Geophys. J. Int. (2022) 228, 1281–1293 https://doi.org/10.1093/gji/ggab392
Advance Access publication 2021 September 24
GJI Seismology
Mayotte seismic crisis: building knowledge in near real-time by
combining land and ocean-bottom seismometers, first results
Jean-Marie Saurel ,
1
Eric Jacques,
1
Chastity Aiken ,
2
Anne Lemoine ,
3
Lise Retailleau,
1,4
Aude Lavayssi
`
ere,
1
Oc
´
eane Foix,
2
Anthony Dofal ,
1,5
Ang
`
ele Laurent,
1
Nicolas Mercury,
3,6
Wayne Crawford,
1
Arnaud Lemarchand,
1
Romuald Daniel,
1
Pascal Pelleau,
2
Maxime B
`
es de Berc,
6
Gr
´
egoire Dectot,
7
Didier Bertil,
3
Agathe Roull
´
e,
3
C
´
eleste Broucke,
8
Alison Colombain,
3
H
´
el
`
ene Jund,
8
Simon Besanc¸on,
1
Pierre Guyavarch,
2
Philippe Kowalski,
1,4
Micka
¨
el Roudaut,
2
Ronan Apprioual,
2
Jean Battaglia,
9
Soumya Bodihar,
1
Patrice Boissier,
1,4
Marie
Paule Bouin,
1
Christophe Brunet,
1,4
K
´
evin Canjamale,
1,4
Philippe Catherine,
1,4
Nicolas Desfete,
1,4
C
´
ecile Doubre,
6
R
´
emi Dretzen,
8
Tom Dumouche,
1
Philippe Fernagu,
2
Val
´
erie Ferrazzini,
1,4
Fabrice R. Fontaine,
1,5
Arnaud Gaillot,
2
Louis G
´
eli,
2
Cyprien Griot,
1,4
Marc Grunberg,
8
Emre Can Guzel,
10
Roser Hoste-Colomer,
3
Sophie Lambotte,
6
Fr
´
ed
´
eric Lauret,
1,4
F
´
elix L
´
eger,
1
Emmanuel Maros,
2
Aline Peltier,
1,4
J
´
er
ˆ
ome Vergne,
8
Claudio Satriano,
1
Fr
´
ed
´
eric Tronel,
7
J
´
er
ˆ
ome Van der Woerd,
6
Yves Fouquet,
2
Stephan J. Jorry,
2
Emmanuel Rinnert,
2
Isabelle Thinon
11
and
Nathalie Feuillet
1
1
Universit
´
e de Paris, Institut de Physique du Globe de Paris, CNRS, F-75005 Paris, France. E-mail: saurel@ipgp.fr
2
IFREMER, Centre de Bretagne, –Unit
´
eG
´
eosciences Marines, 1625 Rte de Ste Anne, 29280 Plouzan
´
e, France
3
BRGM, French Geological Survey, Risk and Prevention Division, F- 45100 Orl
´
eans, France
4
Observatoire volcanologique du Piton de la Fournaise, Institut de Physique du Globe de Paris, F-97418 La Plaine des Cafres, La R
´
eunion, France
5
Universit
´
edeLaR
´
eunion, Laboratoire G
´
eoSciences R
´
eunion, F-97744 Saint Denis, La R
´
eunion, France
6
ITES, Institut Terre Environnement de Strasbourg, UMR 7063, CNRS Universit
´
e de Strasbourg, 5,rueRen
´
e Descartes, 67084 Strasbourg, France
7
BRGM, French Geological Survey, Regional Division (Mayotte), F-97600 Mamoudzou, Mayotte, France
8
EOST, Universit
´
e de Strasbourg/CNRS, 5 rue Descartes, 67084 Strasbourg Cedex, France
9
Universit
´
e Clermont Auvergne, CNRS, IRD, OPGC, Laboratoire Magmas et Volcans, F-63000 Clermont-Ferrand, France
10
Istanbul Technical University, Faculty of Electrical and Electronics Engineering - Graduate School, Electronics and Communications Engineering
Department, 34469 Maslak/Istanbul, Turkey
11
BRGM, French Geological Survey, Georesources, F-45100 Orl
´
eans, France
Accepted 2021 September 22. Received 2021 September 14; in original form 2021 February 24
SUMMARY
The brutal onset of seismicity offshore Mayotte island North of the Mozambique Channel,
Indian Ocean, that occurred in May 2018 caught the population, authorities and scientific
community off guard. Around 20 potentially felt earthquakes were recorded in the first 5 d,
up to magnitude M
w
5.9. The scientific community had little pre-existing knowledge of the
seismic activity in the region due to poor seismic network coverage. During 2018 and 2019,
the MAYOBS/REVOSIMA seismology group was progressively built between four French
research institutions to improve instrumentation and data sets to monitor what we know now
