8
$\begingroup$

The Cumbre Vieja volcano of La Palma Island, Spain, started to erupt 3 months ago.

This week, the volcano stopped emitting lava and there remains only degassing processes. According to this news story, the Spanish National Geographic Institute (IGN) plans to declare the end of the eruption on 25th December if no more activity occurs.

That make me wonder: what determines the end of a volcanic eruption and how can it be predicted?

$\endgroup$
4

2 Answers 2

2
+50
$\begingroup$

The typical cycle of volcanic activity is

  • Filling of a deep magma chamber

The deep magma source is controlled by regional tectonics and how it occurs is not well understood in detail.

  • Movement of the magma to a reservoir at shallow depth

This is often accompanied by increased seismic activity, ground levels rising, changes in ground water levels and changes in heat flow.

  • Extrusion of the magma (now called lava once it erupts) and emptying of the shallow reservoir

If the magma comes in contact with ground water the eruption may be explosive. The release in pressure allows gasses dissolved in the magma to come out of solution (like taking the cap off a seltzer bottle) and blow huge amounts of rock into the sky. Lava that erupts through dry rocks, or lava without a lot of dissolved gasses, will mostly flow out with some fountaining at the site of the eruption.

There may, during the course of the eruption, be further filling of the shallow reservoir and later phases of the eruption sequence. Looking at the gas being emitted by the volcano is a common way of detecting the introduction of fresh magma at depth.

  • A winding down phase where the lava solidifies, seismic activity drops, emission of sulpher dioxide goes down, and ground levels return to pre-eruption levels.

This is interpreted as an exhaustion of the shallow magma reservoir. Until the links to the deep magma source are closed the volcano may erupt again. Prediction of such renewed activity decades, even centuries, in the future is not within current capabilities.

$\endgroup$
1
$\begingroup$

This is a partial answer to the "how can it be predicted" part of the question. A new paper (Duputel et al., 2023) shows how Very-Long-Period (VLP) seismic events can be used to infer the end of an eruption at Piton de la Fournaise (and probably at other volcanoes).

For all eruptions between 2014 and 2015 at Piton de la Fournaise, VLP events were detected primarily when the lava extrusion rate is rapidly decreasing (Fig. S30). This is particularly evident for the three later VLP swarms of the August-October 2015 eruptions, which systematically occurred before sudden drops in lava emissions. These swarms were accompanied by rapid summit deflations (Fig. 3), corresponding to the contraction of the dike above the shallow reservoir (see Fig. S20 and Fig. S22). Although the excitation process of the dike resonance is currently unknown, this suggests that VLP events are associated with rapid pressure variations in the volcano plumbing system. The detection of VLP events is therefore useful for monitoring purposes because they indicate changes in the magma flow that may lead to a permanent or temporary cessation of lava emissions.

The introduction also mentions another method currently used, that is the near real time eruption rate which is measured by satellites, but it comes with its own challenges:

Despite advances in volcano monitoring, it is currently difficult to forecast when an eruption will end. [...] Magma discharge rate can be quantified in near real time from satellite data and volcanic tremor amplitude (Coppola et al., 2009; Hibert et al., 2015). Using such observations, it is possible to predict the end date of eruptions by assuming that they are fed by a single magma reservoir releasing its elastic strain energy as it re-equilibrates with the lithostatic pressure (Bonny and Wright, 2017; Wadge, 1981). Although such forecast performs well for short lived eruptions at Piton de la Fournaise (Bonny and Wright, 2017), it does not work for longer eruptions that are associated with more complicated time-evolutions (Coppola et al., 2009).

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.