Ocean-bottom seismometers (OBSs) allow us to extend seismological research to the oceans to constrain offshore seismicity but also image the marine subsurface. A challenge is the high noise level on OBS records, which is created not only by bottom currents but also by the specific seismometer models used. We present a quantitative noise model for the LOBSTER OBS, which is the main instrument of the DEutscher Geräte-Pool für Amphibische Seismologie (DEPAS), currently the largest European OBS pool, stationed at AWI Bremerhaven. Studying sensor noise in vault conditions and current sensitivity at an oceanographic measurement mast, we can show that the previously reported high noise level of the instrument is caused by the original sensor (Güralp CMG-40T-OBS). We also show that a strong signal that has been reported between 1 and 5 Hz can be attributed to head-buoy cable strumming. We provide a currentdependent quantitative noise model that can be used for experiment design in future deployments and show that the performance of the pool OBS can be improved at moderate cost by replacing the CMG-40T-OBS with a sensor of a proven noise floor below 10 −8 nm=s 2 , for example, a Trillium compact.

Summary To constrain seismic anisotropy under and around the Alps in Europe, we study SKS shear-wave splitting from the region densely covered by the AlpArray seismic network. We apply a technique based on measuring the splitting intensity, constraining well both the fast orientation and the splitting delay. 4 years of teleseismic earthquake data were processed, from 723 temporary and permanent broadband stations of the AlpArray deployment including ocean-bottom seismometers, providing a spatial coverage that is unprecedented. The technique is applied automatically (without human intervention), and it thus provides a reproducible image of anisotropic structure in and around the Alpine region. As in earlier studies, we observe a coherent rotation of fast axes in the western part of the Alpine chain, and a region of homogeneous fast orientation in the Central Alps. The spatial variation of splitting delay times is particularly interesting though. On one hand, there is a clear positive correlation with Alpine topography, suggesting that part of the seismic anisotropy (deformation) is caused by the Alpine orogeny. On the other hand, anisotropic strength around the mountain chain shows a distinct contrast between the Western and Eastern Alps. This difference is best explained by the more active mantle flow around the Western Alps. The new observational constraints, especially the splitting delay, provide new information on Alpine geodynamics.

<p>To constrain seismic anisotropy under and around the Alps in Europe, we study SKS shear-wave splitting from the region densely covered by the AlpArray seismic network. We apply a technique based on measuring the splitting intensity, constraining well both the fast orientation and the splitting delay. 4 years of teleseismic earthquake data were processed automatically (without human intervention), from 724 temporary and permanent broadband stations of the AlpArray deployment including ocean-bottom seismometers. We have obtained an objective image of anisotropic structure in and around the Alpine region, at a spatial resolution that is unprecedented. As in earlier studies, we observe a coherent rotation of fast axes in the western part of the Alpine chain, and a region of homogeneous fast orientation in the central Alps.  The spatial variation of splitting delay times is particularly interesting. On one hand, there is a clear positive correlation with Alpine topography, suggesting that part of the seismic anisotropy (deformation) is caused by the Alpine orogeny. On the other hand, anisotropic strength around the mountain chain shows a distinct contrast between western and eastern Alps. This difference is best explained by the more active mantle flow around the Western Alps. We discuss earlier concepts of Alpine geodynamics in the light of these new observational constraints. </p>

<p>Seismic anisotropy is an important tool for studying geodynamic processes in the Earth, and a common way of constraining it is to analyse shear-wave splitting of seismological body-wave phases,<br>i.p. SKS. Different techniques exist to quantify shear-wave splitting, but they do not always give the same result, raising the question of how stable they are, and whether there are systematic biases. Furthermore, the strength of the splitting ("splitting delay") has generally been more difficult to determine than the other (the "fast orientation").<br>A robust technique for determining shear-wave splitting can be set up<br>based on the splitting intensity method. That technique can in particular also constrain the splitting delay well. Ambient noise can however lead to an underestimation of splitting delay, and it needs to be accounted for, e.g. by a least-squares Wiener filter.<br>We apply that modified splitting intensity method to data from the AlpArray. We have processed 3 years of teleseismic earthquake data for 336 stations of the AlpArray deployment and additional 315 stations of the Italian network to get a potentially broad and more complete image of anisotropic structures in and outside the Alpine region.<br>The technique makes restrictive assumptions, e.g. assuming single-layer anisotropy. Yet, the new constraints, especially the one of the splitting delay are rather useful for understanding the deformation under the mountain belt and around it.</p><p> </p>

<p>Detecting seismic signals and identifying their origin is more and more used for understanding environmental activity. This usually depends on a good signal/noise ratio (S/N), especially for the more distant sources.</p><p>A test area for detection and identification is the urban setting of the University of Vienna, a challenging environment with more than 4000 strong-acceleration events per day. These repetitive noise events would normally classify the site as "too noisy" for any advanced earthquake research.</p><p>With the real-time open database from Wiener Linien it is possible to attribute many of the repetitive seismic signals (e.g. on a Raspberry Shake Citizen Science Station) to the surrounding trams and train lines. The detection challenge was initiated in a Citizen Science Hackathon, where public interest sparked this research. The available train schedule and more than one year of continuous seismic records is sufficient to train and test a machine learning classifier which finds most characteristic features in the signals of commuter trains and trams, such as the energy in each frequency band.</p><p>The labeled dataset can be used to train our detection algorithm to find similar signals and to help determine whether a certain signal is present or not. An additional second seismic Raspberry Shake sensor is installed in the vicinity, to further constrain the directionality of the trains.</p><p>Studying the vibrations of train signals and solving the classification task of these repetitive patterns first can help develop robust methods<br>for seismically loud environments, and might lead to the detection of lower magnitude events such as regional earthquakes or landslides. </p>