Using Artificial Neural Networks with Joint Muon-Gravity Datasets for Volcano Imaging and Monitoring

Imaging shallow subsurface anomalies at volcanoes and volcanic structures is important for understanding magma transport, and time-dependent imaging is an important step toward real-time hazard monitoring. In recent years, the use of machine learning as a novel, data-driven approach to addressing complex inverse problems in the geosciences has gained increasing attention, particularly in the field of seismology. Here we present a physics-based, machine learning method to integrate disparate geophysical datasets for shallow subsurface imaging. We have developed a methodology for imaging static density variations of volcanoes by generating synthetic cosmic-ray muon and gravity datasets using theoretical knowledge of the forward kernels, which we then use to train an artificial neural network (ANN) to interpret subsurface density anomalies. This synthetic data is generated using a suite of possible anomalous density structures, and the accuracy of our trained ANN is determined by comparing against the known forward calculation. Given a comprehensive suite of possible patterns, the ANN should be able to interpolate the best-fit anomalous pattern given data it has never seen before, such as those obtained from field measurements. We use previously published gravity and muography data from the Showa-Shinzan lava dome (Usu, Japan) to test our deep learning model and compare this methodology to a more traditional inversion. Our work thus far has focused on static imaging, however, the power of this approach is its generality, and we explore the feasibility of the ANN method for generating a time series of images if given time-varying geophysical observations. Another advantage of using a (supervised) physics-based approach is its applicability to a range of observables, such as seismic travel times and electrical conductivity, limited by the forward kernels that connect observations to physical parameters, such as density, temperature, composition, porosity, and saturation.