This paper introduces recurrent neural networks (RNN), long short-term memory (LSTM), and gated recurrent unit (GRU) to detect and classify continuous sequences of volcano-seismic events at the Deception Island Volcano, Antarctica. A representative data set containing volcano-tectonic earthquakes, long-period events, volcanic tremors, and hybrid events was used to train these models. Experimental results show that RNN, LSTM, and GRU can exploit temporal and frequency information from continuous seismic data, attaining close to 90%, 94%, and 92% events correctly detected and classified. A second experiment is presented in this paper. The architectures described above, trained with data from campaigns of seismic records obtained in 1995-2002, have been tested with data from the recent seismic survey performed at the Deception Island Volcano in 2016-2017 by the Spanish Antarctic scientific campaign. Despite the variations in the geophysical properties of the seismic events within the volcano across eruptive periods, results provide good generalization accuracy. This result expands the possibilities of RNNs for real-time monitoring of volcanic activity, even if seismic sources change over time.

Deep neural networks (DNNs) could help to identify the internal sources of volcano-seismic events. However, direct applications of DNNs are challenging, given the multiple seismic sources and the small size of available datasets. In this paper, we propose a novel approach in the field of volcano seismology to classify volcano-seismic events based on fully connected DNNs. Two DNN architectures with different weights scheme initialization are studied: stacked denoising autoencoders and deep belief networks. Using a combined feature vector of linear prediction coefficients and statistical properties, we evaluate classification performance on seven different classes of isolated seismic events. These proposed architectures are compared to multilayer perceptron, support vector machine, and random forest. Experimental results show that DNNs can efficiently capture complex relationships of volcano-seismic data and achieve better classification performance with faster convergence when compared to classical models.

Domain-specific problems where data collection is an expensive task are often represented by scarce or incomplete data. From a machine learning perspective, this type of problems has been addressed using models trained in different specific domains as the starting point for the final objective-model. The transfer of knowledge between domains, known as transfer learning (TL), helps to speed up training and improve the performance of the models in problems with limited amounts of data. In this letter, we introduce a TL approach to classify isolated volcano-seismic signals at ``Volcán de Fuego'', Colima (Mexico). Using the well-known convolutional architecture (LeNet) as a feature extractor and a representative data set containing regional earthquakes, volcano-tectonic earthquakes, long-period events, volcanic tremors, explosions, and collapses, our proposal compares the generalization capabilities of the models when we only fine-tune the upper layers and fine-tune overall of them. Compared with the other state-of-the-art techniques, classification systems based on TL approaches provide good generalization capabilities (attaining nearly 94% of events correctly classified) and decreasing computational time resources.

Understanding how deep hierarchical models build their knowledge is a key issue in the usage of artificial intelligence to interpret the reality behind data. Depending on the discipline and models used, such knowledge may be represented in ways that are more or less intelligible for humans, limiting further improvements on the performance of the existing models. In order to delve into the characterization and modeling of volcano-seismic signals, this article emphasizes the idea of deciphering what and how recurrent neural networks (RNNs) model, and how this knowledge can be used to improve data interpretation. The key to accomplishing these objectives is both analyzing the hidden state dynamics associated with their hidden units as well as pruning/trimming based on the specialization of neurons. In this article, we process, analyze, and visualize the hidden states activation maps of two RNN architectures when managing different types of volcano-seismic events. As a result, the class-dependent discriminative behavior of most active neurons is analyzed, thereby increasing the comprehension of the detection and classification tasks. A representative dataset from the deception island volcano (Antarctica), containing volcano-tectonic earthquakes, long period events, volcanic tremors, and hybrid events, is used to train the models. Experimental analysis shows how neural activity and its associated specialization skills change depending on the architecture chosen and the type of event analyzed.

Monitoring continuous volcano-seismic signals is often performed by systems trained on scarce or incomplete datasets. From a machine learning perspective, these types of systems are typically built by learning from seismic records containing information not only on the volcanic dynamics, but also on the complex inner structure of the volcanic edifice. The dual nature of the information content presents a challenge when it comes to modeling events temporally. Here, we show that while existing Recurrent Neural Network-based monitoring systems recognise almost 90% of events annotated in seismic catalogues, the long-range temporal dependencies are still hard to model. We found that Dilated Recurrent Neural Networks based on multi-resolution dilated recurrent skip connections between layers, have the remarkable capability to simultaneously enhance the efficiency of the model, reducing the number of training parameters, while increasing the performance of the model when compared with classical Recurrent Neural Networks in sequence modeling tasks involving very long-term seismic records. Our results offer the potential to increase the reliability of monitoring tools despite the variations in the geophysical properties of the seismic events within the volcano across eruptive periods.