EDP Sciences logo
Submit your paper
Search
Menu
Home All issues Volume 197 (2026) BSGF - Earth Sci. Bull., 197 (2026) 6 Full HTML
Open Access
Issue
BSGF - Earth Sci. Bull.
Volume 197, 2026
Article Number 6
Number of page(s) 23
DOI https://doi.org/10.1051/bsgf/2025029
Published online 22 February 2026

© T. Cavailhes et al., Published by EDP Sciences 2026

Licence Creative CommonsThis 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 use, distribution, and reproduction in any medium, provided the original work is properly cited.

1 Introduction

Volcanic rifts located in subduction back-arc positions exhibit intrinsically coupled magmatic and tectonic activity (e.g., Uyeda and Kanamori, 1979; Taylor, 2013) with complex interactions, particularly when their development occurs within previously thickened and structured continental crusts that have been partially mapped (Le Pichon and Angelier, 1979; Gautier et al., 1999; Jolivet et al., 2015). These rifts, interacting with arc volcanism, concentrate important economic and societal interests, as they host a high potential for geothermal energy (Jolie et al., 2021; Papachristou et al., 2014) and a diversity and large quantity of mineral resources (Naden et al., 2005; Voudouris et al., 2019). They also exhibit CO2 degassing zones that must be taken into account in the carbon cycle (e.g., Bini et al., 2020). These extensional to transtensional zones of active deformation are also responsible for the release of thermal and mechanical energy in an intraplate context, generating volcanic (e.g., Preine et al., 2022), seismic (e.g., Papadopoulos and Pavlides, 1992), gravitational (Camerlenghi et al., 2010) and tsunamogenic (e.g., Petricca and Babeyko, 2019) hazards whose spatial and temporal modes of occurrence remain to be documented.

Normal faults and associated structural heterogeneities (e.g., fractures, fracture corridors) within and around these tectonic-volcanic systems are transient zones of exchange between the mantle, crust, hydrosphere and atmosphere. They reflect a response to extension that sporadically drains the pressure, temperature and chemical gradients associated with volcanic and hydrothermal processes (Pirajno, 2012; Kilias et al., 2013; Loreto et al., 2019). The episodic redistribution of crustal fluids by earthquakes in faults and fractures has been studied for several decades, based on field analyses (e.g., Sibson, 1981; Bonini et al., 2016; Montanaro et al., 2022), vein textures (e.g., Sibson, 1994), geophysics (e.g., Geoffroy et al., 2022; Chiarabba et al., 2022), geochemistry (e.g., Valsami-Jones and Baier, 2001), and numerical modelling (e.g., Marguin et al., 2023). It has been quantified to be particularly effective in the case of extensional systems where normal faults and steeply dipping joints are optimally oriented to dilate during the interseismic period and increase the porosity/structural permeability of the crust in the first 5 kilometres (Muir Wood, 1994). During the seismogenic rupture, the extension is located in the normal fault zone and the cracks in its fault blocks close-up in response to the elastic rebound, inducing a fluid overpressure that pushes the water upwards where the pressures are lower (Muir Wood, 1994). At the same time, co-seismic shearing along the fault creates dilation structures (relays, dilation jogs) that pump in surrounding fluids (Frank, 1965; seismic pumping, Sibson, 1994, 2000). Post-seismic fluid discharge drains the overpressured parts of the crust by fault valve action through rupture of permeability barriers (e.g., hydrothermal self-sealing cap, Brittle-Ductile Transition − BDT), especially at the crustal scale (Fournier et al., 1999; Sibson, 2000; Terakawa et al., 2010). The complexity is that the BDT is generally defined in geophysical terms as the depth above which most earthquake hypocentres are located (e.g., Maggini and Caputo, 2021) whereas beneath volcanic systems with high geothermal gradients, the ’thermal’ BDT, allowing non-localized plastic flow (e.g., dislocation creep of quartz), can be close to the surface (at depths of 2–3 km), have an irregular geometry and be located along structural heterogeneities (Parisio et al., 2019). This zone may correspond to that occupied by the hydrothermal self-sealing cap within the temperature range of 370–450 °C (Fournier et al., 1999), therefore challenging the geophysical definition of the BDT. At all scales, the relationships between inherited and/or newly formed crustal tectonic architecture (Walsh et al., 2002), the long- and short-term rheology of fault systems (e.g., loading, slip, rupture, sealing by fault zone mineralization (Sibson, 1994; Sibson, 2001; Cavailhes et al., 2013)), the location of earthquake hypocentres (micro- and macro-seismicity; Gold and Soter, 1984; Geoffroy et al., 2022; Denrée et al., 2024), perennial hydrothermalism (infiltration, storage, discharge, e.g., Okamoto et al., 2022) and pulsating surface hydrothermalism (e.g., hydrothermal explosions and phreatic eruptions, e.g., Montanaro et al., 2022) remain to be documented in order to understand the interactions between the architecture of deformation types, their plurality, their rheologies and the compartmentalization of crustal fluid circulation (e.g., Ross et al., 2020). Thus, constraining the geometry and dynamics of geothermal systems in active deformation zones, particularly the distribution of fluid compositions, temperatures and pressures within them, enables us to discuss the complex 4D-relationships between the hydraulics of the upper crust and its rheology (e.g., seismicity and microseismicity in the vicinity of the BDT, e.g., Rowland et al., 2012).

