Original Seismograms Research Articles

Testing the algorithm for determining the velocity model of the upper part of the medium based on the first times arrivals

The results of testing the algorithm and the software module created on its basis are presented. The algorithm allows us to build a velocity model of the medium upper part using the times of the first arrivals of seismic signals related to refracted waves. The module implements two possibilities of these times determination. The first involves determining the times before the module is starting. The second implements the automatic determination of such times from the original seismograms using neural networks.

A refined catalogue of focal mechanisms for the Intra‑Carpathian region of Romania: implications for the stress field and seismogenic features assessment

We present a refined and the most complete catalogue of the focal mechanisms for the Intra-Carpathian region of Romania. It contains the high-quality solutions computed for 1217 earthquakes recorded between 1909 and 2018. Primary data gathered from the original seismograms and seismic bulletins have been used to compute the source parameters and focal mechanisms solutions. The focal mechanisms have been obtained using the HASH method by the polarities and S/P amplitudes ratios inversion. Our catalogue provides data necessary for the investigation of the contemporary stress field at different scales with high spatial and temporal resolution. We determined the stress field characteristics through formal inversion of focal mechanisms and also computed the reactivation potential of the active fault systems using the Win‑Tensor program. The stress field is heterogeneous, with SHmax significantly deviating from the first-order stress field direction and also with strong local variations in the stress regime.

Assessing the Fidelity of Seismic Records from Microfilm and Paper Media

Abstract During the analog era of seismology, seismograms were typically written to paper media. The legacy records from this era make up an irreplaceable resource that covers over a century, and many of them remain extant in both their original paper form and as microformat copies. While microformat copies of seismograms drove decades of seismological research, new capabilities have been developed that allow us to revisit the analog data sets and extend application of contemporary analyses to a considerably longer seismological record. Though microformat seismogram copies have been extensively used, we seek to understand possible differences that could arise from using data recovered from such copies compared to using the original seismograms. Identification of these differences provides understanding of the fidelity of information that is recoverable from these different media. We present a case study here, applying a number of quantitative measures to a pair of records (original paper and its 35 mm microfilm copy), and revealing that microfilm-derived time series differ from the original paper in timing and amplitude. Based on a simple analysis of this pair of short-period vertical records, we estimate uncertainties of ±0.1–0.3 for earthquake magnitude (hence, a factor of about 1.3–2.0 in energy) and ±10 km for hypocentral location.

The 9 October 1963 Vajont Catastrophe from the Point of View of the WWSSN-LP Recordings of the TRI-117 Station, Trieste, Italy

AbstractIn this study, we analyze the seismic signal generated by the 1963 Vajont catastrophic landslide recorded at the Worldwide Standardized Seismographic Station Network-Long Period station of Trieste (Italy). The landslide (nearly 260–270 million m3) invaded an artificial reservoir designed for electrical production, and generated a 220 m high wave that flowed over the dam and claimed the lives of approximately 2000 people. The original seismograms have been digitized and analyzed using time–frequency tools and numerical simulations. The results indicate that a seismic signal comparable to that generated by an Ms 3.7 earthquake was generated by the landslide. Furthermore, the calculated nearly 2×1014 J of frictional energy, considering the known parameter of the mass movement, is compatible with a friction coefficient of 0.29, in excellent agreement with the values from previous studies. The seismic efficiency that we calculate (1.12×10−4–4.45×10−4), also taking into account available data on the landslide, is within the range of values previously noted in literature. Finally, via the numerical simulations and adopting an ad hoc crustal model for the area, the origin time of the event is estimated at 21 hr 41 min 42 s UTC. The results confirm the importance of the re-analysis of analog seismograms with modern tools within a multihazard context.