as an on-going exceptional submarine basaltic eruption. After the addition of 3 medium-band
stations on Mayotte island and 1 on Grande Glorieuse island in early 2019, the data recovered
from the Ocean Bottom Seismometers were regularly processed by the group to improve the
location of the earthquakes detected daily by the land network. We first built a new local 1-D
velocity model and established specific data processing procedures. The local 1.66 low V
P
/V
S
ratio we estimated is compatible with a volcanic island context. We manually picked about
C
The Author(s) 2021. Published by Oxford University Press on behalf of The Royal Astronomical Society. This is an Open Access
article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which
permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
1281
Downloaded from https://academic.oup.com/gji/article/228/2/1281/6374867 by Ifremer, Bibliothèque La Pérouse user on 09 November 2021
1282 J.-M. Saurel et al.
125 000 P and S phases on land and sea bottom stations to locate more than 5000 events
between February 2019 and May 2020. The earthquakes outline two separate seismic clusters
offshore that we named Proximal and Distal. The Proximal cluster, located 10 km offshore
Mayotte eastern coastlines, is 20–50 km deep and has a cylindrical shape. The Distal cluster
start 5 km to the east of the Proximal cluster and extends below Mayotte’s new volcanic edifice,
from 50 to 25 km depth. The two clusters appear seismically separated, however our data set
is insufficient to firmly demonstrate this.
Key words: Indian Ocean; Volcano seismology; Volcano monitoring; Africa; Remote sensing
of volcanoes.
1. INTRODUCTION
Before 10 May 2018, Mayotte island, part of the volcanic Comoros
archipelago in the North Mozambique Channel of the Indian Ocean
(Fig. 1), was not considered as a significantly seismically active area
(Bertil & Regnoult 1998). The last reported widely felt earthquakes
occurred around 30 km west of Mayotte: a moment magnitude (M
w
)
5.0 event on 9 September 2011 (EMS98 intensity V estimated) and a
magnitude (M) 5.2 event on 1 December 1993 (Lambert 1997) with
moderate damages (EMS98 intensity VI estimated). Together with
an unfelt M 5.1 event on 23 March 1993, 80 km southwest of May-
otte, these were the only M5+ earthquakes recorded within 100 km
of the island since the advent of the global seismological networks
in 1964 (Storchak et al. 2017;ISC2020). As a consequence, only
one real-time seismic station (RA.YTMZ; R
´
esif 1995) was installed
on the island at the onset of the 2018 seismic crisis. This station was
deployed by BRGM (Bureau de recherches g
´
eologiques et mini
`
eres,
the French geological survey) for the French accelerometric moni-
toring network (R
´
esif-RAP, P
´
equegnat et al. 2008).
On 10 May 2018, the first felt earthquake, quickly followed by
many others, surprised inhabitants. More than 130 M4+ earth-
quakes were recorded in the following months, with the strongest
being a M
w
5.9 on 15 May 2018 (GCMT project, Dziewonski et
al. 1981;Ekstr
¨
om et al. 2012; Lemoine et al. 2020a). After about
50 d of very intense seismic activity, this unprecedented seismic
sequence continued less intensively. During the summer of 2018,
Global Navigation Satellite System (GNSS) data from the locally
continuously recording sites began to show rapid surface displace-
ments of the island (Briole 2018;Cescaet al. 2020; Lemoine et al.