The volcanic archipelago of Milos is the most important volcanic centre of the Aegean arc (Fytikas et al., 1986, Fig. 1 and 2), host the most important high-enthalpy geothermal field in Greece (Fytikas et al., 1976), and has resulted in the most mineralized island in Europe in terms of percentage of mineralized surface area (Liakopoulos et al., 2001; Grosche et al., 2023, Fig. 3). The magmatism at the origin of these exceptional features is located in a post-thickening stretched continental crust, whose tectonic structures are only partially mapped by the scientific community, as they are essentially located at sea (e.g., Fytikas, 1977; Nomikou et al., 2013; Kokkalas and Aydin, 2013, Fig. 1 and 2). The detailed physiographic study and mapping of the structural lineaments will make it possible to document and hierarchize the subsurface processes responsible for the emergent or submerged topography (horsts and grabens) of the Milos archipelago. Using novel structural sketches, this work (i) shows the relationships maintained between the partially emerged peri-island geomorphologies and the major tectonic structures of the Aegean Sea (e.g., Myrtoon Basin, Christiana Basin, Cretan Basin, Mid-Cycladic lineaments) through a land-sea continuum based on multidisciplinary dataset (e.g., tectonics, marine geology, sedimentology, geophysics). (ii) Secondly, the authors examine the spatial relationships between fault networks, their architectures, the distribution of earthquakes (macro- and micro-seismicity), the distribution of recent volcanic events, and the distribution of perennial and sporadic surface hydrothermal events (e.g., phreatic eruptions). The causal links between these mechanisms will be investigated, in particular the possible relationship between seismic rupture of hydrothermal self-sealing cap systems leading to destabilization of surface reservoirs by sudden intrusion of fluids from reservoirs under lithostatic pressure. The ‘phreatic eruptions’ identified on the island, which led to its partial destruction around 2,300 yr ago, are also discussed in this context (e.g., Traineau and Delabakis, 1989).