Challenges and Opportunities in Turning Large U.S. Archives of Analog Seismograms into a Modern Usable Resource

AbstractQuantitative seismology is based firmly on the analysis of actual ground motions, and the transition to digital recording in the 1980s enabled sophisticated new capabilities to extract useful results from waveforms. With some effort, these tools can also be applied to analog records. Focusing on assets available within U.S. institutions, we review the necessary steps and the challenges in enabling “data rescue”—that is, preserving the scientific information latent in large analog seismogram archives and making it usable. They include: determining what assets are available (the analog seismogram archives held by various institutions, with associated metadata on instrument responses, station locations, and timing information); developing a consensus on the top level of a triage process (which analog records most definitely should be rescued?); deciding the level of quality needed in copying original seismograms to media suitable for digitizing; assessing the relative merits of scanning and digitizing; and, the need for a community service in distributing scans and digital records, as they accumulate. The necessary level of effort can benefit from practical experience. For example, specific studies have used digitized versions of analog recordings to model earthquake sources and assess seismic hazard. Other studies have used them to gain experience with nuclear explosion signals recorded at regional distances, noting that regional signals enable explosions to be monitored down to levels much lower than those attainable teleseismically.The opportunities presented by large archives of analog seismograms include the insights they present to current and future seismologists studying earthquakes and explosions, into the practical areas of assessing seismic hazard, monitoring for test ban compliance down to low explosion yields—and prompt characterization of actual explosions should they occur, as well the traditional academic pursuit of a better understanding of earthquake physics.

Geophysical analysis of the 30 July 1972 Sitka, Alaska, earthquake sequence

AbstractThe 1972 Mw 7.6 Sitka earthquake is the largest historical event along the southeastern Alaska portion of the strike-slip Queen Charlotte fault, the transform boundary between the Pacific and North American plates. The fault is one of the fastest moving transform boundaries in the world, having accumulated enough slip since 1972 to produce an event of comparable size in the near future. Thus, understanding the controls on the rupture process of the 1972 mainshock is important for seismic hazard assessment in Alaska. Following the mainshock, the U.S. Geological Survey installed a network of portable seismographs that recorded over 200 aftershocks. These locations were never published, and the original seismograms and digital phase data were misplaced. However, we were able to scan paper copies of the phase data, convert the data to digital form, and successfully relocate 87 aftershocks. The relocations show two clusters of aftershocks along the Queen Charlotte fault, one ∼40 km north of the mainshock epicenter and the other just south of the mainshock, both regions adjacent to portions of the fault that experienced maximum moment release in 1972. Many of the northern aftershocks locate east of the Queen Charlotte fault. This pattern is similar to aftershocks observed in the 2013 Mw = 7.5 Craig, Alaska earthquake. Recent and pre-1971 (1925–1970) seismicity indicates that the regions where aftershocks clustered remained active through time. Gravity, magnetic, and bathymetric anomalies suggest that the structural variations in both the Pacific and North American plates (e.g., age, density, rock type, and thickness) play roles in rupture nucleation and termination along the northern Queen Charlotte fault.

Historical seismogram reproductions for the source parameters determination of the 1902, Atushi (Kashgar) earthquake

The majority of original seismograms recorded at the very beginning of instrumental seismology (the early 1900s) did not survive till present. However, a number of books, bulletins, and catalogs were published including the seismogram reproductions of some, particularly interesting earthquakes. In case these reproductions contain the time and amplitude scales, they can be successfully analyzed the same way as the original records. Information about the Atushi (Kashgar) earthquake, which occurred on August 22, 1902, is very limited. We could not find any original seismograms for this earthquake, but 12 seismograms from 6 seismic stations were printed as example records in different books. These data in combination with macroseismic observations and different bulletins information published for this earthquake were used to determine the source parameters of the earthquake. The earthquake epicenter was relocated at 39.87° N and 76.42° E with the hypocenter depth of about 18 km. We could further determine magnitudes m B = 7.7 ± 0.3, M S = 7.8 ± 0.4, M W = 7.7 ± 0.3 and the focal mechanism of the earthquake with strike/dip/rake − 260°± 20/30°± 10/90°± 10. This study confirms that the earthquake likely had a smaller magnitude than previously reported (M8.3). The focal mechanism indicates dominant thrust faulting, which is in a good agreement with presumably responsible Tuotegongbaizi-Aerpaleike northward dipping thrust fault kinematic, described in previous studies.

