In October 2016, we acquired an approximately 15-km-long seismic profile along a linear transect across the East Bay region of the San Francisco Bay area. Our goal was to image previously unknown strands of the Hayward Fault zone and to better delineate the structure and geometry of the main trace of the Hayward Fault. Our profile started near the southern border of San Leandro, California at the San Francisco Bay shoreline, trended ENE through the northern edge of Castro Valley, California, and ended approximately 5 km WSW of San Ramon, California. The data were analyzed using refraction tomography modeling, reflection processing, and guided-wave analysis. The analyzed data are presented in separate reports by Strayer and others (submitted to BSSA). The seismic data were generated at 26 shotpoints: 16 shotpoints located along the profile (inline shotpoints) and 10 shotpoints offset from the profile and located within known or suspected fault traces (guided-wave shotpoints). Most shotpoints used explosive sources to generate the seismic waves, but three of the shotpoints used repeated hits from a 227-kg (500 lb) accelerated weight dropped ~2 feet above a steel plate to generate the seismic signal. Data from each shot were recorded by a total of 459 seismographs, mostly deployed along the profile at intervals ranging from 20 to 100 meters. This data release contains the raw field records from all explosive and weight drop shots. The raw field records are in a proprietary Trimble TRD format and consist of continuous seismograph recordings during the time of the data collection. Also included in this data release are multiple SEG-Y files consisting of thirty-second-long traces "cut" from the TRD files and resorted into conventional shot gathers. These data are in standard SEG-Y format, with the data samples in IBM floating point format. Data samples are in units of meters/second, without filtering or other data manipulation.

The U.S. Geological Survey acquired high-resolution P- and S-wave seismic data across the Frijoles Fault strand of the San Gregorio Fault Zone (SGFZ) at northern Año Nuevo, California in 2012. SGFZ is a right-lateral fault system that is mainly offshore, and prior studies provide highly variable slip estimates, which indicates uncertainty about the seismic hazard it poses. Therefore, the primary goal of the seismic survey was to better understand the structure and geometry of the onshore section of the Frijoles Fault strand of the SGFZ. We deployed 118 geophones (channels) at 5-m spacing along a linear profile centered on the mapped surface trace of the Frijoles Fault and co-located active P- and S-wave sources at ~1-m offset from the geophones. Channel numbers increase from west to east along the profile. We generated P-waves using either a seisgun (www.utep.edu/science/ssf/Manuals/betsy_seisgun.pdf, accessed August 2022) or an accelerated weight-drop and S-waves by horizontally striking an aluminum block on both sides with a sledgehammer. We first deployed vertical-component geophones (40-Hz, SercelTM L40A, sensitivity of 22.34 volts/meter/second) to record P-wave sources, after which we replaced the vertical-component geophones with horizontal-component geophones (4.5-Hz, SercelTM L28-LBH, sensitivity of 31.3 volts/meter/second) to record S-wave sources. Refraction cables connected all geophones to two 60-channel Geometrics Stratavisor NX-60TM seismographs with 24-bit analog-to-digital converters. Each shot was recorded at a 0.5-ms sampling rate for two seconds, with data recording at 100 ms before the actual time of the shot. This data release provides the metadata needed to utilize the seismic data. Data Format and Files We combined each seismic trace for a given shot time into a shot gather, and the traces in each shot gather are ordered by channel numbers (1-118) based on the position of the geophones along the profile. Furthermore, we assigned a unique field number (FFID) to each shot gather, and we combined the shot gathers recorded from both seismographs into two SEG-Y files (Barry et al., 1975), 78023.segy (channels 1 to 60) and marine.segy (channels 61 to 118), which are stored in big-Endian, 4-byte IBM-floating-point format (format code 1). Data samples are in millivolts and can be converted to velocity using the geophone sensitivity values. Metadata for all profiles are contained in two text files and one xml file: PIE12.setup.csv, PIE12.location.csv, and PIE12Metadata.xml. The setup file describes the identification of shots recorded by the two seismographs, channel number, recording stations (geophones), and the source type for both SEG-Y files. The location file describes the channel number, latitude, and longitude of all geophone locations. Reference Barry, K.M., Cavers, D.A., and Kneale, C.W., 1975, Recommended standards for digital tape formats: Geophysics, vol. 40, no. 2, p. 344-352, doi: 10.1190/1.1440530.