2 Methodology

This physiographic study brings together and synthesises the multi-scale data and observations previously collected by the scientific community on land and at sea. Figures 2, 4 and 5 are based on topographical data (SRTM, Shuttle Radar Topography Mission), on bathymetric data from GEBCO and EMODnet (General Bathymetric Chart of the Oceans; European marine Observation and Data Network; https://emodnet.ec.europa.eu), on scientific publications for the Myrtoon basin / Antimilos area (Nomikou et al., 2014; Belka et al., 2024) and for the Paleochori bay (Nomikou et al., 2013). Analysis of the geological and geophysical data will allow us to discuss the controlling factors in the physiography (e.g., changes in elevation and slope) of the study area, as well as the spatial relationships between faults, seismicity, phreatomagmatism and hydrothermalism, both present and recent (Quaternary). Mapping of the tectonic lineaments of the Aegean Sea (Fig. 2) and the Milos Archipelago (Figs. 4 and 5) facilitates the identification of geomorphological markers associated with recent deformation episodes. It also locates volcanic edifices with recent or current activity, generally characterized by their pseudo-circularity, their slopes and the presence of a crater with little erosion, especially in the case of craters of historical phreatic eruptions (Fytikas et al., 1976). The word ’lineament’ is a geomorphological term referring to a mappable surface structure, linear, simple or composite, the segments of which are linear or slightly curved, and marks a break with the surrounding geomorphological signature (i.e., pattern, O’leary et al., 1976). Structural or tectonic lineaments are generally the surface expression of subsurface structural heterogeneities, such as recent fracture corridors, active faults (seismogenic or creeping), with no clear link to the kinematics of the latter at the scale of a lineament. The architecture of the mapped tectonic lineaments can be an indicator of the kinematics of a fault network, as can the difference in elevation on either side of a structure (e.g., pull-apart basin). The faults also produce propagating folds on the surface, creating a locally anomalous and/or broken slope, which can be mapped using the same technique (Florinsky, 1996). Limitations of such structural mapping depend on the (i) clarity of the geomorphological expression, (ii) the quality of the topographic / bathymetric data, as well as (iii) artefacts (Sander, 2007).

3 Geologic setting

3.1 South Aegean tectonics and magmatism

The studied area is located in Milos (Cycladic islands, Greece) which is part of the South Aegean Volcanic Arc, which results from the northwards subduction of the African plate underneath the Aegean microplate in an overall convergent setting (Pe and Piper, 1972; Le Pichon and Angelier, 1981; Kassaras et al., 2020, Figs. 1 and 2). The subduction started ⁓145 My ago and, since the upper Oligocene (⁓30 Ma), the position of the magmatic arc has started its southward motion due to slab retreat (Jolivet and Brun, 2010; Ring et al., 2010). A coeval regional back-arc extension started ⁓30 Ma ago, as expressed by a continental crust thinning across the Aegean Sea (Ring et al., 2010; Jolivet et al., 2021). The low degree of continental stretching and its long duration (40 Ma) without any oceanic crust formation suggest a complex tectonic context compared to other back-arc basins (Agostini et al., 2010). This crustal thinning results in the development of deep offshore basins bounded by seismogenic faults and a Moho discontinuity located at relatively shallow depths (Anastasakis and Piper, 2005; Nomikou et al., 2014, Figs. 2A and 2B). Authors have previously related the aforementioned extension across the Aegean Sea to either (i) a simple back-arc extension along normal faults (e.g., Feuillet, 2013; Roche et al., 2019) or to (ii) major NE-SW strike-slip faults segmenting this extension (Kokkalas and Aydin, 2013; Sakellariou and Tsampouraki-Kraounaki, 2019; Gueydan et al., 2025). Indeed, major NE-SW strike-slip faults such as the Mid Cycladic Lineament (MCL, Fig. 2B), parallel to the modern convergence direction, would have compartmentalized the extension and may have localized magma injections into the extensional quadrants of crustal strike-slip faults (Mascle and Martin, 1990; Piper and Perissoratis, 2003; Kokkalas and Aydin, 2013). The current geomorphological basins coincide with the Plio-Quaternary basins, demonstrating geodynamic coherence over the last 5 Ma (Anastasakis and Piper, 2005; Anastasakis et al., 2006; Sakellariou and Tsampouraki-Kraounaki, 2019, Fig. 2B). Present-day GPS velocities through the Aegean domain indicate a SW-directed translation relative to Eurasia at about 30.5 mm yr−1 (± 2 mm yr−1), perpendicular to the Hellenic arc (e.g., Reilinger et al., 2010).