Seismic parameters re-determined from historical seismograms of 1935-Erdek–Marmara Island and 1963-Çınarcık Earthquakes

In this study, the original seismograms of the 1935-Erdek–Marmara Island and 1963-Cinarcik Earthquakes, recorded at local and regional distances, were vectorized. The epicentral locations have been calculated using available readings from original records and also ISS and seismic station bulletins for 04.01.1935-14:41 and 16:20 Marmara Island–Erdek Earthquakes and 18.09.1963-16:58 Cinarcik Earthquake. The epicenter determinations show that the first event in 04.01.1935 was located at 40.72N–27.72E, while the second one occurred at 40.61N–27.43E, indicating that both were located near the Marmara Island. Another finding is that the 1963 event was located at 40.80N–29.18E, near the Princes’ Island fault. Furthermore, moment tensor inversion method was applied on these earthquakes by using original seismograms, which provided an opportunity to illuminate the seismotectonic features of Marmara Region based on the retrieved fault mechanism solutions. For the first time, the fault mechanisms for 04.01.1935-14:41 and 16:20 Earthquakes were determined using moment tensor inversion from the original seismic waveforms. Likewise, the result obtained for the fault mechanism of 1963 Cinarcik Earthquake showed normal fault mechanism with much shallower depth than estimated before. Our preferred solutions showed that the fault mechanisms for the three events are normal faults and coincide with the seismotectonic structure of the Marmara Region.

P-wave tomography and dynamics of the crust and upper mantle beneath western Tibet

The continental collision between the Indian and Asian plates plays a key role in the geologic and tectonic evolution of the Tibetan plateau. In this article we present high-resolution tomographic images of the crust and upper mantle derived from a large number of high-quality seismic data from the ANTILOPE project in western Tibet. Both local and distant earthquakes were used in this study and 35,115 P-wave arrival times were manually picked from the original seismograms. Geological and geochemical results suggested that the subducting Indian plate has reached northward to the Lhasa terrane, whereas our new tomography shows that the Indian plate is currently sub-horizontal and underthrusting to the Jinsha river suture at depths of ~100 to ~250km, suggesting that the subduction process has evolved over time. The Asian plate is also imaged clearly from the surface to a depth of ~100km by our tomography, and it is located under the Tarim Basin north of the Altyn Tagh Fault. There is no obvious evidence to show that the Asian plate has subducted beneath western Tibet. The Indian and Asian plates are separated by a prominent low-velocity zone under northern Tibet. We attribute the low-velocity zone to mantle upwelling, which may account for the warm crust and upper mantle beneath that region, and thus explain the different features of magmatism between southern and northern Tibet. But the upwelling may not penetrate through the whole crust. We propose a revised geodynamic model and suggest that the high-velocity zones under Lhasa terrane may reflect a cold crust which has interrupted the crustal flow under the westernmost Tibetan plateau.

Short-Period Source Model of the 2011 Mw 9.0 Off the Pacific Coast of Tohoku Earthquake

We have constructed a short‐period source model for the 2011 M w 9.0 Off the Pacific Coast of Tohoku earthquake using strong‐motion records at stations near the source fault. The observed strong motions contain five wavepackets that correspond to specific strong‐motion generation areas (SMGAs). The origins of the wavepackets were retrieved from the original seismograms using a semblance analysis. We determine the locations of the SMGAs based on sources extracted from the corresponding wavepackets. The short‐period source model consists of five SMGAs (SMGA1–5), which are located west of the hypocenter and along the down‐dip edge of the source fault. SMGA1 is located in the Miyagi‐Oki source region west of the hypocenter, SMGA2 in the middle Sanriku‐Oki source region north of the hypocenter, and SMGA3 in the southern Sanriku‐Oki source region west of the hypocenter. SMGA4 and SMGA5 are located near the down‐dip edge of the mainshock source fault, extending from offshore Fukushima prefecture to offshore Ibaraki prefecture. At some stations near the source fault, impulsive waves are also seen on the recorded seismograms. Ground motions corresponding to these impulsive waves cannot be accounted for using the conventional uniform SMGA model. We attempt instead to simulate the observed ground motions at the Onagawa Nuclear Power Plant at a depth of 128 m very near the source fault, using a heterogeneous source model. The impulsive waves are well simulated using the heterogeneous model with higher stress parameters within a small subarea inside the SMGAs.

