@article{HammerBeyreutherOhrnberger2012,
  author    = {Hammer, Conny and Beyreuther, Moritz and Ohrnberger, Matthias},
  title     = {A seismic-event spotting system for volcano fast-response systems},
  series = {Bulletin of the Seismological Society of America},
  volume    = {102},
  journal   = {Bulletin of the Seismological Society of America},
  number    = {3},
  publisher = {Seismological Society of America},
  address   = {Albany},
  issn      = {0037-1106},
  doi       = {10.1785/0120110167},
  pages     = {948 -- 960},
  year      = {2012},
  abstract  = {Volcanic eruptions are often preceded by seismic activity that can be used to quantify the volcanic activity. In order to allow consistent inference of the volcanic activity state from the observed seismicity patterns, objective and time-invariant classification results achievable by automatic systems should be preferred. Most automatic classification approaches need a large preclassified data set for training the system. However, in case of a volcanic crisis, we are often confronted with a lack of training data due to insufficient prior observations. In the worst case (e. g., volcanic crisis related reconfiguration of stations), there are even no prior observations available. Finally, due to the imminent crisis there might be no time for the time-consuming process of preparing a training data set. For this reason, we have developed a novel seismic-event spotting technique in order to be less dependent on previously acquired data bases and classification schemes. We are using a learning-while-recording approach based on a minimum number of reference waveforms, thus allowing for the build-up of a classification scheme as early as interesting events have been identified. First, short-term wave-field parameters (here, polarization and spectral attributes) are extracted from a continuous seismic data stream. The sequence of multidimensional feature vectors is then used to identify a fixed number of clusters in the feature space. Based on this general description of the overall wave field by a mixture of multivariate Gaussians, we are able to learn particular event classifiers (here, hidden Markov models) from a single waveform example. To show the capabilities of this new approach we apply the algorithm to a data set recorded at Soufriere Hills volcano, Montserrat. Supported by very high classification rates, we conclude that the suggested approach provides a valuable tool for volcano monitoring systems.},
  language  = {en}
}
@article{HammerFaehOhrnberger2017,
  author    = {Hammer, Conny and F{\"a}h, Donat and Ohrnberger, Matthias},
  title     = {Automatic detection of wet-snow avalanche seismic signals},
  series = {Natural hazards : journal of the International Society for the Prevention and Mitigation of Natural Hazards},
  volume    = {86},
  journal   = {Natural hazards : journal of the International Society for the Prevention and Mitigation of Natural Hazards},
  publisher = {Springer},
  address   = {New York},
  issn      = {0921-030X},
  doi       = {10.1007/s11069-016-2707-0},
  pages     = {601 -- 618},
  year      = {2017},
  abstract  = {Avalanche activity is an important factor when estimating the regional avalanche danger. Moreover, a complete and detailed picture of avalanche activity is needed to understand the processes that lead to natural avalanche release. Currently, information on avalanche activity is mainly obtained through visual observations. However, this involves large uncertainties in the number and release times, influencing the subsequent analysis. Therefore, alternative methods for the remote detection of snow avalanches in particular in non-observed areas are highly desirable. In this study, we use the excited ground vibration to identify avalanches automatically. The specific seismic signature of avalanches facilitates the objective detection by a recently developed classification procedure. A probabilistic description of the signals, called hidden Markov models, allows the robust identification of corresponding signals in the continuous data stream. The procedure is based upon learning a general background model from continuous seismic data. Then, a single reference waveform is used to update an event-specific classifier. Thus, a minimum amount of training data is required by constructing such a classifier on the fly. In this study, we processed five days of continuous data recorded in the Swiss Alps during the avalanche winter 1999. With the restriction of testing large wet-snow avalanches only, the presented approach achieved very convincing results. We successfully detect avalanches over a large volume and distance range. Ninety-two percentage of all detections (43 out of 47) could be confirmed as avalanche events; only four false alarms are reported. We see a clear dependence of recognition capability on run-out distance and source-receiver distance of the observed events: Avalanches are detectable up to a source-receiver distance of eight times the avalanche length. Implications for analyzing a more comprehensive data set (smaller events and different flow regimes) are discussed in detail.},
  language  = {en}
}
@article{HammerOhrnbergerFaeh2013,
  author    = {Hammer, Conny and Ohrnberger, Matthias and Faeh, Donat},