The U.S. Geological Survey acquired high-resolution P- and S-wave seismic data across the Frijoles Fault strand of the San Gregorio Fault Zone (SGFZ) at northern Año Nuevo, California in 2012. SGFZ is a right-lateral fault system that is mainly offshore, and prior studies provide highly variable slip estimates, which indicates uncertainty about the seismic hazard it poses. Therefore, the primary goal of the seismic survey was to better understand the structure and geometry of the onshore section of the Frijoles Fault strand of the SGFZ. We deployed 118 geophones (channels) at 5-m spacing along a linear profile centered on the mapped surface trace of the Frijoles Fault and co-located active P- and S-wave sources at ~1-m offset from the geophones. Channel numbers increase from west to east along the profile. We generated P-waves using either a seisgun (www.utep.edu/science/ssf/Manuals/betsy_seisgun.pdf, accessed August 2022) or an accelerated weight-drop and S-waves by horizontally striking an aluminum block on both sides with a sledgehammer. We first deployed vertical-component geophones (40-Hz, SercelTM L40A, sensitivity of 22.34 volts/meter/second) to record P-wave sources, after which we replaced the vertical-component geophones with horizontal-component geophones (4.5-Hz, SercelTM L28-LBH, sensitivity of 31.3 volts/meter/second) to record S-wave sources. Refraction cables connected all geophones to two 60-channel Geometrics Stratavisor NX-60TM seismographs with 24-bit analog-to-digital converters. Each shot was recorded at a 0.5-ms sampling rate for two seconds, with data recording at 100 ms before the actual time of the shot. This data release provides the metadata needed to utilize the seismic data. Data Format and Files We combined each seismic trace for a given shot time into a shot gather, and the traces in each shot gather are ordered by channel numbers (1-118) based on the position of the geophones along the profile. Furthermore, we assigned a unique field number (FFID) to each shot gather, and we combined the shot gathers recorded from both seismographs into two SEG-Y files (Barry et al., 1975), 78023.segy (channels 1 to 60) and marine.segy (channels 61 to 118), which are stored in big-Endian, 4-byte IBM-floating-point format (format code 1). Data samples are in millivolts and can be converted to velocity using the geophone sensitivity values. Metadata for all profiles are contained in two text files and one xml file: PIE12.setup.csv, PIE12.location.csv, and PIE12Metadata.xml. The setup file describes the identification of shots recorded by the two seismographs, channel number, recording stations (geophones), and the source type for both SEG-Y files. The location file describes the channel number, latitude, and longitude of all geophone locations. Reference Barry, K.M., Cavers, D.A., and Kneale, C.W., 1975, Recommended standards for digital tape formats: Geophysics, vol. 40, no. 2, p. 344-352, doi: 10.1190/1.1440530.

In September 2021, the U.S. Geological Survey acquired high-resolution P- and S-wave data near seismic station CE.57213 in Fremont, California, approximately 100 m east of the mapped trace of the Hayward Fault. We acquired the seismic data to evaluate the time-averaged shear-wave velocity in the upper 30 m (VS30) and to better understand ground-shaking near the station CE.57213. The seismic data were acquired using a linear array of SmartSolo 3-component nodal seismometers (nodes), which continuously recorded at 2000 samples per second (0.5-ms sampling rate). We deployed 60 nodes, spaced at 2-m increments, along a 180-m-long, northeast-southwest-trending linear array. We generated P-wave seismic sources (shots) adjacent to each node at a 1-m offset using a 3.5-kg sledgehammer to vertically strike a steel plate on the ground surface. S-wave sources (shots) were generated adjacent to each node by horizontally striking an aluminum block with a 3.5-kg sledgehammer. For each shot point, we extracted approximately 2 seconds of data from each node following the shot times, combined the seismic traces into a single shot gather, and stored the data in SEG-Y format (Barry et al, 1975). This report provides the metadata needed to analyze the seismic data.

In September 2021, the U.S. Geological Survey acquired high-resolution P- and S-wave data near seismic station CE.57213 in Fremont, California, approximately 100 m east of the mapped trace of the Hayward Fault. We acquired the seismic data to evaluate the time-averaged shear-wave velocity in the upper 30 m (VS30) and to better understand ground-shaking near the station CE.57213. The seismic data were acquired using a linear array of SmartSolo 3-component nodal seismometers (nodes), which continuously recorded at 2000 samples per second (0.5-ms sampling rate). We deployed 60 nodes, spaced at 2-m increments, along a 180-m-long, northeast-southwest-trending linear array. We generated P-wave seismic sources (shots) adjacent to each node at a 1-m offset using a 3.5-kg sledgehammer to vertically strike a steel plate on the ground surface. S-wave sources (shots) were generated adjacent to each node by horizontally striking an aluminum block with a 3.5-kg sledgehammer. For each shot point, we extracted approximately 2 seconds of data from each node following the shot times, combined the seismic traces into a single shot gather, and stored the data in SEG-Y format (Barry et al, 1975). This report provides the metadata needed to analyze the seismic data.