Collisional tectonics between the Eurasian and Philippine Sea plates from tomography evidences in Southeast China

Abstract The upper mantle structure of Southeast China is important for us to understand the deformation and mantle dynamics process associated with the interaction between the Eurasian plate and Philippine Sea (PHS) slab. We determined a detailed three-dimensional P-wave velocity (Vp) structure of the crust and upper mantle down to 400 km depth beneath Southeast China by applying teleseismic tomography to 6869 high-quality P-wave arrival times. The data were collected very carefully from the original seismograms of 635 teleseismic events recorded by 65 broadband stations deployed in Southeast China. Our images show that the high-Vp PHS slab subducts toward the north along the Ryukyu trench at the latitude of about 24°N and extends down to 350 km depth and even more. High-Vp anomalies are imaged in the upper mantle under central and southern Taiwan, which represent the subducted Eurasian plate. Break-off Eurasian plate at a big angle subducting eastward is revealed under central Taiwan at depths from the upper mantle to 400 km. While continuous Eurasian plate under South Taiwan is mainly imaged from the Moho down to 400 km depth, a torn mantle window within the Eurasian continent beneath central and northern Taiwan created by the northward motion of the Philippine Sea plate is the upwelling path of the asthenosphere. The tomographic images also show the low-Vp anomalies spread widely under the coastal areas of Mainland China and Taiwan Strait. The structure of the crust and upper mantle suggests that the mountain building process in the central part of Taiwan is mainly attributed to the subduction–collision tectonics at the boundary between the Eurasian continental lithosphere and the subducting oceanic lithosphere of the PHS slab.

Deep structure of the central uplift belt in the Qiangtang terrane Tibetan Plateaufrom teleseismic P-wave tomography

We have determined a detailed three-dimensional P-wave velocity structure of the crust and upper mantle down to a depth of 200 km beneath the central uplift belt of Qiangtang by applying Fast Marching Teleseismic Tomography (FMTT) to 9532 high-quality P-wave arrival times. The travel time data were determined by careful interpretations of original seismograms of 506 teleseismic events recorded by a network of 54 broadband stations deployed in the Qiangtang terrane on the Tibetan Plateau during September 2008 and November 2010. Our tomographic imaging beneath the central uplift belt of Qiangtang shows a deep boundary rupture structure for which the lithospheric velocity structure is significantly different on the northern and southern sides. Tomographic images along 33癗 and 33.5癗 show that lithospheric structures beneath the sides of the central uplift belt have different scales. The crust at the sides of the central uplift belt exhibits E-W trending high-velocity anomalies. However very low velocities are found within the crust beneath the northern Qiangtang basin overlying and high-velocity anomalies in the lithospheric mantle. Imaging along north-south profiles (88.5癊 and 89.5癊) shows that the central uplift of Qiangtang is a significant tectonic boundary fault zone.

Ratio-to-moving-average seismograms: a strategy for improving correlation detector performance

SUMMARY Correlation detectors are becoming a standard method for identifying seismic signals from repeating sources. These highly sensitive, source-specific detectors frequently facilitate a reduction in the detection threshold by around an order of magnitude. Detections are typically declared when the value of the correlation coefficient (CC), or a related statistic, exceeds significantly some measure of the variability of values over a longer time window. The performance of correlation detectors is often compromised by the presence of short duration, high-amplitude signals, which influence excessively the value of the CC. We suggest replacing the original seismograms with waveforms in which the value of each sample is replaced by the ratio of that value to a centred moving average of absolute values of the original waveform. These ratio-to-moving-average (RMA) seismograms are relatively featureless over long time intervals, but resemble greatly the original waveforms over short time windows and hence still capture the characteristic seismic fingerprint of a given source. We demonstrate a correlation detection calculation, which fails due to the presence of a high-amplitude signal interfering with part of the correlation window, but which succeeds when RMA seismograms are used due to the diminished influence of the interfering signal. We also demonstrate an example from an aftershock sequence where the CC traces are heavily modulated due to the high dynamic range of the original waveforms. This makes the setting of detection thresholds difficult and results in multiple peaks, which do not correspond to events in the vicinity of the master event. Repeating the calculation using RMA seismograms results in CC traces with a more well-defined detection threshold and most of the spurious detections are lost. The ability to set lower detection thresholds without increasing greatly the number of false alarms facilitates the robust detection of lower magnitude events.

