Ambient Seismic Noise Analysis Associated with the 2010 Eruption of Merapi Volcano Using Temporal Variations of Randomness and Background Noise

We present the combination of permutation entropy (PE) and power spectral density (PSD) analysis on continuous seismic data recorded by short-period seismic stations during the 2010 Merapi volcano eruption. The calculation of PE aims at characterizing the randomness level in seismic noise, while the PSD parameters use to detect the background noise level in various frequency bands. It was previously observed that a significant reduction of randomness before the volcano eruption could be indicated as one of the short-term precursors due to the lack of high frequencies (>1 Hz) in the noise wave-field caused by high absorption losses as the hot magma uprises to the upper crust. The results show no significant reduction in signal randomness before the eruption series. The characteristic of events during the preeruptive period and the crisis tends to be chaotic (PE in the range 0.9 to 1). Further calculations show that the standard deviation in PE decreased in four days before the first eruption onset on 26 October. PE was stable at the highest values (very close to 1) and gradually returned to the previous fluctuation after the eruption onset. The level of background noise in the low- and high-frequency bands appeared to have the same tendency. The two main eruptions correspond to the two highest peaks of noise levels.