Abstract

This work offers a novel methodological framework to address the problem of generating data-driven earthquake shaking fields at different vibration periods, which are key to support decision making and civil protection planning. We propose to analyse the entire profiles of spectral accelerations and project their information content to unsampled locations in the system, based on the theory of Object Oriented Spatial Statistics. The proposed methodology combines a non-ergodic ground motion model with a fully functional model for the residual term, the latter consisting of (i) the spatially-varying systematic effects due to source, site and path, and (ii) the remaining aleatory error. The proposed methodology allows to generate multiple shaking scenarios conditioned on the data, jointly and consistently for all the vibration periods, overcoming the intrinsic limitations of existing multivariate approaches to the problem. The approach is tested on a vast dataset of ground motion records collected in the study-area of the Po Plain (Northern Italy), for which a region-specific fully non-ergodic GMM was previously calibrated. Our validation tests demonstrate the potentiality of the approach, which is capable to effectively simulate spectral acceleration profiles, while keeping the ability to capture the main physical features of ground motion patterns in the region.

Highlights

  • Seismic shaking maps are tools to support decision making at a given site and a key-topic for civil protection planning and engineering purposes such as for loss assessment and risk analysis

  • In the field of engineering seismology, the effect of the earthquake shaking is generally described through empirical ground motion models (GMMs), which quantify the conditional distribution of a ground motion intensity measure IM, given a set of explanatory parameters related to a given earthquake event e and an observing site s in a study area D

  • Spectral accelerations fSAsðTÞ; T 2 T g in a GMM have been here interpreted as functional data, distributed in space

Read more

Summary

Introduction

Seismic shaking maps are tools to support decision making at a given site and a key-topic for civil protection planning and engineering purposes such as for loss assessment and risk analysis. A functional characterization of the dependence among spectral acceleration profiles is here used to provide data-driven simulations of shaking scenarios (Fig. 1), which are key for several engineering applications (e.g., Hacıefendioglu and Alpaslan 2014; Hacıefendioglu et al 2015) In this conceptual framework, the present study takes advantage of the large availability of seismic records in Northern Italy to build a novel model for the spectral profiles in the region, based on a non-ergodic GMM developed by Lanzano et al (2017) for the same area.

Basic definitions
Dataset and assumptions
Conditional simulation of the corrective term
Unconditional simulation of the error term
Simulation of the spectral acceleration
Generation of shaking fields in the study area
Spatial analysis of the corrective term
Analysis of the aleatory term s
Shaking fields
Model validation
Comparison with finite-dimensional approaches
Case UVT
Cases MVT-a and MVT-b
Application to an independent event
Conclusions and discussion
Findings
Compliance with ethical standards
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call