Search

The global Collaboratory for the Study of Earthquake Predictability (CSEP, https://cseptesting.org) advances rigorous research into earthquake forecasting through open science practices. These include, foremost, a dedication to testing forecasts in a truly prospective manner, along with strict standards for reproducibility, community-endorsed model evaluation approaches, and open-source software. Now operating for over 15 years, CSEP recently re-invented itself with a modern community-developed open-source Python software toolkit (pyCSEP), a broader range of model classes and evaluations, and better open access to models, data and test results through reproducibility packages. This reinvention facilitated a surge in new research, driven largely by a global and engaged early career researcher community. In this overview talk, I will review the progress and new insights this has generated., The 28th IUGG General Assembly (IUGG2023) (Berlin 2023)

The Collaboratory for the Study of Earthquake Predictability (CSEP) is an open and global community whose mission is to accelerate earthquake predictability research through rigorous testing of probabilistic earthquake forecast models and prediction algorithms. pyCSEP supports this mission by providing open-source implementations of useful tools for evaluating earthquake forecasts. pyCSEP is a Python package that contains the following modules: (1) earthquake catalog access and processing, (2) representations of probabilistic earthquake forecasts, (3) statistical tests for evaluating earthquake forecasts, and (4) visualization routines and various other utilities. Most significantly, pyCSEP contains several statistical tests needed to evaluate earthquake forecasts, which can be forecasts expressed as expected earthquake rates in space–magnitude bins or specified as large sets of simulated catalogs (which includes candidate models for governmental operational earthquake forecasting). To showcase how pyCSEP can be used to evaluate earthquake forecasts, we have provided a reproducibility package that contains all the components required to re-create the figures published in this article. We recommend that interested readers work through the reproducibility package alongside this article. By providing useful tools to earthquake forecast modelers and facilitating an open-source software community, we hope to broaden the impact of the CSEP and further promote earthquake forecasting research.

SUMMARY The Regional Earthquake Likelihood Models (RELM) experiment, conducted within the Collaboratory for the Study of Earthquake Predictability (CSEP), showed that the smoothed seismicity (HKJ) model by Helmstetter et al. was the most informative time-independent earthquake model in California during the 2006–2010 evaluation period. The diversity of competing forecast hypotheses and geophysical data sets used in RELM was suitable for combining multiple models that could provide more informative earthquake forecasts than HKJ. Thus, Rhoades et al. created multiplicative hybrid models that involve the HKJ model as a baseline and one or more conjugate models. In retrospective evaluations, some hybrid models showed significant information gains over the HKJ forecast. Here, we prospectively assess the predictive skills of 16 hybrids and 6 original RELM forecasts at a 0.05 significance level, using a suite of traditional and new CSEP tests that rely on a Poisson and a binary likelihood function. In addition, we include consistency test results at a Bonferroni-adjusted significance level of 0.025 to address the problem of multiple tests. Furthermore, we compare the performance of each forecast to that of HKJ. The evaluation data set contains 40 target events recorded within the CSEP California testing region from 2011 January 1 to 2020 December 31, including the 2016 Hawthorne earthquake swarm in southwestern Nevada and the 2019 Ridgecrest sequence. Consistency test results show that most forecasting models overestimate the number of earthquakes and struggle to explain the spatial distribution of epicenters, especially in the case of seismicity clusters. The binary likelihood function significantly reduces the sensitivity of spatial log-likelihood scores to clustering, however; most models still fail to adequately describe spatial earthquake patterns. Contrary to retrospective analyses, our prospective test results show that none of the models are significantly more informative than the HKJ benchmark forecast, which we interpret to be due to temporal instabilities in the fit that forms hybrids. These results suggest that smoothing high-resolution, small earthquake data remains a robust method for forecasting moderate-to-large earthquakes over a period of 5–15 yr in California.

Operational earthquake forecasting for risk management and communication during seismic sequences depends on our ability to select an optimal forecasting model. To do this, we need to compare the performance of competing models with each other in prospective forecasting mode, and to rank their performance using a fair, reproducible and reliable method. The Collaboratory for the Study of Earthquake Predictability (CSEP) conducts such prospective earthquake forecasting experiments around the globe. One metric that has been proposed to rank competing models is the Parimutuel Gambling score, which has the advantage of allowing alarm-based (categorical) forecasts to be compared with probabilistic ones. Here we examine the suitability of this score for ranking competing earthquake forecasts. First, we prove analytically that this score is in general improper, meaning that, on average, it does not prefer the model that generated the data. Even in the special case where it is proper, we show it can still be used in an improper way. Then, we compare its performance with two commonly-used proper scores (the Brier and logarithmic scores), taking into account the uncertainty around the observed average score. We estimate the confidence intervals for the expected score difference which allows us to define if and when a model can be preferred. Our findings suggest the Parimutuel Gambling score should not be used to distinguishing between multiple competing forecasts. They also enable a more rigorous approach to distinguish between the predictive skills of candidate forecasts in addition to their rankings., Comment: 29 pages, 14 figures. Work presented at vEGU21 as vPico presentation