Thumbnail: Fig. 1 Refer to the following caption and surrounding text. Fig. 1

Toponymic map showing the main localities and tectonic structures in the Aegean region mentioned in this study. See Section 4 (“Physiographic analysis”) and Figure 2 for more details. The studied Milos Archipelago forms part of the Aegean volcanic arc.

Carte toponymique indiquant les principales localités et structures tectoniques de la région égéenne mentionnées dans cette étude. Voir la section 4 (« Analyse physiographique ») et la Figure 2 pour plus de détails. L’archipel de Milos étudié fait partie de l’arc volcanique égéen.
Thumbnail: Fig. 2 Refer to the following caption and surrounding text. Fig. 2

(A) Topo-bathymetric and relief colored map of the Aegean Sea derived from SRTM (3 arc-second ca 30 m grid) and EMODNet Digital Bathymetry (1/16 arc-minute ca 115 m grid). (B) Associated morphotectonic sketch based on a detailed human-made lineaments mapping. Most of the lineaments are structural in origin and have been voluntarily overinterpreted to enhance discussion. The density of lineaments mostly depends on topo-bathymetric resolution and intensity of erosion, in particular on land. Milos appears to be located at the junction between the N110 trending southern part of the Myrtoon basin (My), the N-S trending structures dissecting the Aegean Sea, the Mid Cycladic Lineament (MCL) and­­ the N110 trending extensional faults leading to both the Cretan (C) and Christiana (Chr) basins. Hierarchy between neotectonics faults offsets can be positively correlated to the bathymetric relief. Am, Amorgos basin; An, Andros basin; Ar, Argolic basin; As, Astypalaea basin; C, Cretan basin; Chr, Christiana basin; Co, Corinth rift; CSK, Christiana-Santorini-Kolumbo rift; GoG, Gulf of Gokova; H, Heraklion Basin; Ka, Kamilonissi basin; Ma, Maleas basin; Mi-I, Mirthes-Ikaria fault; Myk, Mykonos; My, Myrtoon Basin; NI, North Ikaria Basin; NK, North Karpathos basin; SaG, Saronic Gulf; Si, Sikinos Basin; SK, South Karpathos basin.

(A) Carte topo-bathymétrique en couleurs de la mer Égée dérivée des données SRTM (grille de 3 secondes d’arc, soit environ 30 m) et de la bathymétrie numérique EMODNet (grille de 1/16 minute d’arc, soit environ 115 m). (B) Schéma structural associé basé sur une cartographie détaillée des linéaments. La plupart des linéaments sont d’origine tectonique et ont été volontairement surinterprétés afin d’enrichir la discussion. La densité des linéaments dépend principalement de la résolution topo-bathymétrique et de l’intensité de l’érosion, en particulier à terre. Milos semble être située à la jonction entre la partie sud du bassin de Myrtoon (My) orientée N110, les structures orientées N-S qui dissèquent la mer Égée, le Mid Cycladic Lineament (MCL) et les failles d’extension orientées N110 responsables des bassins crétois (C) et de Christiana (Chr). La hiérarchie entre les rejets des failles néotectoniques peut être positivement corrélée au relief bathymétrique. Am, Amorgos basin; An, Andros basin; Ar, Argolic basin; As, Astypalaea basin; C, Cretan basin; Chr, Christiana basin; Co, Corinth rift; CSK, Christiana-Santorini-Kolumbo rift; GoG, Gulf of Gokova; H, Heraklion Basin; Ka, Kamilonissi basin; Ma, Maleas basin; Mi-I, Mirthes-Ikaria fault; Myk, Mykonos; My, Myrtoon Basin; NI, North Ikaria Basin; NK, North Karpathos basin; SaG, Saronic Gulf; Si, Sikinos Basin; SK, South Karpathos basin.
Thumbnail: Fig. 3 Refer to the following caption and surrounding text. Fig. 3

Toponymy and synthetic geological map of Milos Island, based on the bibliography cited in the text and this study. The main tectonic structures (e.g., the Zephyria and Fyriplaka grabens), Quaternary volcanic centres (e.g., Trachylas, Fyriplaka), and mines (e.g., Cape Vani, Thiorychia) are also indicated.

