Abstract: The Pacific Tsunami Warning Center (PTWC) in Honolulu, Hawaii, routinely analyses most earthquakes with a 5.5 or larger magnitude occurring around the world. The PTWC first issues an unofficial observatory message (OM) containing a set of preliminary parameters for these events. When the estimated magnitude crosses the 6.5 magnitude threshold, however, the protocol calls for the issuance of at least a tsunami information bulletin. For years, PTWC geoscientists assumed that the inclusion of more stations in the initial analysis automatically improved the quality of the preliminary source parameters. We assessed the validity of these assumptions by matching 627 observatory messages issued by the PTWC between 2003 and 2017 with the official tsunami messages that followed. We then computed the epicentral offsets, magnitude residuals, and response times against the source parameters listed in later, more authoritative earthquake catalogs. These statistics reveal that for 54\% of the earthquakes both the OMs and the official tsunami messages reported the same magnitude despite up to 20 additional minutes of processing time. Paradoxically, in another 17\% of the cases, the magnitude residuals worsened instead of improving. In the remaining 29\%, the earthquake magnitude estimates saw improvements characterized by a median of 0.3 magnitude unit. These results show that, except when dealing with some very large or complex earthquakes, the quality of PTWC’s earthquake preliminary parameters benefits little by the extra message delays incurred by adding more stations to the initial analysis. We conclude that in the overwhelming majority of cases waiting to include more stations in the initial analyses, or manually reviewing individual station magnitudes before issuing the first official message product turns into a waste of otherwise precious time, something particularly crucial when warning of an impending tsunami in the near field.
Media has not been submitted for this Presentation
Abstract: Analysis of the seismograms recorded by the seismic station array provides the fundamental information of earthquake source spectra and site effects. Previous studies use different sections of the seismograms, e.g. the direct P and S waves and coda waves, to study the two terms. Both stress drops derived from the source spectra and site effects have practical implications for the strong ground motion prediction. In this study, our aim is to comprehensively compare the source spectra and site effects derived from direct waves and coda waves separately and to explore the self-similar scaling of the earthquake source properties. We analyze around 1500 earthquakes with the local magnitude (ML) range between 1.2 and 3.5 in the San Jacinto Fault region. Both the spectra decomposition method (Peter et al., 2006 and Trugman & Shearer, 2017) and the improved Coda-Q method (Wang & Shearer, 2017) are applied to derive the source spectra and site effects. To fit the source spectra by using the Brune-type crack source model, both self-similarity and none self-similarity are explored by assuming the constant stress drops or the moment dependent stress drops. The error estimations of both methods will be discussed in this study. This work will provide a systematic comparison of source spectra and site effects from direct waves and coda waves. Furthermore, another comprehensive analysis of the earthquake source spectra and site effects will be conduct in the southern California.
Media has not been submitted for this Presentation
Abstract: In collaboration with the VA, the U.S. Geological Survey has developed structural health monitoring (SHM) software that utilizes vibration inputs to continually analyze and archive the response characteristics of a building in near real-time. The SHM software is built on the Earthworm (EW) system (Johnson et al., 1995), which is an open data processing platform that allows any continuous waveform data to be collected into ring buffers from a digitizer for further analyses (http://www.isti2.com/ew). The SHM software initially determines baselines for a suite of structural response parameters, and then continuously examines the response for changes in these parameters. The structural parameters monitored currently are inter-story drift ratios, shear-wave travel times throughout the building, and base-shear capacity-demand ratio. The SHM software is integrated with a web-enabled SHM data management framework to support aggregation, storage, and reporting of SHM data obtained and analyzed from instrumented hospital buildings to record strong shaking from earthquakes. By analyzing and characterizing the threshold values for building-specific engineering demand parameters, the SHM software can determine inspection priority to be low, moderate, high or very high and thus assist efforts in evaluating the safety and integrity of buildings in the aftermath of an earthquake. The SHM software is scalable—to support an arbitrary number of sensors, and it is extensible—to accommodate new data streams without the need to rewrite storage and display logic. The SHM software works on site or remotely. The software was validated using both ambient and low- and high-intensity shaking data inputs to a full-scale seven-story reinforced concrete building section tested on the UC San Diego shake table.