Tracing the Indian lithospheric mantle beneath central Tibetan Plateau using teleseismic tomography

We determined a detailed three-dimensional P-wave velocity structure of the crust and upper mantle down to 400 km depth beneath central Tibet by applying teleseismic tomography to 28,146 high-quality P-wave arrival times. The data were collected very carefully from the original seismograms of 1625 teleseismic events recorded by 131 broadband stations of five portable seismic networks deployed in central Tibet. Our results show that the Indian lithospheric mantle has subducted beneath central Tibet and its frontier has passed through the Bangong-Nujiang Suture and extended northward beneath the Qiangtang Terrane at 34° north latitude. A prominent low-velocity anomaly is revealed within the Tethyan Himalayan Sequences close to the Yarlung-Zangbo Suture which is the boundary between the Tethyan Himalayan Sequences and the Lhasa Terrane. The subduction of the Indian lithospheric mantle has caused the east–west extension in central Tibet. The Indian slab beneath the region has not sutured with the Asian lithospheric mantle in the north. The obtained tomographic images can explain many geological and geophysical features such as strong seismic anisotropy in the upper mantle and K-rich and ultra-potassic lavas in northern Tibet.

High-resolution 3-DP-wave model of the Alpine crust

The 3-D P-wave velocity structure of the Alpine crust has been determined from local earthquake tomography using a set of high-quality traveltime data. The application of an algorithm combining accurate phase picking with an automated quality assessment allowed the repicking of first arriving P-phases from the original seismograms. The quality and quantity of the repicked phase data used in this study allows the 3-D imaging of large parts of the Alpine lithosphere between 0 and 60 km depth. Our model represents a major improvement in terms of reliability and resolution compared to any previous regional tomographic studies of the Alpine crust. First-order anomalies like crust—mantle boundary (Moho) and the Ivrea body in the Western Alps are well resolved and in good agreement with previous studies. In addition, several (consistent) small-scale anomalies are visible in the tomographic image. A clear continuation of the lower European crust beneath the Adriatic Moho in the Central Alps is not observed in our results. The absence of such a signature may indicate the eclogitization of the subducted European lower crust in the Central Alps. In agreement with previous results, the additional analysis of focal depths in our new 3-D P-wave model shows that all studied earthquakes in the northern foreland have occurred within the European crust. Waveforms and focal depths suggest that at least one of the analysed events south of the Alps is located in the Adriatic mantle.

Exploring self-affine properties in seismograms

Self-affine properties have been observed in a large variety of rough profiles and time series from natural data sets. In this work, seismograms used for oil prospecting, which contain information of distinct subsurface features collected by seismic waves reflected or scattered at their interfaces, are taken into consideration. It is expected that any self-affine property, measured by the Hurst exponent H, depends on the depth. For each seismic trace, H is evaluated locally within a moving window, which is chosen narrow enough to reveal space dependency but also wide enough to display scale invariance. With the use of color code diagrams, it is possible to draw two-dimensional diagrams that show the local dependence of H for the analyzed seismogram. The reliability of the method is tested by the investigation of seismograms that contain ground roll components, as well as multiple reflections. The effect of different kinds of filter in the scaling properties is also investigated. In this case, comparisons are drawn among the diagrams obtained from original seismograms and those subjected to appropriate filter to eliminate spurious components.

Search for deep slab segments under Alaska

The evolution of the northeastern Pacific Basin is very complicated, featured by many ancient plates, microplates, ridge subduction and a series of slab windows. In this work, we collected 15,804 teleseismic arrival times from original seismograms of 889 distant earthquakes to determine a three-dimensional P-wave velocity structure down to 700 km depth beneath Alaska using a local and teleseismic joint inversion method. Our results show that the Pacific slab imaged as high-velocity (high-V) anomalies is subducting down to 300–400 km depth and it becomes deeper westwards under south-central and western Alaska. While in eastern Alaska, the Pacific slab is visible down to only about 90 km depth. Beneath western Alaska, high-V anomalies at 400–600 km depths are revealed, which represent the extinct Kula plate, and a gap between the subducted Pacific slab and the Kula slab is considered to represent the ancient Kula–Pacific spreading center. In southeastern Alaska, a large low-velocity (low-V) anomaly is found, which may reflect the upwelling mantle in the Pacific–Juan de Fuca slab window near the subducted edge of the Pacific plate. Our results support the existence of the Pacific–Juan de Fuca slab window suggested by the previous studies.