  title     = {Classifying seismic waveforms from scratch: a case study in the alpine environment},
  series = {Geophysical journal international},
  volume    = {192},
  journal   = {Geophysical journal international},
  number    = {1},
  publisher = {Oxford Univ. Press},
  address   = {Oxford},
  issn      = {0956-540X},
  doi       = {10.1093/gji/ggs036},
  pages     = {425 -- 439},
  year      = {2013},
  abstract  = {Nowadays, an increasing amount of seismic data is collected by daily observatory routines. The basic step for successfully analyzing those data is the correct detection of various event types. However, the visually scanning process is a time-consuming task. Applying standard techniques for detection like the STA/LTAtrigger still requires the manual control for classification. Here, we present a useful alternative. The incoming data stream is scanned automatically for events of interest. A stochastic classifier, called hidden Markov model, is learned for each class of interest enabling the recognition of highly variable waveforms. In contrast to other automatic techniques as neural networks or support vector machines the algorithm allows to start the classification from scratch as soon as interesting events are identified. Neither the tedious process of collecting training samples nor a time-consuming configuration of the classifier is required. An approach originally introduced for the volcanic task force action allows to learn classifier properties from a single waveform example and some hours of background recording. Besides a reduction of required workload this also enables to detect very rare events. Especially the latter feature provides a milestone point for the use of seismic devices in alpine warning systems. Furthermore, the system offers the opportunity to flag new signal classes that have not been defined before. We demonstrate the application of the classification system using a data set from the Swiss Seismological Survey achieving very high recognition rates. In detail we document all refinements of the classifier providing a step-by-step guide for the fast set up of a well-working classification system.},
  language  = {en}
}
@article{BeyreutherHammerWassermannetal.2012,
  author    = {Beyreuther, Moritz and Hammer, Conny and Wassermann, Joachim and Ohrnberger, Matthias and Megies, Tobias},
  title     = {Constructing a hidden Markov Model based earthquake detector: application to induced seismicity},
  series = {Geophysical journal international},
  volume    = {189},
  journal   = {Geophysical journal international},
  number    = {1},
  publisher = {Wiley-Blackwell},
  address   = {Malden},
  issn      = {0956-540X},
  doi       = {10.1111/j.1365-246X.2012.05361.x},
  pages     = {602 -- 610},
  year      = {2012},
  abstract  = {The triggering or detection of seismic events out of a continuous seismic data stream is one of the key issues of an automatic or semi-automatic seismic monitoring system. In the case of dense networks, either local or global, most of the implemented trigger algorithms are based on a large number of active stations. However, in the case of only few available stations or small events, for example, like in monitoring volcanoes or hydrothermal power plants, common triggers often show high false alarms. In such cases detection algorithms are of interest, which show reasonable performance when operating even on a single station. In this context, we apply Hidden Markov Models (HMM) which are algorithms borrowed from speech recognition. However, many pitfalls need to be avoided to apply speech recognition technology directly to earthquake detection. We show the fit of the model parameters in an innovative way. State clustering is introduced to refine the intrinsically assumed time dependency of the HMMs and we explain the effect coda has on the recognition results. The methodology is then used for the detection of anthropogenicly induced earthquakes for which we demonstrate for a period of 3.9 months of continuous data that the single station HMM earthquake detector can achieve similar detection rates as a common trigger in combination with coincidence sums over two stations. To show the general applicability of state clustering we apply the proposed method also to earthquake classification at Mt. Merapi volcano, Indonesia.},
  language  = {en}
}
@article{HammerOhrnberger2012,
  author    = {Hammer, Conny and Ohrnberger, Matthias},
  title     = {Forecasting seismo-volcanic activity by using the dynamical behavior of volcanic earthquake rates},
  series = {Journal of volcanology and geothermal research},
  volume    = {229},
  journal   = {Journal of volcanology and geothermal research},
  number    = {11},
  publisher = {Elsevier},
  address   = {Amsterdam},
  issn      = {0377-0273},
  doi       = {10.1016/j.jvolgeores.2012.01.016},
  pages     = {34 -- 43},
  year      = {2012},