Toponymie et carte géologique synthétique de l’île de Milos réalisée à partir de la bibliographie citée dans le texte et cette étude. Les principales structures tectoniques (e.g., le graben de Zephyria, le graben de Fyriplaka), les centres volcaniques du Quaternaire (e.g., Trachylas, Fyriplaka) et les mines (e.g., Cap Vani, Thiorychia) sont également indiqués.

3.2 Synthesis on the Milos geology

The geology of Milos has been studied throughout the last century (e.g., Sonder, 1924; Fytikas et al., 1976, 1986; Zhou et al., 2021). The island is made of four main geological units: (i) the alpine metamorphic basement, (ii) the Neogene sedimentary rocks, (iii) the volcanic sequences and (iv) the alluvial cover (Fytikas, 1977, Fig. 3).

The metamorphic basement of Milos belongs to the Aegean continental microplate where the Cycladic Blueschist Unit appears divided by the Trans-cycladic thrust in two nappes which developed during Paleocene/Eocene-aged subduction processes (Marsellos et al., 2012; Grasemann et al., 2018). These alpine metasedimentary rocks including micaschists, marbles, chloritic schists and quartzites, presently exposed in the south of Milos, were derived from Mesozoic protoliths and were metamorphosed to blue schists and green schists facies, respectively at 64 Ma ± 6 Ma and 35.2 ± 1 Ma (Fytikas, 1977; Grasemann et al., 2018). Isoclinal folds trend E-W to NW-SE with axial planes sometimes reworked as NW-SE trending shear-zone (Grasemann et al., 2018).

Upper Miocene to early Pliocene volcaniclastic sediments have been non-conformably deposited into WNW-SSE-trending extensional post-orogenic basins overlying the metamorphic basement of the Aegean micro-plate (Mercier, 1981; Van Hinsbergen et al., 2004). Prior to the onset of the Upper Pliocene volcanism, a remarkable stage of subsidence (0.9–1 km) occurred in the area during the early Pliocene development of these extensional basins (5.0–4.4 Ma; Van Hinsbergen et al., 2004).

