View over the Maars de Moya on Petite-Terre (Moya, Mayotte, 2012). © BRGM - Dominique Tardy

Earthquake swarm in Mayotte: Scientific FAQ

In these FAQ, the BRGM presents the knowledge it has collected on the earthquake swarm that began in Mayotte on 10 May 2018. They correspond to a preliminary analysis and supplement the FAQ published by the Mayotte Prefecture on its website.

What is an earthquake swarm?

An "earthquake swarm" is an episode of multiple earthquakes that occur in a particular zone over several days or weeks, as opposed to the usual pattern of one main shock followed by several smaller aftershocks. Because of their nature as a "swarm", these episodes cannot be characterised in terms of magnitude and duration.

Have there been other earthquake swarms in the world?

These phenomena are quite common and have been recorded in many regions of the world. Seismic swarms also occur in mainland France, for example in late 2017 in the Maurienne area. Earthquake swarms are not the same as the seismic sequences associated with several major and successive earthquakes (of comparable magnitude), each of which is followed by aftershocks. Those closest to Mayotte were observed in 1985 to the north of the Davie Ridge (offshore from Mozambique), then in 1989 and 2005 in Ethiopia (Afar region), in tectonically active zones associated with different rift systems.

Have these swarms produced any major earthquakes?

The earthquake swarms observed so far do not always reach their maximum at the same time. For example:

  • Sequence to the north of the Davie Ridge (1985): a severe magnitude 6.3 earthquake occurring at the start of the sequence was followed many smaller quakes.
  • Dôbi seismic crisis (Afar, 1989): a magnitude 6.2 earthquake occurred at the start of the crisis, followed over 48 hours by a succession of magnitude ≥6 earthquakes (6 altogether), each producing aftershocks. This is referred to as a seismic sequence.
  • The Manda-Hararo Dabbahu crisis (Afar region, volcano-tectonic episode in 2005-2010): a magnitude 5.6 earthquake occurred 6 after the start of the sequence. Other magnitude 4.9 to 5.4 earthquakes occurred as the sequence continued.

Could Mayotte experience an earthquake of greater magnitude than those already recorded? 

The record for Mayotte until now was the magnitude 5.2 earthquake in 1993. This has been substantially exceeded during the current sequence. The 1993 earthquake has been used as a reference for hazard studies and to define the seismicity zone officially. The entire département of Mayotte is within the "moderate" seismicity zone (3rd in rank in the French seismic zoning plan). This means, as in other "moderately" seismic zones in France (Oléron, Vendée, Massif central), that only "moderate" seismic observations are provided for, a fact that also accounts for the limited extent of current knowledge on local seismicity.

Today, we no longer have a local reference with which to define a possible maximum magnitude. It is therefore not possible to exclude the occurrence, during this swarm or any subsequent episode, of an earthquake of greater magnitude than those already observed.

Read the report on "Predicting the next destructive earthquake" ("Prédire où et quand aura lieu le prochain séisme destructeur") on the Plan Séisme website (in French)

Could these earthquakes form a new volcano?

Volcanic eruptions or the formation of dykes (when lava or magmatic rock rise through a fissure) have been observed following earthquake swarms, but they do not occur systematically. One of the hypotheses envisaged by scientists, as of September 2018, is that the current swarm is caused by volcano-tectonic events. Recent surveys at sea have confirmed that underwater eruptions are currently occurring and therefore coincide with the seismic swarm.

In fact, the swarm appears to have several compounded causes.

It is clear that the initial phenomenon was caused by a tectonic event for which hypocentre mechanisms at the source have been described and that are typical of ruptures along faults.

The volcanic component is being studied thanks to additions to the observation network and new measurements taken at sea. Information from the processing of GPS data available for the island (IGN data) by a team from the Geology Laboratory of the Ecole normale supérieure de Paris has shown that significant movements have occurred on Mayotte since July 2018. These movements could not be detected at the beginning of the crisis, but reveal the presence of a volcanic component as one of the causes of the current seismic sequence (see the summary note drafted by Pierre Briole). Furthermore, on 11 November, an atypical, very low frequency signal was detected by international networks. The signal was also visible at the Chiconi seismic station. Signals of this type are characteristic of volcanic phenomena. These phenomena can be observed directly because of the appearance of a new volcanic cone, fluid plumes and spills during marine campaigns. Rocks collected from the area are currently being analysed and this should provide more information on the nature of the eruption.

These observations therefore back up the hypothesis of a combination of tectonic and volcanic effects accounting for a geological phenomenon involving a seismic sequence and a volcanic event. The hypothesis will need to be confirmed by future scientific studies and different types of observations.

The eruption observed (volcanic cone and effusion) is located 50 km from land at a depth of 3500 m. No significant impact on Mayotte is currently expected, but these aspects require additional studies to monitor changes to the ongoing phenomenon.

Could a tsunami occur?

An earthquake can cause a tsunami when strong tremors, typically more than 6.5 in magnitude, combine with vertical movement of the ocean floor caused by rupturing, to form waves. These conditions do not appear to be present together in Mayotte as the movements triggered by the earthquakes observed seem to be essentially horizontal ("strike-slip" faulting).

Tsunamis can also be triggered when submarine landslips occur after the ocean bottom is destabilised by earth tremors or after unstable volcanic edifices have formed. The configuration of the ocean floor around Mayotte (which plunges very steeply beyond the coral barrier) is not conducive to the amplification of possible tsunami waves, but the tsunami risk is being assessed given that there are several possible sources.

How long is this earthquake swarm likely to last?