2020a). Their elastic modelling evidence a large regional deflation
centred east of Mayotte’s shorelines (Cesca et al. 2020; Lemoine
et al. 2020a). On 11 November 2018, a very low frequency tremor
was recorded worldwide (Satriano et al. 2019; Cesca et al. 2020;
Lemoine et al. 2020a), confirming that the seismic crisis was very
likely of magmatic origin. This was later confirmed by the discov-
ery of a new submarine volcanic edifice offshore Mayotte during
the MAYOBS1 scientific expedition onboard RV Marion Dufresne
in May 2019 (Feuillet et al. 2021). This large eruption began either
on June 18 (Cesca et al. 2020) or on 3 July 2018 (Lemoine et al.
2020a).
Since the onset of the crisis, collaborations were progressively
established between French research institutes to improve the under-
standing and knowledge of the on-going crisis. These collaborations
enhanced the seismic monitoring of the region, which included in-
stalling additional real-time sites onshore, access to real-time data
recorded by existing regional stations and offshore deployments.
During the first year of the crisis, the monitoring network thus
evolved rapidly (Lemoine et al. 2020a). In March 2019, 1 month
after funding, four seismic stations were installed onshore (three
on Mayotte island and one on Grande Glorieuse island) and six
Ocean Bottom Seismometers (OBS) were deployed offshore, within
a radius of 40 km around the seismically active area. A seismol-
ogy team was created among the researchers, engineers, and stu-
dents belonging to the participating French institutions (BRGM;
Institut de Physique du Globe de Paris—IPGP; Institut Franc¸ais
de Recherche pour l’Exploitation de la Mer—IFREMER; Insti-
tut National des Sciences de l’Univers du Centre National de la
Recherche Scientifique—INSU-CNRS). In July 2019, the Mayotte
seismo-volcanic monitoring network (R
´
eseau de surveillance vol-
canologique et sismologique de Mayotte—REVOSIMA) was cre-
ated with all four institutions. The aim of the team is to process
newly acquired data as quickly as possible and to obtain first hand
results in almost real-time, to improve the daily monitoring and the
knowledge of the volcano-seismic crisis.
In this paper, we review the local seismic network improvements
since the beginning of the crisis and our scientific developments. We
detail how this collaborative work was orchestrated for maximum
efficiency and how it led to an improved local velocity model and
a seismic catalogue from February 2019 up to May 2020 to better
document the Mayotte 2018-ongoing seismo-volcanic crisis.
2. SEISMIC NETWORK EVOLUTION
AND DATA PROCESSING
From May 2018 to June 2019, the Mayotte local real-time seis-
mic network progressively evolved from 1 to 8 stations. Daily data
analysis protocols have also been continuously adapted in several
Institutes by our group to take advantage of the increasing number
of local stations and to produce better locations for the detected
events (see details in Section 2.2). In 2019, we also developed a
new protocol to efficiently process OBSs data during pickathons,
when, at the same place, several analysts dedicate a few days to
work together on the newly recovered data.