  abstract  = {We present a novel approach for short-term forecasting of volcano seismic activity. Volcanic earthquakes can be seen as a response mechanism of the earth crust to stresses induced by magma injection. From this point of view the temporal evolution of seismicity can be represented as a diffusion process which compensates pressure differences. By means of this dynamical approach we are able to estimate the system behavior in the near future which in turn allows us to forecast the evolution of the earthquake rate for the next time span from actual and past observations. For this purpose we model the earthquake rate as a random walk process embedded in a moving and deforming potential function. The center of the potential function is given by a moving average of the random walk's trace. We successfully apply this procedure to estimate the next day seismicity at Soufriere Hills volcano, Montserrat, over a time period of six years. When comparing the dynamical approach to the well known method of material failure forecast we find much better predictions of the critical stages of volcanic activity using the new approach.},
  language  = {en}
}
@article{NiemzCescaHeimannetal.2020,
  author    = {Niemz, Peter and Cesca, Simone and Heimann, Sebastian and Grigoli, Francesco and von Specht, Sebastian and Hammer, Conny and Zang, Arno and Dahm, Torsten},
  title     = {Full-waveform-based characterization of acoustic emission activity in a mine-scale experiment},
  series = {Geophysical journal international / the Royal Astronomical Society, the Deutsche Geophysikalische Gesellschaft and the European Geophysical Society},
  volume    = {222},
  journal   = {Geophysical journal international / the Royal Astronomical Society, the Deutsche Geophysikalische Gesellschaft and the European Geophysical Society},
  number    = {1},
  publisher = {Oxford Univ. Press},
  address   = {Oxford},
  issn      = {0955-419X},
  doi       = {10.1093/gji/ggaa127},
  pages     = {189 -- 206},
  year      = {2020},
  abstract  = {Understanding fracturing processes and the hydromechanical relation to induced seismicity is a key question for enhanced geothermal systems (EGS). Commonly massive fluid injection, predominately causing hydroshearing, are used in large-scale EGS but also hydraulic fracturing approaches were discussed. To evaluate the applicability of hydraulic fracturing techniques in EGS, six in situ, multistage hydraulic fracturing experiments with three different injection schemes were performed under controlled conditions in crystalline rock at the Aspo Hard Rock Laboratory (Sweden). During the experiments the near-field ground motion was continuously recorded by 11 piezoelectric borehole sensors with a sampling rate of 1 MHz. The sensor network covered a volume of 30x30x30 m around a horizontal, 28-m-long injection borehole at a depth of 410 m. To extract and characterize massive, induced, high-frequency acoustic emission (AE) activity from continuous recordings, a semi-automated workflow was developed relying on full waveform based detection, classification and location procedures. The approach extended the AE catalogue from 196 triggered events in previous studies to more than 19600 located AEs. The enhanced catalogue, for the first time, allows a detailed analysis of induced seismicity during single hydraulic fracturing experiments, including the individual fracturing stages and the comparison between injection schemes. Beside the detailed study of the spatio-temporal patterns, event clusters and the growth of seismic clouds, we estimate relative magnitudes and b-values of AEs for conventional, cyclic progressive and dynamic pulse injection schemes, the latter two being fatigue hydraulic fracturing techniques. While the conventional fracturing leads to AE patterns clustered in planar regions, indicating the generation of a single main fracture plane, the cyclic progressive injection scheme results in a more diffuse, cloud-like AE distribution, indicating the activation of a more complex fracture network. For a given amount of hydraulic energy (pressure multiplied by injected volume) pumped into the system, the cyclic progressive scheme is characterized by a lower rate of seismicity, lower maximum magnitudes and significantly larger b-values, implying an increased number of small events relative to the large ones. To our knowledge, this is the first direct comparison of high resolution seismicity in a mine-scale experiment induced by different hydraulic fracturing schemes.},
  language  = {en}
}
@article{HammerNeuberg2009,
  author    = {Hammer, Conny and Neuberg, J{\"u}rgen W.},
  title     = {On the dynamical behaviour of low-frequency earthquake swarms prior to a dome collapse of Soufri{\`e}re Hill volcano, Montserrat},
  issn      = {0094-8276},
  doi       = {10.1029/2008GL036837},
  year      = {2009},
  abstract  = {A series of low-frequency earthquake swarms prior to a dome collapse on Soufri{\`e}re Hills volcano, Montserrat, are investigated with the emphasis on event rate and amplitude behaviour. In a single swarm, the amplitudes of consecutive events tend to increase with time, while the rate of event occurrence accelerates initially and then decelerates toward the end of the swarm. However, when consecutive swarms are considered, the average event rates seem to follow the material failure law, and the time of the dome collapse can be successfully estimated using the inverse event rate. These patterns in amplitude and event rate are interpreted as fluctuations in magma ascent velocity, which result in both the generation of low-frequency events as well as cyclic ground deformation accompanying the swarm activity.},
  language  = {en}
}