Classical Poissonian probabilistic seismic hazard assessment (PSHA) relies on the rate of independent mainshocks to estimate future seismic hazard levels. Real earthquakes, however, cluster in time and space. How different is hazard assessed by Poissonian PSHA from that generated by clustered seismicity? In this synthetic study, we simulate short‐ and long‐term catalogs and ground motions using a generic, temporal epidemic‐type aftershock sequences (ETAS) model assuming a single point source in a subduction zone setting, and we compare the ETAS‐simulated hazard with approximations based on the declustered Poisson approach (DP), the nondeclustered Poisson approach (NDP), and the recently proposed sequence‐based PSHA (Iervolino et al., 2014). Our comparisons reveal that the extent to which the approximations are good depends on the probability levels (return periods) of interest: (1) at (low) building design‐oriented probability (i.e., 10% or 2% in 50 yr), the approximate PSHA methods give peak ground acceleration (PGA) estimates to within ± 7% of the ETAS hazard; (2) at high probability level (>∼45% in 50 yr), the methods interestingly overestimate the ETAS hazard, a subtle consequence of the poor Poisson approximation to the skewed ETAS distributions; and (3) none of the approximate methods reproduce the multiple exceedance curves implied by the ETAS model, suggesting that cumulative seismic risk assessment with damage‐dependent fragility curves should account for ETAS‐like sequences. We also investigate the relation between short‐term (conditional) and long‐term hazard, and propose a hazard analog of the Omori–Utsu and Utsu–Seki scaling laws as a function of initial magnitude, elapsed time, and long‐term hazard. This model can provide a quick prediction of ensemble‐averaged short‐term hazard and the time needed for the hazard levels to return to the long‐term average. We conclude that realistic multigenerational earthquake clustering has both obvious and more subtle effects on long‐ and short‐term hazard, and should be considered in refined hazard assessments.

For government officials and the public to act on real-time forecasts of earthquakes, the seismological community needs to develop confidence in the underlying scientific hypotheses of the forecast generating models by assessing their predictive skill. For this purpose, the Collaboratory for the Study of Earthquake Predictability (CSEP) provides cyberinfrastructure and computational tools to evaluate earthquake forecasts. Here, we introduce pyCSEP, a Python package to help earthquake forecast developers embed model evaluation into the model development process. The package contains the following modules: (1) earthquake catalog access and processing, (2) data models for earthquake forecasts, (3) statistical tests for evaluating earthquake forecasts, and (4) visualization routines. pyCSEP can evaluate earthquake forecasts expressed as expected rates in space-magnitude bins, and simulation-based forecasts that produce thousands of synthetic seismicity catalogs. Most importantly, pyCSEP contains community-endorsed implementations of statistical tests to evaluate earthquake forecasts, and provides well defined file formats and standards to facilitate model comparisons. The toolkit will facilitate integrating new forecasting models into testing centers, which evaluate forecast models and prediction algorithms in an automated, prospective and independent manner, forming a critical step towards reliable operational earthquake forecasting.

Biomedical engineering (2022). doi:10.1515/bmt-2021-0379, Published by de Gruyter, Berlin [u.a.]

The development of robust forecasts of human‐induced seismicity is highly desirable to mitigate the effects of disturbing or damaging earthquakes. We assess the performance of a well‐established statistical model, the epidemic‐type aftershock sequence (ETAS) model, with a catalog of∼93,000∼93,000microearthquakes observed at the Preston New Road (PNR, United Kingdom) unconventional shale gas site during, and after hydraulic fracturing of the PNR‐1z and PNR‐2 wells. Because ETAS was developed for slower loading rate tectonic seismicity, to account for seismicity caused by pressurized fluid, we also generate three modified ETAS with background rates proportional to injection rates. We find that (1)the standard ETAS captures low seismicity between and after injections but is outperformed by the modified model during high‐seismicity periods, and (2)the injection‐rate driven ETAS substantially improves when the forecast is calibrated on sleeve‐specific pumping data. We finally forecast out‐of‐sample the PNR‐2 seismicity using the average response to injection observed at PNR‐1z, achieving better predictive skills than the in‐sample standard ETAS. The insights from this study contribute toward producing informative seismicity forecasts for real‐time decision making and risk mitigation techniques during unconventional shale gas development.

