@article{KriegerowskiPetersenVasyuraBathkeetal.2018, author = {Kriegerowski, Marius and Petersen, Gesa Maria and Vasyura-Bathke, Hannes and Ohrnberger, Matthias}, title = {A Deep Convolutional Neural Network for Localization of Clustered Earthquakes Based on Multistation Full Waveforms}, series = {Seismological research letters}, volume = {90}, journal = {Seismological research letters}, number = {2}, publisher = {Seismological Society of America}, address = {Albany}, issn = {0895-0695}, doi = {10.1785/0220180320}, pages = {510 -- 516}, year = {2018}, abstract = {Earthquake localization is both a necessity within the field of seismology, and a prerequisite for further analysis such as source studies and hazard assessment. Traditional localization methods often rely on manually picked phases. We present an alternative approach using deep learning that once trained can predict hypocenter locations efficiently. In seismology, neural networks have typically been trained with either single-station records or based on features that have been extracted previously from the waveforms. We use three-component full-waveform records of multiple stations directly. This means no information is lost during preprocessing and preparation of the data does not require expert knowledge. The first convolutional layer of our deep convolutional neural network (CNN) becomes sensitive to features that characterize the waveforms it is trained on. We show that this layer can therefore additionally be used as an event detector. As a test case, we trained our CNN using more than 2000 earthquake swarm events from West Bohemia, recorded by nine local three-component stations. The CNN successfully located 908 validation events with standard deviations of 56.4 m in east-west, 123.8 m in north-south, and 136.3 m in vertical direction compared to a double-difference relocated reference catalog. The detector is sensitive to events with magnitudes down to M-L = -0.8 with 3.5\% false positive detections.}, language = {en} } @misc{WoithPetersenHainzletal.2018, author = {Woith, Heiko and Petersen, Gesa Maria and Hainzl, Sebastian and Dahm, Torsten}, title = {Review: Can Animals Predict Earthquakes?}, series = {Bulletin of the Seismological Society of America}, volume = {108}, journal = {Bulletin of the Seismological Society of America}, number = {3A}, publisher = {Seismological Society of America}, address = {Albany}, issn = {0037-1106}, doi = {10.1785/0120170313}, pages = {1031 -- 1045}, year = {2018}, abstract = {In public perception, abnormal animal behavior is widely assumed to be a potential earthquake precursor, in strong contrast to the viewpoint in natural sciences. Proponents of earthquake prediction via animals claim that animals feel and react abnormally to small changes in environmental and physico-chemical parameters related to the earthquake preparation process. In seismology, however, observational evidence for changes of physical parameters before earthquakes is very weak. In this study, we reviewed 180 publications regarding abnormal animal behavior before earthquakes and analyze and discuss them with respect to (1) magnitude-distance relations, (2) foreshock activity, and (3) the quality and length of the published observations. More than 700 records of claimed animal precursors related to 160 earthquakes are reviewed with unusual behavior of more than 130 species. The precursor time ranges from months to seconds prior to the earthquakes, and the distances from a few to hundreds of kilometers. However, only 14 time series were published, whereas all other records are single observations. The time series are often short (the longest is 1 yr), or only small excerpts of the full data set are shown. The probability density of foreshocks and the occurrence of animal precursors are strikingly similar, suggesting that at least parts of the reported animal precursors are in fact related to foreshocks. Another major difficulty for a systematic and statistical analysis is the high diversity of data, which are often only anecdotal and retrospective. The study clearly demonstrates strong weaknesses or even deficits in many of the published reports on possible abnormal animal behavior. To improve the research on precursors, we suggest a scheme of yes and no questions to be assessed to ensure the quality of such claims.}, language = {en} } @article{PetersenCescaKriegerowski2019, author = {Petersen, Gesa Maria and Cesca, Simone and Kriegerowski, Marius}, title = {Automated quality control for large seismic networks}, series = {Seismological research letters}, volume = {90}, journal = {Seismological research letters}, number = {3}, publisher = {Seismological Society of America}, address = {Albany}, organization = {AlpArray Working Grp}, issn = {0895-0695}, doi = {10.1785/0220180342}, pages = {1177 -- 1190}, year = {2019}, abstract = {As a consequence of the rapid growing worldwide seismic data set, a huge variety of automatized data-processing methods have been developed. To perform automatized waveform-based seismological studies aiming for magnitudes or source process inversion, it is crucial to identify network stations with erroneous transfer functions, gain factors, or component orientations. We developed a new tool dedicated to automated station quality control of dense seismic networks and arrays. The python-based AutoStatsQ toolbox uses the pyrocko seismic data-processing environment. The toolbox automatically downloads data and metadata for selected teleseismic events and performs different tests. As a result, relative gain factors, sensor orientation corrections, and reliable frequency bands are computed for all stations in a chosen time period. Relative gain factors are calculated for all stations and events in a time domain based on maximum P-phase amplitudes. A Rayleigh-wave polarization analysis is used to identify deviating sensor orientations. The power spectra of all stations in a given frequency range are compared with synthetic ones, accessing Global Centroid Moment Tensor (CMT) solutions. Frequency ranges of coinciding synthetic and recorded power spectral densities (PSDs) may serve as guidelines for choosing band-pass filters for moment tensor (MT) inversion and help confirm the corner frequency of the instrument. The toolbox was applied to the permanent and temporary AlpArray networks as well as to the denser SWATH-D network, a total of over 750 stations. Stations with significantly deviating gain factors were identified, as well as stations with inverse polarity and misorientations of the horizontal components. The tool can be used to quickly access network quality and to omit or correct stations before MT inversion. Electronic Supplement: List of teleseismic events and tables of median, mean, and standard deviation of relative gain factors, and figures of relative gain factors of all event-station pairs, waveform example showing inverse polarity of horizontal components on ZS.D125, histograms of median, mean, and standard deviation of the correction angles, examples of synthetic and recorded frequency spectra of ZS.D046 and NI.VINO.