Relational Database for Horizontal-to-Vertical Spectral Ratios

Frequency-dependent horizontal-to-vertical spectral ratios (HVSRs) of Fourier amplitudes from three-component recordings can provide useful information for site response modeling. However, such information is not incorporated into most ground-motion models, including those from Next-Generation Attenuation projects, which instead use the time-averaged shear-wave velocity (VS) in the upper 30 m of the site and sediment depth terms. To facilitate utilization of HVSR, we developed a publicly accessible relational database. This database is adapted from a similar repository for VS data and provides microtremor-based HVSR data (mHVSR) and supporting metadata, but not parameters derived from the data. Users can interact with the data directly within a web portal that contains a graphical user interface (GUI) or through external tools that perform cloud-based computations. Within the database GUI, the median horizontal-component mHVSR can be plotted against frequency, with the mean and mean ± one standard deviation (representing variability across time windows) provided. Using external interactive tools (provided as a Jupyter Notebook and an R script), users can replot mHVSR (as in the database) or create polar plots. These tools can also derive parameters of potential interest for modeling purposes, including a binary variable indicating whether an mHVSR plot contains peaks, as well as the fitted properties of those peaks (frequencies, amplitudes, and widths). Metadata are also accessible, which includes site location, details about the instruments used to make the measurements, and data processing information related to windowing, antitrigger routines, and filtering.

Ranking and Selection of Earthquake Ground-Motion Models Using the Stochastic Area Metric

We introduce the cumulative-distribution-based area metric (AM)—also known as stochastic AM—as a scoring metric for earthquake ground-motion models (GMMs). The AM quantitatively informs the user of the degree to which observed or test data fit with a given model, providing a rankable absolute measure of misfit. The AM considers underlying data distributions and model uncertainties without any assumption of form. We apply this metric, along with existing testing methods, to four GMMs in order to test their performance using earthquake ground-motion data from the Preston New Road (United Kingdom) induced seismicity sequences in 2018 and 2019. An advantage of the proposed approach is its applicability to sparse datasets. We, therefore, focus on the ranking of models for discrete ranges of magnitude and distance, some of which have few data points. The variable performance of models in different ranges of the data reveals the importance of considering alternative models. We extend the ranking of GMMs through analysis of intermodel variations of the candidate models over different ranges of magnitude and distance using the AM. We find the intermodel AM can be a useful tool for selection of models for the logic-tree framework in seismic-hazard analysis. Overall, the AM is shown to be efficient and robust in the process of selection and ranking of GMMs for various applications, particularly for sparse and small-sized datasets.

Relational database used for ground-motion model development in the NGA-sub project

The Next-Generation Attenuation for subduction zone regions project (NGA-Sub) has developed data resources and ground motion models for global subduction zone regions. Here we describe the NGA-Sub database. To optimize the efficiency of data storage, access, and updating, data resources for the NGA-Sub project are organized into a relational database consisting of 20 tables containing data, metadata, and computed quantities (e.g. intensity measures, distances). A database schema relates fields in tables to each other through a series of primary and foreign keys. Model developers and other users mostly interact with the data through a flatfile generated as a time-stamped output of the database. We describe the structure of the relational database, the ground motions compiled for the project, and the means by which the data can be accessed. The database contains 71,340 three-component records from 1880 earthquakes from seven global subduction zone regions: Alaska, Central America and Mexico, Cascadia, Japan, New Zealand, South America, and Taiwan. These data were processed on a component-specific basis to minimize noise effects in the data and remove baseline drifts. Provided ground motion intensity measures include peak acceleration, peak velocity, and 5%-damped pseudo-spectral accelerations for a range of oscillator periods.

Constraints on near-source saturation models for avoiding over-saturation of response spectral ordinates in RVT-based stochastic ground-motion simulations

Inversions of empirical data and ground-motion models to find Fourier spectral parameters can result in parameter combinations that produce over-saturation of short-period response spectral ordinates. While some evidence for over-saturation in empirical data exists, most ground-motion modellers do not permit this scaling within their models. Host-to-target adjustments that are made to published ground-motion models for use in site-specific seismic hazard analyses frequently require the identification of an equivalent set of Fourier spectral parameters. In this context, when inverting response spectral models that do not exhibit over-saturation effects, it is desirable to impose constraints upon the Fourier parameters to match the scaling of the host-region model. The key parameters that determine whether over-saturation arises are the geometric spreading rate (γ) and the exponential rate within near-source saturation models (hβ). The article presents the derivation of simple nonlinear constraints that can be imposed to prevent over-saturation when undertaking Fourier spectral inversions.

New conditional, scenario-based, and non-conditional cumulative absolute velocity models for subduction tectonic settings

The PEER NGA-Sub ground-motion intensity measure database is used to develop new conditional ground-motion models (CGMMs), a set of scenario-based models, and non-conditional models to estimate the cumulative absolute velocity ([Formula: see text]) of ground motions from subduction zone earthquakes. In the CGMMs, the median estimate of [Formula: see text] is conditioned on the estimated peak ground acceleration ([Formula: see text]), the time-averaged shear-wave velocity in the top 30 m of the soil ([Formula: see text]), the earthquake magnitude ([Formula: see text]), and the spectral acceleration at the period of 1 s ([Formula: see text]). Multiple scenario-based [Formula: see text] models are developed by combining the CGMMs with pseudo-spectral acceleration ([Formula: see text]) ground-motion models (GMMs) for [Formula: see text] and [Formula: see text] to directly estimate [Formula: see text] given an earthquake scenario and site conditions. Scenario-based [Formula: see text] models are capable of capturing the complex ground-motion effects (e.g. soil non-linearity and regionalization effects) included in their underlying [Formula: see text]/[Formula: see text] GMMs. This approach also ensures the consistency of the [Formula: see text] estimates with a [Formula: see text] design spectrum. In addition, two non-conditional [Formula: see text] GMMs are developed using Bayesian hierarchical regressions. Finally, we present comparisons between the developed models. The comparisons show that if non-conditional GMMs are properly constrained, they are consistent with scenario-based GMMs. The [Formula: see text] GMMs developed in this study advance the performance-based earthquake engineering practice in areas affected by subduction zone earthquakes.

A Comparison of Nonergodic Ground-Motion Models based on Geographically Weighted Regression and the Integrated Nested Laplace Approximation

Different nonergodic Ground-Motion Models based on spatially varying coefficient models are compared for ground-motion data in Italy. The models are based different methodologies: Multi-source geographically weighted regression (Caramenti et al., 2020), and Bayesian hierarchical models estimated with the integrated nested Laplace approximation (Rue et al., 2009). The different models are compared in terms of their predictive performance, their spatial coefficients, and their predictions. Models that include spatial terms perform slightly better than a simple base model that includes only event and station terms, in terms of out-of sample error based on cross-validation. The Bayesian spatial models have slightly lower generalization error, which can be attributed to the fact that they can include random effects for events and stations. The different methodologies give rise to different dependencies of the spatially varying terms on event and station locations, leading to between-model uncertainty in their predictions, which should be accommodated in a nonergodic seismic hazard assessment.

Estimating Ground-Motion Models via Bayesian Inference using Stan

We provide a simple introduction to the estimation of ground-motion models via Bayesian inference and the probabilistic programming language Stan.We show one ca implement a simple ground-motion model in Stan, and how can run the program from the computer environment R.We also show how one can access the results, and plot summaries of estimated parameters.A large number of different Stan models for the development https://github.com/pstafford/StanGMMTutorial.