The duration of a swarm is impossible to predict. The measurements now being taken give the number and characteristics of the earthquakes occurring, but there is no law that can predict when the phenomenon will end.

In other cases of earthquake swarms, seismic activity has gradually lessened over several consecutive days. The fact that fewer earthquakes are observed over one or two days cannot be interpreted as indicating the end of the phenomenon.

Earthquake swarms described as examples in the literature (bibliographic study under way) and located in volcanic-tectonic contexts similar to that in Mayotte last from a few hours to several days. We have not yet identified any swarms of similar magnitudes or lasting for such a long time. Earthquake swarms in other contexts (in the Alps, for example) can last for a long time but with smaller magnitudes.

Where are the earthquakes located, and are they moving closer to Mayotte?

The location of these seismic events is determined by analysing the signals measured by the monitoring stations. Measurements carried out by BRGM since the beginning of the swarm show that they are located around two areas east of Petite-Terre. The monitoring carried out since the beginning of the crisis shows no evidence that the epicentres are moving closer together.

The new data, acquired during missions at sea, may lead to recalculating the location of epicentres, with models different from those used at the beginning of the phenomenon. These models assume a greater depth than the current model, but the overall distance between the source and the stations (hypocentral distance) is the same as that indicated by BRGM.

Why do different institutes give different magnitudes and locations?

Magnitudes and locations are not direct measurements but result from calculations based on data measured by seismological stations.

BRGM estimates the location and magnitude of earthquakes from 9 stations less than 1,400 km from the epicentres, including three stations in Mayotte (at a distance of 40 to 50 km), and two stations in Grande Comore (located about 270 km from the epicentres). The international networks use more distant seismic monitoring stations, the nearest being 480 km from Mayotte. The uncertainties are therefore greater.

In addition, there are different models for representing wave propagation, which leads to results that may vary from one establishment to another. Studies are underway to identify the most suitable models and hypotheses for the Mayotte context.

Why are earthquakes strongly felt even when they are moderate?

How a earthquake is felt depends at once on its characteristics, its effects on the spot and individual perceptions.  The following factors come into play:

  • the rupture mechanism at source: waves produced by earthquakes of the same magnitude can be amplified more strongly in the direction of the rupture (comparable to the Doppler effect). In a swarm, ruptures can be of different types. Further studies on this point are being considered;
  • the effects of wave propagation can differ geographically: North-South differences in perception are known in the Caribbean, but not documented in Mayotte;
  • local site effects can amplify ground movements depending on the topography or the nature of the soil.  These effects are known and documented in Mayotte;
  • effects due to the building where the earthquake is felt: amplification at higher levels, responses of the building and furnishings;
  • individual factors: it is recognised that perceptions differ with individuals and with the same individual depending on conditions: day/night, emotional state, activity, etc.

What are the causes of the earthquakes now occurring?

Seismic events occur regularly and quite frequently along the 500 km of the Comoros archipelago, including earthquakes of around magnitude 5 throughout the zone. The current episode is therefore in line with the fairly moderate seismic activity known to occur to the north of the Mozambique Channel.

Given current knowledge, the seismic events now occurring can be attributed to a combination of dynamic geological phenomena:

  • Tectonic causes: due to the relative movements of tectonic plates, some sectors in the region experience seismic activity along major structures, especially the East African Rift and the Davie Ridge to the west of the archipelago. Active tectonic structures associated with the Comoros Archipelago are still poorly documented, but the seismic events recorded testify to ruptures along their faults.
  • A volcanic component: this hypothesis is supported in particular by studies of movements on the surface and by marine observation campaigns.

We are therefore witnessing a geological phenomenon involving a seismic sequence and a volcanic event. The use of data recently acquired for scientific research will make it possible to determine the origin of these phenomena and their interactions.

What is the cause of the monochrome waves detected on 11 November?

On 11 November, a highly atypical signal was observed by international seismic stations at a distance of several thousand kilometres from Mayotte: an almost monochrome wave at very low frequency (period of 17 seconds, or 0.06Hz). The source of these waves appears to be the swarm zone. In general, waves of this kind that propagate over long distances are generated by high-magnitude earthquakes (magnitude 5 and above) at shallow depths (< 30 km). But in the swarm zone, only small earthquakes (less than 3.5), which cannot trigger a signal of this type, were observed at the relevant time. Furthermore, no strong tremors were sensed by the island's population.

 YTMZ, RAE55, MCHI in Mayotte, SBC in the Comoros, SBV and ABPO in Madagascar. The Mayotte stations recorded four successive small earthquakes but no strong tremors. The surface wave first appeared at MCHI, then SBC, then at the Malagasy stations. It was not seen at YTMZ and RAE55 because their sensors cannot detect waves of this type.

Signals at 6 stations: YTMZ, RAE55, MCHI in Mayotte, SBC in the Comoros, SBV and ABPO in Madagascar. The Mayotte stations recorded four successive small earthquakes but no strong tremors. The surface wave first appeared at MCHI, then SBC, then at the Malagasy stations. It was not seen at YTMZ and RAE55 because their sensors cannot detect waves of this type.

The mechanism that triggered the wave is so far unexplained. The hypotheses advanced by researchers, such as magma chamber resonance, have yet to be confirmed by additional data. However, the phenomenon observed seems to be a further indication of a volcanic component in the ongoing earthquake swarm.


Read the FAQ and safety advice distribute by the Mayotte Prefecture (in French)

BRGM - 3 avenue Claude-Guillemin - BP 36009 45060 Orléans Cedex 2 - France Tel.: +33 (0)2 38 64 34 34