Volcanism in the Milos archipelago started subsequently, about 3.5 Ma ago, with three main phases of explosive and extrusive activity, building in concert with tectonic uplifts of the present-day islands, with a probable subaerial emergence at 1.44 ± 0.08 Ma (Fytikas et al., 1986; Stewart, 2003; Stewart and McPhie, 2003). The Milos volcanic succession is composed of calc-alkaline volcanics, from basaltic andesite to rhyolite with a predominance of andesites and dacites (Fytikas et al., 1986; Stewart, 2003). Recent volcanic edifices such as Trachilas( −317 ka, Zhou et al., 2021) and Fyriplaka (− 37 ka, Kutterolf et al., 2021) are located at the tip of the NW-SE graben of the Milos gulf (0.38-0.1 Ma, phase 3 in Kokkalas and Aydin, 2013). Volcanism has continued up to the 1st century A.D. through hydrothermal and phreatic eruptions (Fytikas et al., 1986; −27 ka, Green Lahar, Principe et al., 2002; Traineau and Dalabakis, 1989) resulting in lahars deposits and / or tsunamites, particularly in Aghia Kyriaki (Photos-Jones et al., 1999; Hall et al., 2003). Onshore active solfatares and fumaroles and offshore hydrothermalism are related to this volcanic activity (Fitzsimons et al., 1997; Martelat et al., 2020; Khimasia et al., 2021). Geothermal manifestations are clearly associated with the Plio-Quaternary structural heterogeneities which localize and favor the hydrothermal circulation (Angelier, 1977; Fytikas, 1989, Ochmann et al., 1989). Hydrothermal circulation and related transition phases (e.g., boiling) led to the formation of economic minerals such as alunite, montmorillonite, sulphur, galena, manganese, iron oxides and epithermal gold; kaolin, baryte, perlite and bentonite, which are presently mined (Plimer, 2000; Liakopoulos et al., 2001 and therein references; Kilias et al., 2001; Naden et al., 2003; Schaarschimidt et al., 2021). The geothermal gradients in the five exploration wells drilled in the eastern region of the island, to depths of 1000 to 1200 metres, exhibit thermal gradients from 36 °C to 55 °C per 100 metres (Fytikas et al., 1989). Four wells were situated within the Zephyria graben, while the fifth was positioned along the border fault of the major N-S graben, to the north of Adamantas. The production zones are clearly identified within the fractured and brecciated metamorphic basement of the island. These wells show that conditions for boiling water at salinities between 20% and that of fresh water are common throughout the Milos hydrothermal system and explain the favourable conditions for the development of phreatic explosions by decompression of hydrothermal fluids (Fournier, 1989; Chiodini et al., 2023).

3.3 Synthesis on the Milos tectonics

Milos is situated within a complex system of crossing neotectonic grabens that can be interpreted as part of 10 km-scale submerged volcanic rifts (Pe-Piper and Piper, 2005; Nomikou et al., 2013; Preine et al., 2022, Figs. 1, 2B, and 4). Plio-Quaternary deformations are therefore complex, multi-directional, partly synchronous, cross-cutting, seismogenic, and interact with volcanic processes (Fytikas, 1989; Papanikolaou et al., 1993; Kokkalas and Aydin, 2013). They can be summarized as follows:

  • During the Pliocene, NE-SW extension has been described into the volcaniclastic series despite the fact that the related fault stress analysis has shown relatively dispersed main stress axes (Fig. 86 in work by Jarigue, 1978).

  • Late Pliocene NW-SE extension was expressed along ENE striking faults systems showing 20–80 m of vertical displacement (Angelier, 1979; Mercier, 1981; Phase I in Kokkalas and Aydin, 2013). This phase was responsible for the exposure of the metamorphic basement in the south of Milos and likely controls the settlement of dacitic domes and lavas (Kokkalas and Aydin, 2013, see Fig. 3 for the geological map).

  • During the lower Pleistocene (noted “Calabrian” (between 1.8 Ma and 0.8 Ma) in Mercier, 1981), unconformable sedimentation has been described and tentatively related to the lower quaternary poly-directional compressional event recognized in the back-arc domain (Angelier, 1977; Jarigue, 1978; Fig. 2 in Mercier, 1981). This compressive event would be due to an acceleration of the subduction which would have temporally stopped the rhyolitic production (Angelier, 1977; Jarigue, 1978).

  • During the Middle to the Upper Pleistocene, a N-S directed faulting/fracturing event is consistent with an E-W direction of extension in both onshore and offshore domains of Milos (Mercier, 1981; Anastasakis and Piper, 2005; between 1, 1 Ma and 0.38 Ma, phase 2 in Kokkalas and Aydin, 2013). Simultaneously, the Vromolimni-Kondaro fault bounding the western side of the Milos gulf shows strong tectonic activity, triggering Upper Pleistocene hydrothermal circulations and manganese mineralization (Fig. 3, Liakopoulos et al., 2001).

  • Based on field analysis, a Quaternary reactivation of the existing NW-SE normal faults has been detailed in the Figure 4 of Angelier (1979). This NW-SE-trending network of normal faults displaces the Quaternary deposits and is likely responsible for the current-day physiography of Milos, in particular the NW-SE elongation of its gulf, along the seismogenic Achivadolimni North-eastward-dipping fault (Ganas et al., 2022).

