Seismicity at active volcanoes provides crucial constraints on the
dynamics of magma systems and complex fault activation processes
preceding and during an eruption. We characterize time-dependent
spectral features of volcanic earthquakes at Axial Seamount with
unsupervised machine learning methods, revealing mixed frequency signals
that emerge from the continuous waveforms about 15 hours before eruption
onset. The events migrate along pre-existing fissures, suggesting that
they represent brittle crack opening driven by influx of magma or
volatiles. These results demonstrate the power of novel machine learning
algorithms to characterize subtle changes in magmatic processes
associated with eruption preparation, offering new possibilities for
forecasting Axial's anticipated next eruption. This novel method is
generalizable and can be employed to identify similar precursory signals
at other active volcanoes.