2.1 Development of the monitoring network
2.1.1 The in-land network
At the beginning of the seismic crisis, Mayotte’s seismicity was
monitored by the BRGM with the only local real-time seismic
station from the French R
´
esif-RAP accelerometric RA network
(YTMZ; R
´
esif 1995) and some regional stations from international
networks (IRIS/IDA, GEOFON and GEOSCOPE in Madagascar,
Kenya, Seychelles and La R
´
eunion). In the Comoros archipelago,
the Observatoire Volcanologique du Karthala (OVK) monitors the
Karthala volcano (Grande Comore) since 1988. The OVK seismic
network, while located 250 km northwest from Mayotte (Fig. 1b),
was crucial for characterizing Mayotte’s seismicity, particularly at
Downloaded from https://academic.oup.com/gji/article/228/2/1281/6374867 by Ifremer, Bibliothèque La Pérouse user on 09 November 2021
Mayotte seismic crisis: first results from marine and land seismic deployments 1283
Figure 1. Maps of seismological land stations and Ocean Bottom Seismometers (OBSs) used in this study with the red volcano symbol representing the
location of the new volcanic edifice (Feuillet et al. 2021). (a) Map showing the regional stations contributing to the monitoring (yellow inverted triangles).
Red rectangles highlights the Comoros archipelago area shown in (b) and the Mayotte area shown in (c). (b) Map of the Comoros archipelago showing the
temporary stations used for previously-developed velocity profiles (orange squares), historical M > 5,0 earthquakes close to Mayotte (white stars) and the
Karthala seismic network (yellow inverted triangles). (c) Map of the local land stations (green inverted triangles, most of them installed in 2019) and all OBSs
deployed between February 2019 and May 2020 [blue inverted triangles, the blue hue depends on the MAYOBS deployment timing as seen in (d)]. The MAYO
station (orange square) is the temporary station used by Dofal et al.(2018) receiver function study. The white dots represent the 5000 manually relocated
earthquakes. (d) Time evolution of the number of recording stations since the onset of the crisis. All regional stations (dark grey) are available in real-time.
Local stations (added to the regional stations; in grey)—significantly increased in number in 2019. The number of OBSs (added to the local and regional
stations; in light grey) varies over time depending on the MAYOBS deployment period (blue scale on bottom right numerated as the MAYOBS oceanographic
campaigns that recovered them). The dashed line represents the evolution in time of the number of real-time stations.
the beginning of the sequence. Real-time seismic data recorded at
Karthala volcano were shared during the first weeks of the crisis
via its French partner in the Indian Ocean, the Observatoire Vol-
canologique du Piton de la Fournaise (OVPF-IPGP). In June 2018, a
medium-band station was installed at a Mayotte school (ED.MCHI),
cofunded by the local representation of the ministries of environ-
ment (DEAL) and of education as well as the BRGM, within the
program Edusismo (Virieux 2000). By the end of June 2018, the
seismicity decreased after one and a half month of intense activ-
ity. At that time, the BCSF-R
´
eNaSS (Bureau Central Sismologique
Franc¸ais—R
´
eseau National de Surveillance Sismique), in charge of
the macroseismic and intensity investigations on the national ter-
ritory, performed a macroseismic field survey on Mayotte (Sira et
al. 2018) and took advantage of this mission to install two Rasp-
berryShake instruments (B
`
es de Berc et al. 2019) from the AM
network (Raspberry Shake Community et al. 2016): RAE55 and
RCBF0. RCBF0 station unfortunately failed after 2 weeks. The lo-
cal onshore network (Fig. 1) was not upgraded again until March
2019 with the installation of three stations: two Guralp CMG40T
medium-band sensors from the 1T temporary network (Feuillet, Van
der Woerd and RESIF, 2022)—MTSB and PMZI; and one Nano-
metrics Trilium120PA broad-band sensor from the QM network—
KNKL. Safety and rapid deployment on a small island, as well as
network geometry, were key criteria to choose the station sites. The
stations were therefore installed in city halls or public buildings to
ensure a stable power supply and protection from theft. In the same
time, a seismic station with triggered data transmission from the QM
network (GGLO) was installed on Grande Glorieuse, a small island
250 km northeast from Mayotte, completing the regional network in
March 2019. Finally, two RaspberryShake instruments were added
to the AM network by R
´
eNaSS in June 2019 on Mayotte: R1EE2
and R0CC5 (colocated with RA.YTMZ station).