Submitted by Maria Helena Ribeiro (helena.ribeiro@lisboa.ucp.pt) on 2021-09-28T13:08:16Z No. of bitstreams: 1 152419014_Max Werner_DPDFA.pdf: 1259960 bytes, checksum: 3a69e2f667bbaf5d74ebce8bd1a64f1f (MD5) Made available in DSpace on 2021-09-28T13:08:16Z (GMT). No. of bitstreams: 1 152419014_Max Werner_DPDFA.pdf: 1259960 bytes, checksum: 3a69e2f667bbaf5d74ebce8bd1a64f1f (MD5) Previous issue date: 2021-01-26

The Kathmandu Valley is within a seismically active region with only few recorded strong-motion data. Geophysical information in the Valley is also sparse. In addition, the absence of an open database which compiles in situ geophysical tests, borehole records, and geotechnical laboratory data is affecting the advancement of knowledge in the region. This article presents SAFER/GEO-591 database, named after the Engineering and Physical Science Research Council (EPSRC)-funded project Seismic Safety and Resilience of Schools in Nepal (SAFER). SAFER/GEO-591 contains data from groundwater wells and boreholes originally commissioned for research and commercial purposes. This work describes (1) the quality assessment and harmonization process conducted on the dataset, (2) the variation of shear-wave velocity (V_S) measurements and geotechnical parameters with depth and elevation in the Valley, (3) the current understanding of the Valley sediment/bedrock topography, and finally (4) new geological cross sections. A companion article presents an updated V_(S30) map across the Valley based on the contributions of this article. The database can be downloaded from the University of Bristol repository via DOI: https://doi.org/10.5523/bris.3gjcvx51lnpuv269xsa1yrb0rw.

The 2019 Ridgecrest sequence provides the first opportunity to evaluate Uniform California Earthquake Rupture Forecast v.3 with epidemic‐type aftershock sequences (UCERF3‐ETAS) in a pseudoprospective sense. For comparison, we include a version of the model without explicit faults more closely mimicking traditional ETAS models (UCERF3‐NoFaults). We evaluate the forecasts with new metrics developed within the Collaboratory for the Study of Earthquake Predictability (CSEP). The metrics consider synthetic catalogs simulated by the models rather than synoptic probability maps, thereby relaxing the Poisson assumption of previous CSEP tests. Our approach compares statistics from the synthetic catalogs directly against observations, providing a flexible approach that can account for dependencies and uncertainties encoded in the models. We find that, to the first order, both UCERF3‐ETAS and UCERF3‐NoFaults approximately capture the spatiotemporal evolution of the Ridgecrest sequence, adding to the growing body of evidence that ETAS models can be informative forecasting tools. However, we also find that both models mildly overpredict the seismicity rate, on average, aggregated over the evaluation period. More severe testing indicates the overpredictions occur too often for observations to be statistically indistinguishable from the model. Magnitude tests indicate that the models do not include enough variability in forecasted magnitude‐number distributions to match the data. Spatial tests highlight discrepancies between the forecasts and observations, but the greatest differences between the two models appear when aftershocks occur on modeled UCERF3‐ETAS faults. Therefore, any predictability associated with embedding earthquake triggering on the (modeled) fault network may only crystalize during the presumably rare sequences with aftershocks on these faults. Accounting for uncertainty in the model parameters could improve test results during future experiments.

Operational earthquake forecasting protocols commonly use statistical models for their recognized ease of implementation and robustness in describing the short‐term spatiotemporal patterns of triggered seismicity. However, recent advances on physics‐based aftershock forecasting reveal comparable performance to the standard statistical counterparts with significantly improved predictive skills when fault and stress‐field heterogeneities are considered. Here, we perform a pseudoprospective forecasting experiment during the first month of the 2019 Ridgecrest (California) earthquake sequence. We develop seven Coulomb rate‐and‐state models that couple static stress‐change estimates with continuum mechanics expressed by the rate‐and‐state friction laws. Our model parameterization supports a gradually increasing complexity; we start from a preliminary model implementation with simplified slip distributions and spatially homogeneous receiver faults to reach an enhanced one featuring optimized fault constitutive parameters, finite‐fault slip models, secondary triggering effects, and spatially heterogenous planes informed by pre‐existing ruptures. The data‐rich environment of southern California allows us to test whether incorporating data collected in near‐real time during an unfolding earthquake sequence boosts our predictive power. We assess the absolute and relative performance of the forecasts by means of statistical tests used within the Collaboratory for the Study of Earthquake Predictability and compare their skills against a standard benchmark epidemic‐type aftershock sequence (ETAS) model for the short (24hr after the two Ridgecrest mainshocks) and intermediate terms (one month). Stress‐based forecasts expect heightened rates along the whole near‐fault region and increased expected seismicity rates in central Garlock fault. Our comparative model evaluation not only supports that faulting heterogeneities coupled with secondary triggering effects are the most critical success components behind physics‐based forecasts, but also underlines the importance of model updates incorporating near‐real‐time available aftershock data reaching better performance than standard ETAS. We explore the physical basis behind our results by investigating the localized shut down of pre‐existing normal faults in the Ridgecrest near‐source area.