In June of 2020, the U.S. Geological Survey conducted a high-resolution seismic survey at Edwards Air Force Base in Kern County, California. Seismic data were acquired using 601 DTCC SmartSolo 3-component nodal seismometer systems (“nodes”), which continuously recorded at 2000 samples per second. Nodes were deployed 5 meters apart along a southwest-northeast trend to create an approximately 3-km-long linear profile. P-wave seismic sources were generated primarily using a 500-lb (227-kg) accelerated weight drop at each recording station. P-wave sources were also generated at every 40 stations using downhole explosions. Fault-zone-guided waves were generated using explosive sources that were placed within a mapped trace of a nearby fault (Kramer Hills Fault zone), located approximately 1 km southeast of the seismic profile. Shot gathers were created in SEG-Y format (Barry et al, 1975) by extracting several seconds of data from each node for each recorded shot time. This report provides the metadata needed to analyze the seismic data.

In June of 2020, the U.S. Geological Survey conducted a high-resolution seismic survey at Edwards Air Force Base in Kern County, California. Seismic data were acquired using 601 DTCC SmartSolo 3-component nodal seismometer systems (“nodes”), which continuously recorded at 2000 samples per second. Nodes were deployed 5 meters apart along a southwest-northeast trend to create an approximately 3-km-long linear profile. P-wave seismic sources were generated primarily using a 500-lb (227-kg) accelerated weight drop at each recording station. P-wave sources were also generated at every 40 stations using downhole explosions. Fault-zone-guided waves were generated using explosive sources that were placed within a mapped trace of a nearby fault (Kramer Hills Fault zone), located approximately 1 km southeast of the seismic profile. Shot gathers were created in SEG-Y format (Barry et al, 1975) by extracting several seconds of data from each node for each recorded shot time. This report provides the metadata needed to analyze the seismic data.

This three-dimensional (3D) seismic velocity model includes a detailed domain covering the greater San Francisco Bay urban region and a regional domain at a coarser resolution covering a larger region. Version 21.1 updates only the detailed domain with adjustments to the elastic properties east and north of the San Francisco Bay. There are no changes to the underlying 3D geologic model or the regional domain seismic velocity model. Version 21.1 of the detailed domain fits seamlessly inside version 21.0 of the regional domain without any jumps in elastic properties across the boundary between the two domains. The model was constructed by assigning elastic properties (density, Vp, Vs, Qp, and Qs) to grids of points based on the geologic unit and depth from the ground surface. The model is stored in HDF5 files using the GeoModelGrids (https://geomodelgrids.readthedocs.io) storage scheme. GeoModelGrids provides a high-level interface for accessing the model. The model can also be accessed using the HDF5 application programming interface provided with many programming languages and tools, but the user will be responsible for all coordinate transformations.

This three-dimensional (3D) seismic velocity model includes a detailed domain covering the greater San Francisco Bay urban region and a regional domain at a coarser resolution covering a larger region. Version 21.1 updates only the detailed domain with adjustments to the elastic properties east and north of the San Francisco Bay. There are no changes to the underlying 3D geologic model or the regional domain seismic velocity model. Version 21.1 of the detailed domain fits seamlessly inside version 21.0 of the regional domain without any jumps in elastic properties across the boundary between the two domains. The model was constructed by assigning elastic properties (density, Vp, Vs, Qp, and Qs) to grids of points based on the geologic unit and depth from the ground surface. The model is stored in HDF5 files using the GeoModelGrids (https://geomodelgrids.readthedocs.io) storage scheme. GeoModelGrids provides a high-level interface for accessing the model. The model can also be accessed using the HDF5 application programming interface provided with many programming languages and tools, but the user will be responsible for all coordinate transformations.

Between 1993 and 1997, the U.S. Geological Survey acquired high-resolution, marine seismic-reflection profile data across submerged portions of known and inferred upper crustal fault zones throughout the greater San Francisco Bay area. This particular dataset was acquired in 1997 during USGS Field Activity J4-97-SF using the vessel David Johnston. The dataset includes navigational data in ASCII format, gif images of the seismic-profile lines, and seismic data in industry-standard SEG-Y format. These data are also available via GeoMapApp (http://www.geomapapp.org/) and Virtual Ocean (http://www.virtualocean.org/) earth science exploration and visualization applications.