2.1.2 The OBS network
At the end of February 2019, six short-period OBSs from the
INSU-IPGP OBS facility were deployed from a barge around the
Downloaded from https://academic.oup.com/gji/article/228/2/1281/6374867 by Ifremer, Bibliothèque La Pérouse user on 09 November 2021
1284 J.-M. Saurel et al.
earthquake locations that were known at the time (Bertil et al.
2019). The INSU-IPGP OBSs are free-fall instruments with 1-yr
maximum autonomy, whose sensors are a 3-channel geophone and
a broad-band hydrophone. There have been several recoveries and
redeployments of OBSs since then, but there have always been
4–16 OBSs deployed (Fig. 1d), which have greatly improved the az-
imuthal coverage of the local seismic network, as will be discussed
later in the paper. The INSU-IPGP OBSs were regularly comple-
mented with IFREMER micrOBSs or LotOBSs, hosting a similar
geophone and a short-period hydrophone but with an autonomy,
respectively, limited to 45 or 120 d maximum. After the initial de-
ployment, area specific location protocol establishment and much
more precise location of the seismicity, the OBS network geometry
remained stable and a number of positions were re-occupied during
subsequent deployments (Figs 1c and d).
Data acquisition protocol differs depending on the instruments—
land or ocean-bottom based. Data from land stations are acquired
and transferred in real-time, except the regional QM.GGLO sta-
tion from which only triggered data can be transmitted via a low-
bandwidth satellite link. The onshore data are centralized at the
IPGP data centre and then made available through the same pro-
tocol on its public SeedLink server. The three stations installed on
Mayotte in March 2019 were not equipped with internet connec-
tion during the first months of recording but were brought online
in May 2019. Their data were manually collected from the internal
storage just before the departure of the RV Marion Dufresne for the
MAYOBS1 oceanographic campaign in May 2019 (Feuillet 2019).
The OBSs are regularly serviced for maintenance and data recov-
ery (every 3–4 months for INSU-IPGP, every month for micrOBSs).
As of May 2020, nine different OBSs deployments have been con-
ducted. When the OBSs are recovered, their data are downloaded and
time-corrected for internal clock drift and converted to miniSEED
format using L-Cheapo tools (Orcutt & Constable 1996). These data
are then integrated in the main waveform database at the IPGP data
centre, along with the local and regional land station data.
2.2 Data analysis
2.2.1 Daily monitoring
For daily monitoring of the crisis, only the real-time data from land
stations can be used. Several institutes successively participated in
the day-to-day seismic data processing, using SeisComP3 (Weber et
al. 2007) with slightly different setups. The BRGM office in May-
otte was involved first and maintained a seismic catalogue from the
beginning of the crisis in May 2018 (Bertil et al. 2018, 2019). The
events from Bertil et al.(2018, 2019) were located from manu-
ally picked waveforms using the LocSAT algorithm (Bratt & Bache
1988) and a slightly modified IASPEI91 velocity model (Kennett
& Engdahl 1991). During the summer of 2019, with the addition
of new local stations available in real-time, the R
´
eNaSS hosted by
the Ecole et Observatoire des Sciences de la Terre in Strasbourg
(EOST) improved the settings of its SeisComP3 automatic detec-
tion system by using STA/LTA automatic picking and grid-search
location. They took over the role of manually picking all automat-
ically detected events for the newly created R
´
eseau de surveillance
Volcanologique et Sismologique de Mayotte (REVOSIMA) work-
ing group. The R
´
eNaSS also used LocSAT but with the original
IASPEI91 model. In April 2020, the daily manual picking and lo-
cation duty was transferred to OVPF-IPGP in La R
´
eunion, which
uses the NonLinLoc (NLL) location software (Lomax et al. 2014)
and a new hybrid velocity model described in Section 3. The NLL
configuration and 1-D local velocity model used by OVPF-IPGP in
La R
´
eunion were developed during MAYOBS1 campaign in 2019
(Feuillet et al. 2019;Saurelet al. 2019; Feuillet
2019).