SWAS: A shear-wave analysis system for semi-automatic measurement of shear-wave splitting above small earthquakes

The complexity of shear wave-arrivals above small earthquakes makes the polarisations and time-delays of shear-wave splitting above small earthquakes difficult to measure. We report a semi-automatic shear-wave analysis system, SWAS, that appears to combine the benefits of both visual and automatic techniques. Initially, SWAS automatically estimates shear-wave polarisations and picks shear-wave arrivals by an expert system, which provides sufficiently accurate initial measurements for visual adjustment. SWAS then allows easy comparison and adjustment of picks between screen images of original seismograms, seismograms rotated into anisotropic polarisations, and polarisation diagrams (hodograms), with immediate plotting in various standard or non-standard configurations. This speeds up visual measurements by well over an order of magnitude, and typically allows records of almost all small earthquakes ( M ≥ −1.0) to be reliably measured for shear-wave splitting polarisations and time-delays. SWAS was developed and tested for data from the SIL seismic network in Iceland.

Phase identification analysis using directionality and rectilinearity functions in three component seismograms

An efficient technique for phase identification analysis is introduced using newly defined directionality and conventional rectilinearity functions obtained from three component seismograms. The directionality function,D, is defined asD(t)=1-|eb(t s )·e1(t)| where eb(t s ) and e1(t) are vectors obtained from two different short time windows at timest=t s andt=t i , respectively of a three-component seismogram set. The vector eb(t s ) is the eigenvector associated with the largest eigenvalue of the covariance matrix whose elements are values of covariance between component time series with a short time duration covering a basis phase at timet=t s . The vector e1(t) is the eigenvector associated with the largest eigenvalue of the covariance matrix computed for another phase at timet=t i . Computation of the directionality is done for a fixed basis phase at timet=t s and other phases to be compared at variable timest=t si . The first arrival phase is generally used as the basis phase, because this phase could be the most uncontaminated among all the phases in seismic records. The rectilinearity function obtained from the ratio of the largest and the second-largest eigenvalues of a covariance matrix in three component seismograms is also used in this study to increase confidence and accuracy of phase identification. The directionality and rectilinearity functions are computed along the time points of three component seismograms, and resulting time series are compared with original seismograms by inspection. These functions are not used as filters to be applied to seismograms by multiplications as used in previous researches, in order to avoid identification ambiguity caused by mixing of the characteristics of the functions and amplitudes of seismograms. This approach has improved performances in accuracy, stability, and computing time compared with the previously existing ones and is especially suitable for identification of phases in weak or moderate seismicity regions where routine identifications with large amount of data is not necessary. The technique was applied to a set of local, regional and teleseismic data, and body wave and surface wave phases were effectively identified.

Azimuthal variation of the P phase in Icelandic receiver functions

SUMMARY A curious observation has been made on radial receiver functions calculated from teleseisms recorded by 29 broad-band seismometers distributed over Iceland. The arrival time of the direct P phase of the radial receiver functions depends critically upon the azimuth of the teleseismic source. For a seismic station in West Iceland, the direct P phase of the radial receiver function arrives consistently later for easterly source azimuths than for westerly source azimuths. The reverse applies for stations in East Iceland. In the original seismograms, the delayed P phase of the receiver function appears up to 450 ms later on the radial than on the vertical component. The seismometer locations in East and West Iceland are separated by the Neovolcanic Zone, a constructive plate boundary. The delayed P phases occur for seismic rays travelling across this zone. However, it is not obvious how wave propagation across the plate boundary zone could cause the observed delays. The tentative explanation proposed here involves the regional dip of the Icelandic lava sequences towards the Neovolcanic Zone. A dipping interface at shallow depth results in a P–S converted phase arriving shortly after the P phase. These phases cannot be separated in the radial receiver functions, given the bandwidth of the observed signals. However, a calculation of receiver functions from estimates of the P, SV and SH wavefields clearly reveals a P–S converted phase at about 500 ms for easterly source azimuths in West Iceland and for westerly source azimuths in East Iceland. The amplitudes of the direct P phase and the P–S phase converted at a dipping interface would be expected to vary strongly with azimuth in accordance with the observed behaviour.