• Detection
• Neural
• Networks
• Microseismic
• Seismic

In the last years, the number of dense seismic networks deployed around the world has grown exponentially and will continue to grow in the next years, producing larger and larger datasets. Furthermore, the rising popularity of new technologies for seismic data acquisition such as fiber optics and nodal seismometers, are making seismological datasets grow in size and variety at an exceptionally fast rate, pushing the limit of current data storage infrastructures and, in particular, of data analysis procedures. This data explosion lead seismology to enter in the so-called BIGDATA era. Among the different seismological applications where these massive datasets are usually collected, fluid-induced seismicity studies (both natural and induced) are certainly the most relevant and are a perfect playground for data intensive techniques. In these applications we generally deal with seismic sequences characterized by a large number of weak earthquakes overlapping each other or with short inter-event times; in these cases, pick-based detection and location methods may struggle to correctly assign picks to phases and events, and errors can lead to missed detections and/or reduced location resolution, which can have significant consequences if real-time seismicity information are used for risk assessment frameworks and, more in general, in the interpretation of the evolution of seismicity in the space-time-magnitude domain. Within this MSc thesis I have developed a new workflow for automatic and robust microseismic event detection and location that combines waveform-based methods with Deep-Learning (DL) techniques, with the aim to obtain more reliable microseismicity catalogues. In particular I use, Convolutional Neural Networks (CNNs), successfully employed for classification of images and/or speech signals to analyze coherence matrices (the output of waveform-stacking methods) and classify seismic events from noise, hence to detect earthquakes in an automated fashion without any user input (Fig.1). In this way we remove the dependency on fixed or dynamic threshold levels within the detection procedure that may generate missed detections, if the threshold is too high, or many false events if it is too low.