While a daily manual screening of continuous waveforms with
WebObs (Beauducel et al. 2020) and identification of every event
has been performed since early 2019, the composite earthquake cat-
alogue is based on the automatic detection of events from land sta-
tions. The land stations influence the magnitude detection threshold
for two reasons. First, the land-based seismic network has evolved
since May 2018, and second, the chosen sites that offered equipment
safety are often affected by anthropogenic noise, implying a daily
variation of the magnitude detection threshold. Since the drastic
improvement of the local land network in spring 2019, the detec-
tion threshold is around M2.0 at nighttime and around M2.5 during
daytime.
2.2.2 OBS integration
After each OBSs data recovery, we manually pick phases on the OBS
data and offline land stations to improve the locations in the exist-
ing earthquake catalogue during dedicated pickathons that continue
today. Adding new phase picks significantly improved the location
of the earthquakes already detected and first located only by the
land network. Since there are already a great number of events to
relocate in the existing catalogue, we do not search for new events
from the OBS data set. Searching for new events in the OBS data
set is a work in progress and will be reported in future studies.
During each pickathon, the time span of the OBSs data we need
to process is divided across three teams using the same software
setup and each team manually locates earthquakes in descending
order of magnitude (Saurel et al. 2019; Fig. 2). When the OBSs
are recovered from an oceanographic research vessel, such as RV
Marion Dufresne, large enough to board a team of around 10 ana-
lysts, we divided the work in three 4-hr-shifts (i.e. 2 or 3 analysts
by shift) we can manually pick earthquakes 24/7. Otherwise, the
MAYOBS/REVOSIMA seismology group meet—either virtually
or at one of the group institutes—for 2 d to process the recently re-
covered data, produce graphics demonstrating the evolution of the
crisis, and interpret the results together. Whether the phase pick-
ing, location, and interpretation are done onboard or on land, we
use the same setup/configuration, starting database, and software
to analyse the recently recovered data. We assigned uncertainty to
each phase, from a common predefined list of values. Time un-
certainties assigned to the S-phase were always equal to or larger
than the uncertainty assigned to the P-phase. Impulsive P-phase
polarity onset is also reported for first motion source mechanisms
studies later use. We were able to locate events as small as M 0.8,
but despite improvement since July 2019 and the use of more sta-
tions in the STA/LTA automatic processing, their detection is far
from complete and mainly depends on the land station daily noise
level. We estimate that the magnitude of completeness is below M
3.0 (see Section 4) whatever the time allocated to the processing
during each pickathon and the number of events processed. We
typically process around 1000 earthquakes during pickathons when
performed onboard scientific cruises and around 500 earthquakes
during pickathons conducted on land. So far, our catalogue contains
more than 5000 manually picked earthquakes from February 2019
to May 2020, relocated using combined land and OBS data.
Downloaded from https://academic.oup.com/gji/article/228/2/1281/6374867 by Ifremer, Bibliothèque La Pérouse user on 09 November 2021
Mayotte seismic crisis: first results from marine and land seismic deployments 1285
Figure 2. Typical pickathon organization. After each OBS recovery, the deployment time-span is divided in three groups, that is Group 1, Group 2 and
Group 3. Each group processes the same number of events, by decreasing magnitude order, using the same software setup and the same picking guidelines
and recommendations. The daily catalogue of events, automatically detected and manually confirmed on land-based network data (BRGM then R
´
eNaSS and
OVPF-IPGP), is synchronized in the pickathon database. It also feeds in real-time the REVOSIMA/MAYOBS catalogue. When OBS recovery is performed by
a research vessel, each group alternates a 4-hr day and night shift. At the end of the pickathon, the relocated events, with additional manual picks and polarities
on OBSs data, is merged within the REVOSIMA/MAYOBS database for further use for monitoring and research. The same procedure is applied to on land
pickathons, only the work is completed during 2.5 normal work days instead of 24-hr shifts.