}, language = {en} } @article{BueyuekakpınarAktarPetersenetal.2021, author = {B{\"u}y{\"u}kakp{\i}nar, P{\i}nar and Aktar, Mustafa and Petersen, Gesa Maria and K{\"o}seoğlu, Ay{\c{s}}eg{\"u}l}, title = {Orientations of broadband stations of the KOERI seismic network (Turkey) from two independent methods}, series = {Seismological research letters / Seismological Society of America}, volume = {92}, journal = {Seismological research letters / Seismological Society of America}, number = {3}, publisher = {Seismological Society of America}, address = {Boulder, Colo.}, issn = {0895-0695}, doi = {10.1785/0220200362}, pages = {1512 -- 1521}, year = {2021}, abstract = {The correct orientation of seismic sensors is critical for studies such as full moment tensor inversion, receiver function analysis, and shear-wave splitting. Therefore, the orientation of horizontal components needs to be checked and verified systematically. This study relies on two different waveform-based approaches, to assess the sensor orientations of the broadband network of the Kandilli Observatory and Earthquake Research Institute (KOERI). The network is an important backbone for seismological research in the Eastern Mediterranean Region and provides a comprehensive seismic data set for the North Anatolian fault. In recent years, this region became a worldwide field laboratory for continental transform faults. A systematic survey of the sensor orientations of the entire network, as presented here, facilitates related seismic studies. We apply two independent orientation tests, based on the polarization of P waves and Rayleigh waves to 123 broadband seismic stations, covering a period of 15 yr (2004-2018). For 114 stations, we obtain stable results with both methods. Approximately, 80\% of the results agree with each other within 10 degrees. Both methods indicate that about 40\% of the stations are misoriented by more than 10 degrees. Among these, 20 stations are misoriented by more than 20 degrees. We observe temporal changes of sensor orientation that coincide with maintenance work or instrument replacement. We provide time-dependent sensor misorientation correction values for the KOERI network in the supplemental material.}, language = {en} } @article{NiemzDahmMilkereitetal.2021, author = {Niemz, Peter and Dahm, Torsten and Milkereit, Claus and Cesca, Simone and Petersen, Gesa Maria and Zang, Arno}, title = {Insights into hydraulic fracture growth gained from a joint analysis of seismometer-derived tilt signals and scoustic emissions}, series = {Journal of geophysical research : Solid earth}, volume = {126}, journal = {Journal of geophysical research : Solid earth}, number = {12}, publisher = {American Geophysical Union}, address = {Washington}, issn = {2169-9313}, doi = {10.1029/2021JB023057}, pages = {14}, year = {2021}, abstract = {Hydraulic fracturing is performed to enhance rock permeability, for example, in the frame of geothermal energy production or shale gas exploitation, and can potentially trigger induced seismicity. The tracking of increased permeabilities and the fracturing extent is often based on the microseismic event distribution within the stimulated rock volume, but it is debated whether the microseismic activity adequately depicts the fracture formation. We are able to record tilt signals that appear as long-period transients (<180 s) on two broadband seismometers installed close (17-72 m) to newly formed, meter-scale hydraulic fractures. With this observation, we can overcome the limitations of the microseismic monitoring alone and verify the fracture mapping. Our analysis for the first time combines a catalog of previously analyzed acoustic emissions ([AEs] durations of 20 ms), indirectly mapping the fractures, with unique tilt signals, that provide independent, direct insights into the deformation of the rock. The analysis allows to identify different phases of the fracturing process including the (re)opening, growth, and aftergrowth of fractures. Further, it helps to differentiate between the formation of complex fracture networks and single macrofractures, and it validates the AE fracture mapping. Our findings contribute to a better understanding of the fracturing processes, which may help to reduce fluid-injection-induced seismicity and validate efficient fracture formation.
Plain Language Summary Hydraulic fracturing (HF) describes the opening of fractures in rocks by injecting fluids under high pressure. The new fractures not only can facilitate the extraction of shale gas but can also be used to heat up water in the subsurface in enhanced geothermal systems, a corner stone of renewable energy production. The fracture formation is inherently accompanied by small, nonfelt earthquakes (microseismic events). Occasionally, larger events felt by the population can be induced by the subsurface operations. Avoiding such events is important for the acceptance of HF operations and requires a detailed knowledge about the fracture formation. We jointly analyze two very different data sets recorded during mine-scale HF experiments: (a) the tilting of the ground caused by the opening of the fractures, as recorded by broadband seismometers-usually deployed for earthquake monitoring-installed close to the experiments and (b) a catalog of acoustic emissions, seismic signals of few milliseconds emitted by tiny cracks around the forming hydraulic fracture. The novel joint analysis allows to characterize the fracturing processes in greater detail, contributing to the understanding of the physical processes, which may help to understand fluid-injection-induced seismicity and validate the formation of hydraulic fractures.}, language = {en} }