Probabilistic forecasting of earthquake‐producing fault ruptures informs all major decisions aimed at reducing seismic risk and improving earthquake resilience. Earthquake forecasting models rely on two scales of hazard evolution: long‐term (decades to centuries) probabilities of fault rupture, constrained by stress renewal statistics, and short‐term (hours to years) probabilities of distributed seismicity, constrained by earthquake‐clustering statistics. Comprehensive datasets on both hazard scales have been integrated into the Uniform California Earthquake Rupture Forecast, Version 3 (UCERF3). UCERF3 is the first model to provide self‐consistent rupture probabilities over forecasting intervals from less than an hour to more than a century, and it is the first capable of evaluating the short‐term hazards that result from multievent sequences of complex faulting. This article gives an overview of UCERF3, illustrates the short‐term probabilities with aftershock scenarios, and draws some valuable scientific conclusions from the modeling results. In particular, seismic, geologic, and geodetic data, when combined in the UCERF3 framework, reject two types of fault‐based models: long‐term forecasts constrained to have local Gutenberg–Richter scaling, and short‐term forecasts that lack stress relaxation by elastic rebound.

Reliable probabilistic seismic hazard analysis (PSHA) for Nepal is a long-term goal that different researchers have been working on in the last decade. Especially after the 2015 Gorkha earthquake, several classical probabilistic seismic hazard analysis studies have been carried out for the entire Nepal. Herein, an alternative simulation-based PSHA is performed for the Kathmandu basin generating a stochastic catalogue of events using information about seismogenic zones from recent research studies. Different ground motion prediction equations have been adopted. Results show the necessity for a ground motion prediction model that is tailored for Nepal.

Traffic data identified from free, satellite imagery, such as historical imagery available on Google Earth, are collected and employed as possible indicator of business downtime after the 2015 Gorkha Earthquake in Nepal. Business downtime is a crucial factor when assessing losses and it is also of great importance to insurance and reinsurance companies, particularly the downtime of private businesses, for which limited data is available. Currently, indicators of business recovery after natural disaster events either focus on damage or they are directly collected through costly procedures and none of the approaches available quantifies longer term recovery. In many cases, such data, when available, are not easy to access. There is a need for both time- and cost-effective methods to quantify business downtime that are not necessarily derived from physical damage and are able to describe both short-term and, potentially, long-term business recovery. The aim of this quantification refers to the recovery phase rather than the immediate emergency response in theaftermath of a natural disaster. Car and lorry numbers of four areas of Kathmandu were counted from historical images available on Google Earth. These four areas correlated to the four areas in which data related to business closures were obtained through interviews during a field investigation made by the authors in Kathmandu in November 2016 (18 months after the event). Food and health business closure data correlate very satisfactory with the calculated drop in car numbers.

The Global Earthquake Activity Rate (GEAR1) seismicity model uses an optimized combination of geodetic strain rates, hypotheses about converting strain rates to seismicity rates from plate tectonics, and earthquake-catalog data to estimate global mw ≥ 5.767$ shallow (≤ 70 km) seismicity rates. It comprises two parent models: a strain rate-based model, and a smoothed-seismicity based model. The GEAR1 model was retrospectively evaluated and calibrated using earthquake data from 2005-2012, resulting in a preferred log-linear, multiplicative combination of the parent forecasts. Since October 1, 2015, the GEAR1 model has undergone prospective evaluation within the Collaboratory for the Study of Earthquake Predictability (CSEP) testing center. We present initial prospective forecast test results for the GEAR1 model, its tectonic and seismicity components, and for the first iteration of the strain rate-based model, during the period October 1, 2015 to September 7, 2017. During the evaluation period, observed earthquakes are consistent with the GEAR1 forecast and comparative test results likewise support that GEAR1 is more informative than either of its components alone. Based on a combination of retrospective and prospective testing, the tectonic forecasts do not effectively anticipate observed spatial earthquake distribution, largely due to over-localization of the model with respect to observed earthquake distributions.