Media has not been submitted for this Presentation
Abstract: Aftershock forecast modeling is an important tool for investigating the locations and magnitudes of historical earthquakes as well as for short-term earthquake forecasting. Although the 16 July 1936 M6 Milton-Freewater earthquake is the largest historical earthquake in eastern Oregon, having been widely felt in eastern Washington, northeastern Oregon, and northern Idaho, its location is uncertain. Various studies have reported its epicenter as lying within 30 km of the intersection of the Hite and Wallula faults. In the absence of reported coseismic surface rupture, we estimate the mainshock location and magnitude from aftershock forecast modeling that considers the numbers of reported aftershocks at six different locations in the day following the quake. The estimated epicenter and magnitude derived from the aftershock modeling compare favorably to the ISC-GEM solution, an S-P time recorded at Spokane, the directions of observed horizontal motions, and the distribution of ground failure and liquefaction. The aftershock modeling constrains the epicenter of the 1936 earthquake with an accuracy of about 5 km and is most consistent with an epicenter midway between Umapine and Milton-Freewater, above an estimated 10 km long subsurface rupture between them. This suggested epicenter and rupture plane are consistent with the elongation of ground failure along the Wallula fault and the fault’s strike along one of the focal planes indicating that the earthquake may have primarily ruptured the eastern end of NW-trending Wallula fault. Aftershock forecast modeling indicates a mainshock magnitude between 6.2 to 6.4, in agreement with reported instrumental magnitudes but higher than the range derived from seismic intensity data (5.1 to 5.5). This discrepancy may result from a systematic underestimation of the seismic intensities in the epicentral region: felt reports justify a maximum Modified Mercalli intensity of VIII instead of the VII previously assigned.
Media has not been submitted for this Presentation
Abstract: Tectonic classification of earthquakes is a key component of the USGS Global ShakeMap system because it serves as the basis for the selection of ground motion prediction equations (GMPEs). GMPE selection has a significant impact on the estimated ground motion intensities and therefore downstream USGS products (e.g., PAGER, ShakeCast). The current method of earthquake regionalization provides a seismotectonic classification of the event (e.g., “subduction zone interface” or “active crustal”), which is then mapped to an appropriate GMPE. The existing algorithm needs to be updated to account for issues that were not originally anticipated, such as the classification of induced earthquakes, and to account for the uncertainty of the classification. Uncertainty in the classification arises from many sources, including 1) uncertainty in the location of the earthquake and its treatment as a point source, 2) uncertainty in the focal mechanism, and 3) the precision of the boundaries between different tectonic environments. In the presence of such inherent uncertainty in the classification, we propose a probabilistic approach, where all classes are assigned a likelihood, rather than reporting a single “best guess” classification. The resulting ground motions are computed as a weighted combination of the GMPEs that are assigned to each region, where each region may be assigned one or more GMPEs with varying weights. This approach accounts for the uncertainty inherent to the regionalization, and helps to minimize artificial changes in ShakeMap versions due to a change in the tectonic classification. Our primary motivation is for assigning GMPEs in ShakeMap, but we do anticipate several other applications, such as the development of probabilistic seismic hazard models and earthquake early warning alerts.
Media has not been submitted for this Presentation
Abstract: Pantanal is a sedimentary basin of quaternary age, located in the Center-West region of Brazil, in the Upper Paraguay River Basin, constituting a seismogenic zone. The physiography of the area is characterized by the presence of fluvial megafans, mostly sandy, modern and associated with fault lines and structural lineaments. The Taquari Megafan is the largest of the Pantanal fluvial fans, constituting, therefore, the most remarkable feature of the basin seismicity. The region is structured by faults of reverse type generated under compressive stress, coinciding with epicenters of current earthquakes, which are evidenced by instrumental data. In the Mato Grosso do Sul state, the first seismographic station installed was the Aquidauana (AQDA) – MS, in May 2003, being an important landmark for seismology, as part of a research project titled “Integrated Seismographic Network of Brazil (BRASIS)”, with the Institute of Astronomy, Geophysics and Atmospheric Sciences (IAG – USP). The seismic data analyzed presented magnitudes ≥ 3.5, between 2003 and 2017. Six seismic events were recorded in Pantanal, four of which occurred in the Taquari Megafan, with magnitudes ranging from 3.5 to 4.8 Mb. Thus, it was possible to identify a major seismogenic zone in the Pantanal Sedimentary Basin, in the Nhecolândia region. The two most significant earthquakes that had their calculated focal mechanisms occurred in Coxim (2009) and Miranda (2015), with magnitudes 4.8 and 4.0Mb, respectively, with depths around 5 km, as the result of transpressive and compressive stresses in both cases. Despite being a region in subsidence, the neotectonic stresses in the upper crust are compressive. The Pantanal Basin crustal thinning might be considered one of the hypothesis of its origin which is related to its seismicity.
Media has not been submitted for this Presentation