To enable this collaborative work, we use techniques and configu-
rations developed during the last 10 yr for the daily routine process-
ing of earthquakes in IPGP volcanic and seismologic observatories.
Waveform data and event databases are held by a SeisComP3 in-
stance (Weber et al. 2007). Each earthquake analyst uses their own
laptop and, regardless of their laptop’s Operating System, they run
a VirtualBox pre-configured Linux machine with the SeisComP3
Origin Locator GUI client (scolv, Weber et al. 2007). We use NLL
software and the new local 1-D velocity model described in the
following section to locate the events. Magnitudes are computed
with the embedded local magnitude (M
L
) formula in SeisComP3
(Richter 1958). The horizontal signals are converted to a Wood-
Anderson seismometer response (Urhammer & Collins 1990)be-
fore measuring their pick amplitude. We have not yet calibrated this
magnitude as very few of the earthquakes since 2019 have been
characterized with a moment tensor magnitude by the global mon-
itoring agencies (GCMT project, Dziewonski et al. 1981;Ekstr
¨
om
et al. 2012).
3. IMPROVED 1-D LOCAL VELOCITY
MODEL
A local velocity model is essential to provide precise location of
this dense swarm seismicity. One of the first challenges in improv-
ing earthquake locations was then to build a reasonable 1-D local
velocity model, because only global models were available so far.
This was done onboard RV Marion Dufresne during the MAYOBS1
campaign (Feuillet 2019).WeusedthedataofthefirstOBSsre-
covery and of three local land stations (see Section 2). We first
produced modified Wadati diagrams (Chatelain 1978) and consid-
ered two different existing velocity profiles from the area. The first
profile, named ‘Coffin449’, is based on a P-wave velocity (V
P
)
profile derived from a 1980 active-seismic sonobuoy. That experi-
ment was located 100 km southeast of Mayotte (Coffin et al. 1986,
instrument 449; Fig. 1b) and extended beyond 10 km depth with
a Moho interface 15 km deep (Jacques et al. 2019). The second
profile, named ‘ADofal’, is based on a S-wave velocity (V
S
)pro-
file determined from receiver functions (Dofal et al. 2018) using the
MAYO temporary station deployed on Mayotte island between 2011
and 2014 (RHUM-RUM project, doi:10.15778/RESIF.YV2011;
Fig. 1b). After adding phases from OBSs to 100 events during
an onboard pickathon, we located them with Hypo71 (Lee & Lahr
1972) and the ‘Coffin449’ model with a V
P
/V
S
ratio of 1.80 ex-
trapolated from Eastern and Central Afar studies (Jacques et al.
1999; Grandin et al. 2011). The modified Wadati diagrams (Chate-
lain 1978) indicated a local (OBS and Mayotte land stations)
V
P
/V
S
ratio of 1.66 and a regional V
P
/V
S
ratio of 1.72 (Figs 3a
and b).
We then tested different combinations of velocity model parame-
ters (2 velocity profiles and 3 V
P
/V
S
ratios) on a more complete data
set of 800 events with OBS phases, using NLL software (Lomax et
al. 2014). Contrary to Hypo71, NLL allows the use of depth vari-
ations of the V
P
/V
S
ratios and different velocity models depending
on the station. Its probabilistic approach (Lomax et al. 2014) also
makes the reported ellipsoidal errors more meaningful and easier
to interpret than estimated horizontal and vertical errors given by
Hypo71. We first used only local stations (Mayotte land stations
and OBSs) arrivals to assess the best local 1-D velocity model.
We compared the distributions of maximum ellipsoidal error for
the 2 velocity models and 3 V
P
/V
S
ratios for the 800-earthquakes
MAYOBS1 data set. The ellipsoidal error major axis ranges be-
tween 2 and 10 km, with most of the events at 4 ± 2 km. The
Downloaded from https://academic.oup.com/gji/article/228/2/1281/6374867 by Ifremer, Bibliothèque La Pérouse user on 09 November 2021