran
Abstract
This paper presents the methodology and practical example for the application of optimization process to select earthquake scenarios which best represent probabilistic earthquake hazard in a given region. The method is based on simulation of a large dataset of potential earthquakes, representing the long-term seismotectonic characteristics in a given region. The simulation process uses Monte-Carlo simulation and regional seismogenic source parameters to generate a synthetic earthquake catalogue consisting of a large number of earthquakes, each characterized with magnitude, location, focal depth and fault characteristics. Such catalogue provides full distributions of events in time, space and size; however, demands large computation power when is used for risk assessment, particularly when other sources of uncertainties are involved in the process. To reduce the number of selected earthquake scenarios, a mixed-integer linear program formulation is developed in this study. This approach results in reduced set of optimization-based probabilistic earthquake scenario, while maintaining shape of hazard curves and full probabilistic picture by minimizing the error between hazard curves driven by full and reduced sets of synthetic earthquake scenarios. To test the model, the regional seismotectonic and seismogenic characteristics of northern Iran are used to simulate a set of 10,000-year worth of events consisting of some 84,000 earthquakes. The optimization model is then performed multiple times with various input data, taking into account probabilistic seismic hazard for Tehran city as the main constrains. The sensitivity of the selected scenarios to the user-specified site/return period error-weight is also assessed. The methodology could enhance run time process for full probabilistic earthquake studies like seismic hazard and risk assessment. The reduced set is the representative of the contributions of all possible earthquakes; however, it requires far less computation power. The authors have used this approach for risk assessment towards identification of effectiveness-profitability of risk mitigation measures, using optimization model for resource allocation. Based on the error-computation trade-off, 62-earthquake scenarios are chosen to be used for this purpose.
Year
DOI
Venue
2016
10.1016/j.cageo.2015.10.003
Computers & Geosciences
Keywords
Field
DocType
Optimization,Seismic hazard,Monte-Carlo,Earthquake scenarios,Tehran
Data mining,Magnitude (mathematics),Earthquake scenario,Earthquake simulation,Computer science,Return period,Incremental Dynamic Analysis,Risk management,Probabilistic logic,Seismic hazard
Journal
Volume
Issue
ISSN
86
C
0098-3004
Citations 
PageRank 
References 
3
0.59
0
Authors
2
Name
Order
Citations
PageRank
Mohammad Reza Zolfaghari161.12
E. Peyghaleh230.59