Cyclic activity of the Fuego de Colima volcano (Mexico): insights from satellite thermal data and nonlinear models

Abstract

Abstract. The Fuego de Colima volcano (Mexico) shows a complex eruptive behavior, with periods of rapid and slow lava dome growth punctuated by explosive activity. We reconstructed the weekly discharge rate average between 1998 and 2018 by means of satellite thermal data integrated with published discharge rate data. By using spectral and wavelet analysis, we found a multiyear long-term, multi-month intermediate-term, and multi-week short-term cyclic behavior during the period of the investigated eruptive activity like that of many other dome-forming volcanoes. We use numerical modeling in order to investigate the nonlinear cyclic eruptive behavior considering a magma feeding system composed of a dual or a single magma chamber connected to the surface through an elastic dyke developing into a cylinder conduit in the shallowest part. We investigated cases in which the periodicity is controlled by (i) the coupled deep–shallow magma reservoirs, (ii) the single shallow chamber, and (iii) the elastic shallow dyke when it is fed by a fixed influx rate or constant pressure. Due to the limitations of the current modeling approach, there is no single configuration that can reproduce all the periodicities on the three different timescales. The model outputs indicate that the observed multiyear periodicity (1.5–2.5 years) can be described by the fluctuations controlled by a shallow magma chamber with a volume of 20–50 km3 coupled with a deep reservoir of ca. 500 km3, connected through a deep elastic dyke. The multi-month periodicity (ca. 5–10 months) appears to be controlled by the shallow magma chamber for the same range of volumes. The short-term multi-week periodicity (ca. 2.5–5 weeks) can be reproduced considering a fixed influx rate or constant pressure at the base of the shallower dyke. This work provides new insights on the nonlinear cyclic behavior of Fuego de Colima and a general framework for comprehension of the eruptive behavior of andesitic volcanoes.

Stochastic Modeling of Explosive Eruptive Events at Galeras Volcano, Colombia

“…Many of the results found in this study on Galeras may be explained by a simple conceptual model of a volcanic system fed by two interconnected shallow and deep magma reservoirs, as already proposed for e.g., Soufriere Hills, Montserrat (Melnik and Costa, 2014), or Colima, Mexico (Massaro et al, 2019). The deep reservoir feeds the shallow one through a dyke-like structure, and the magma transfer from the deep to the shallow reservoir is strongly controlled by the pressure gradient between the two reservoirs.…”

Detecting Multiscale Periodicity From the Secular Effusive Activity at Santiaguito Lava Dome Complex (Guatemala)

“…These inferences are in good agreement with the results of Wallace et al (2020), who explored the shift in eruption style at Santiaguito from early-2015 to late-2016, showing an escalation in explosivity which could have been in uenced by the injection of a higher temperature, relatively volatile-rich magma into the feeding system. With respect to other studied volcanic systems (Melnik and Costa 2014; Massaro et al 2019), the behaviour of Santiaguito lava dome complex can be even more complicated as the upper feeding system is connected to multiple vents. This is because clustered vents may re ect different conduits characterizing the shallow feeding system, which may contribute to dissipation of pressurization differently from a single conduit system.…”

“…As in Massaro et al (2019), the heat ux radiated from the Santiaguito dome complex was converted into averaged magma discharge rates to reconstruct the yearly, monthly, and weekly time series from February 2000 to August 2021 (Fig. 2b-c-d).…”

“…Following Torrence and Compo (1998) and Massaro et al (2019), we applied the wavelet analysis on our yearly, monthly, and weekly averaged discharge rate time series (Fig. 2).…”

“…The criterion for applying the local wavelet analysis is very similar to those employed with classic spectral methods since it is a generalization of the Fourier transform. Successively, this can be compared with the "global" wavelet power spectrum, which is de ned as the averaged variance contained in all wavelet coe cients of the same frequency (i.e., Torrence and Compo 1998;Cazelles et al 2008;Massaro et al 2019).…”