Focal mechanism solutions of 122 earthquakes occurred between 2024 and 2025 in the Southeastern Alps
收藏资源简介：
This dataset contains quality-controlled focal mechanism solutions for 122 earthquakes that occurred in the Southeastern Alps between January 2024 and October 2025, with local magnitudes Md 1.0–4.6.First-motion polarities were automatically classified using the Convolutional First Motion (CFM) deep-learning model (Messuti et al., 2023). Focal mechanisms were then computed using SKHASH, a Python implementation of the HASH algorithm for focal-mechanism inversion based on first-motion polarities (Skoumal et al., 2024).The dataset includes hypocentral parameters (origin time, latitude, longitude, depth), local magnitude (Md), nodal-plane orientations (strike, dip, rake), and solution-quality metrics derived from the SKHASH inversion, including polarity misfit, nodal-plane uncertainty, number of polarities, and station distribution ratio (STDR). This dataset also includes Validation_dataset.csv file for the validation dataset used to evaluate the automated CFM–SKHASH workflow. The dataset includes 159 earthquakes from 2014–2023 and compares focal mechanisms estimated using CFM–SKHASH with reference mechanisms from Magrin et al. (2024), which were obtained using manually reviewed polarities and the FPFIT algorithm. In this file, Event_id is given in YearMonthDayHourMinuteSecond format; latitude and longitude are reported in decimal degrees; Magnitude_MD is the duration magnitude; and Depth_km is the hypocentral depth in kilometers. Strike_1, Dip_1, and Rake_1 are the nodal-plane parameters estimated in this study using the CFM–SKHASH workflow, whereas Strike_2, Dip_2, and Rake_2 are the corresponding reference nodal-plane parameters from Magrin et al. (2024), obtained using manually reviewed polarities and the FPFIT algorithm. ReferencesMessuti, G., S. Scarpetta, O. Amoroso, F. Napolitano, M. Falanga, and P. Capuano (2023). CFM: A convolutional neural network for first-motion polarity classification of seismic records in volcanic and tectonic areas, Front. Earth Sci. 11, https://doi.org/10.3389/feart.2023.1223686. Magrin, A., M. Sugan, A. Snidarcig, M. A. Romano, M. Guidarelli, M. Santulin, P. Di Bartolomeo, and A. Saraò (2024). Focal mechanism solutions of 162 earthquakes occurred between 2014 and 2023 in the Southeastern Alps, Zenodo, https://doi.org/10.5281/zenodo.10851786. Skoumal, R. J., J. L. Hardebeck, and P. M. Shearer (2024). SKHASH: A Python package for computing earthquake focal mechanisms, Seismol. Res. Lett. 95(4), 2519–2526, https://doi.org/10.1785/0220230329
本数据集包含2024年1月至2025年10月间东南阿尔卑斯山区发生的122次地震的经过质量管控的震源机制解，这些地震的地方震级（Md）介于1.0至4.6之间。初动极性首先通过卷积初动（Convolutional First Motion, CFM）深度学习模型（Messuti等，2023）完成自动分类。随后，基于初动极性的震源机制反演通过SKHASH完成计算——SKHASH是一款基于HASH算法的Python实现工具包（Skoumal等，2024）。本数据集涵盖震源参数（发震时刻、纬度、经度、深度）、地方震级（Md）、节面参数（走向、倾角、滑动角），以及由SKHASH反演得到的解质量指标，包括极性残差、节面不确定性、初动极性数量与台站分布比（STDR）。 本数据集同时包含用于验证自动化CFM–SKHASH工作流的验证数据集文件Validation_dataset.csv。该验证数据集涵盖2014至2023年间的159次地震，将通过CFM–SKHASH估算得到的震源机制解，与Magrin等（2024）基于人工复核初动极性及FPFIT算法得到的参考震源机制解进行对比。该文件中，Event_id采用"YearMonthDayHourMinuteSecond"格式；纬度与经度以十进制度数表示；Magnitude_MD为持续震级；Depth_km为以千米为单位的震源深度。Strike_1、Dip_1与Rake_1为本研究通过CFM–SKHASH工作流估算得到的节面参数，而Strike_2、Dip_2与Rake_2则为Magrin等（2024）通过人工复核初动极性及FPFIT算法得到的对应参考节面参数。 参考文献 Messuti, G., S. Scarpetta, O. Amoroso, F. Napolitano, M. Falanga, and P. Capuano (2023). CFM: 一款用于火山与构造区域地震记录初动极性分类的卷积神经网络，Front. Earth Sci. 11, https://doi.org/10.3389/feart.2023.1223686. Magrin, A., M. Sugan, A. Snidarcig, M. A. Romano, M. Guidarelli, M. Santulin, P. Di Bartolomeo, and A. Saraò (2024). 2014—2023年东南阿尔卑斯山区162次地震的震源机制解，Zenodo, https://doi.org/10.5281/zenodo.10851786. Skoumal, R. J., J. L. Hardebeck, and P. M. Shearer (2024). SKHASH: 一款用于计算地震震源机制的Python工具包，Seismol. Res. Lett. 95(4), 2519–2526, https://doi.org/10.1785/0220230329
