The use of numerical simulations in probabilistic seismic hazard analysis (PSHA) has achieved a promising level of reliability in recent years. One example is the CyberShake project, which incorporates physics‐based 3D ground‐motion simulations within seismic hazard calculations. Nonetheless, considerable computational time and resources are required due to the significant processing requirements imposed by source‐based models on one hand, and the large number of seismic sources and possible rupture variations on the other. This article proposes to use a less computationally demanding simulation‐based PSHA framework for CyberShake. The framework can accurately represent the seismic hazard at a site, by only considering a subset of all the possible earthquake scenarios, based on a Monte‐Carlo simulation procedure that generates earthquake catalogs having a specified duration. In this case, ground motions need only be simulated for the scenarios selected in the earthquake catalog, and hazard calculations are limited to this subset of scenarios. To validate the method and evaluate its accuracy in the CyberShake platform, the proposed framework is applied to three sites in southern California, and hazard calculations are performed for earthquake catalogs with different lengths. The resulting hazard curves are then benchmarked against those obtained by considering the entire set of earthquake scenarios and simulations, as done in CyberShake. Both approaches yield similar estimates of the hazard curves for elastic pseudospectral accelerations and inelastic demands, with errors that depend on the length of the Monte‐Carlo catalog. With 200,000 yr catalogs, the errors are consistently smaller than 5% at the 2% probability of exceedance in 50 yr hazard level, using only ∼3% of the entire set of simulations. Both approaches also produce similar disaggregation patterns. The results demonstrate the potential of the proposed approach in a simulation‐based PSHA platform like CyberShake and as a ground‐motion selection tool for seismic demand analyses.

中文翻译：

---

使用蒙特卡洛地震目录进行基于模拟的地震灾害评估：在 Cyber​​Shake 中的应用

近年来，数值模拟在概率地震危险性分析 (PSHA) 中的使用已取得了令人鼓舞的可靠性水平。一个例子是 Cyber​​Shake 项目，该项目将基于物理的 3D 地面运动模拟纳入地震危害计算。尽管如此，一方面由于基于震源的模型强加了大量的处理要求，另一方面需要大量的震源和可能的破裂变化，因此需要大量的计算时间和资源。本文建议对 Cyber​​Shake 使用计算要求较低的基于仿真的 PSHA 框架。该框架可以准确地表示一个地点的地震危险，只需考虑所有可能的地震场景的一个子集，基于生成具有指定持续时间的地震目录的蒙特卡罗模拟程序。在这种情况下，只需为地震目录中选择的场景模拟地面运动，并且危险计算仅限于该场景子集。为了验证该方法并评估其在 Cyber​​Shake 平台上的准确性，将所提出的框架应用于加利福尼亚南部的三个站点，并对不同长度的地震目录进行了危害计算。然后将得到的危险曲线与通过考虑整套地震场景和模拟获得的曲线进行基准测试，如 Cyber​​Shake 中所做的那样。两种方法对弹性伪谱加速度和非弹性需求的风险曲线产生相似的估计，误差取决于蒙特卡洛目录的长度。对于 200,000 年的目录，在 50 年危险水平的 2% 超标概率下，误差始终小于 5%，仅使用整个模拟集的约 3%。这两种方法也产生类似的分解模式。结果证明了所提出的方法在基于模拟的 PSHA 平台（如 Cyber​​Shake）以及作为地震需求分析的地面运动选择工具中的潜力。