  • This NE-SW direction of extension is also consistent with the one derived from the focal mechanisms of 1987’s earthquakes at shallow depths, which were aligned along the NW-SE graben (1–10 km; Ochmann et al., 1989). This temporary network (1988–1989) of geophysical observations identified a seismic swarm 10 km beneath the island of Milos, with a hypocentre of 3 to 5 events/day interrupted by peaks of 600 events/day of low magnitude.

  • Delibasis and Drakopoulos (1993) and Papadopoulos (1993) have shown that the 5.3 Mw earthquake and its aftershocks in 1992 were tectonic in origin, related to E-W extension (neotectonic observations); this earthquake showed dextral transtensive kinematics, and the main shock was located 9.6 km below the same NW-SE tectonic graben that shapes the gulf of Milos. Aftershocks have been detected at a depth of around 5 km. This 1992 earthquake, associated with a fault dipping more than 75°, caused (i) the appearance of NW-SE oriented fissures/fractures in which gas fumaroles formed, (ii) liquefaction along these tectonic fissures, sometimes at a considerable distance from the epicentre (12 km), (iii) significant subsurface hydrological changes (artesian wells, turbidity, changes in water table levels), (iv) temperature increases in existing fumaroles, (v) subsidence of the coastal zone (−50 cm) and (vi) rock falls and landslides (Papanikolaou et al., 1993). Abnormal animal behaviour seems to have been reported 12 h before the main shock (Papanikolaou et al., 1993), these testimonies would corroborate the important role of gas emissions, detected by certain animals, as likely seismic precursors (Wang et al., 2006). In 1918, an earthquake with an estimated Mw of between 2.8 and 4.8 also levelled houses and opened new fumaroles (Drakopoulos and Delibazis, 1975; Fytikas et al., 1976). On 8 July 1738, a violent earthquake destroyed the village of Zephyria (Ambraseys, 2009), located in the north-south graben of the same name (Figs. 3 and 5), although the fault that ruptured has not been identified.

  • Stress tensors analysis all over the island has highlighted that the mean direction of extension into the Quaternary deposits seems coevally NW-SE to E-W directed (Fig. 89 in Jarigue, 1978; Kokkalas and Aydin, 2013). This extension is nearly perpendicular to the Quaternary Zephyria graben (Fytikas et al., 1986; 19 Ka in Principe et al., 2002) and consistent with the focal mechanism analysis performed in Delibasis and Drakopoulos (1993). In addition, and surprisingly, the GPS-derived horizontal velocity across Milos is perfectly N-S directed with a velocity of 2.66 mm yr−1 (Bohnhoff et al., 2006). By analysing INSAR data, Ganas et al. (2022) highlight a divergent differential movement of the eastern and western blocks of 1 cm/yr along the active NW-SE Achivadolimni fault, which borders the western margin of the Gulf of Milos (Fig. 3).

4 Physiographic analysis

4.1 Aegean Sea

Figure 2A shows the physiography of the Aegean Sea, where topo-bathymetric ruptures caused by faults bound the Plio-Quaternary offshore basins, linking the present physiography to the geodynamics of the last 5 million years (Sakellariou and Tsampouraki-Kraounaki, 2019). The mapped slope breaks, their linearity and a comparison with published knowledge of the Aegean Sea lead us to propose the tectonic sketch shown in Figure 2B. The volcanic archipelago of Milos appears to lie at the intersection of four major tectonic lineaments:

4.1.1 NE-SW trending lineaments

NE-SW tectonic lineaments are clearly expressed in the topo-bathymetry of the Aegean Sea, in particular by the fault systems of the South Karpathos Basin (SK), the Christiana-Santorini-Kolumbo Rift (CSK), the Gulf of Gokova (GoG), the Mirthes-Ikaria Fault (Mir-I) and the Mid-Cycladic Lineament (MCL, Fig. 1 and 2B). All these structures are subparallel to the Aegean crustal shear-zones described in Mascle and Martin (1990). The SW tip of the MCL is in the Milos sector, although some authors have mapped its presence to the north of the Milos archipelago (e.g., Brun et al., 2016; Sakellariou and Tsampouraki-Kraounaki, 2019), others to the south (e.g., Kokkalas and Aydin, 2013) and others have stopped mapping it to the NE of the archipelago (e.g., Le Pichon et al., 2019; Kassaras et al., 2020). Its behaviour is described as aseismic, demonstrating its current secondary role in active deformation; the MCL is thought to have ceased to function in the early Pliocene, in conjunction with the activity of the North Anatolian Fault (Walcott and White, 1998).

4.1.2 NW-SE trending lineaments

NW-SE oriented tectonic lineaments control the rugged physiography of the Corinth Rift (Co), the Saronic Gulf (SaG), the Argolic Basin (Ar), the Maleas Basin (Ma), the Cretan Basin (C), the Christiana Basin (Chr) and the southern part of the Myrtoon Basin (My) near the island of Antimilos, northwest of Milos (Fig. 2B). These major Aegean structures, which are essentially extensional, follow the cartographic traces of the North Cycladic Detachment and the West Cycladic Detachment, inherited from the Alpine compression phases (Searle and Lamont, 2022); they are also parallel to the direction of the West Anatolian Shear Zone (Papanikolaou and Royden, 2007). Linear slope breaks oriented ∼ N100E and large submarine elevation differences in the Milos area are located on the southern margin of the Myrtoon Basin and along the Cretan Basin, ∼ 50 km southeast of Milos (Sakellariou and Tsampouraki-Kraounaki, 2019).

4.1.3 E-W trending lineaments

An E-W tectonic fabric has also been observed north of the Crete and in the Amorgos grabens sector, particularly as evidenced by the E-W orientation of certain fault relays in the area (Fig. 2B). These tectonic lineaments may partially explain the slope break on the southern edge of Milos (Fig. 2B, Le Pichon et al., 2019). More generally, this E-W tectonic lineament direction is interpreted as being associated with the setting up of the Hellenic subduction (Mascle and Martin, 1990). This direction is also sub-parallel to the detachments associated with the formation of metamorphic domes, which express the extension of the Aegean back-arc in the lower crust during the middle Miocene (e.g., Serifos, located ∼50 km north of Milos, Ducoux et al., 2017).

4.1.4 N-S trending lineaments

The N-S trending tectonic lineaments of the Aegean Sea lie in the continuity of active N-S faults mapped in Crete (e.g., Gramvousa fault, Zacharias fault, Moslopoulou et al., 2014). These N-S fault systems control the tectonic grabens of the Hellenic Basin and are at the origin of the offshore basins south of Milos, at the eastern and western margins of the Myrtoon Basin, i.e., east of Serifos and Kythnos, and continue between mainland Greece and the island of Andros, towards the north of the Aegean Sea (Fig. 2B). In the Milos area, these lineament directions and the associated slope breaks appear to provide the junction between the N100E / NW-SE structures of the Myrtoon basin and the structures with the same direction bordering the Cretan Basin, and also explain the direction of the Zephyria graben (Fytikas, 1989). North of Milos, the N-S tectonic lineaments are parallel to the Eocene syn-orogenic trans-cycladic thrust (Grasemann et al., 2018; Glodny and Ring, 2022). At the scale of the eastern Mediterranean, the N-S tectonic lineaments are sub-parallel to the magnetic anomalies measured in the subducted Tethyan oceanic crust south of the Hellenic subduction (Granot, 2016).

The Quaternary activity of numerous offshore faults is also revealed in seismic profiles from the Folegandros Basin, the Milos Basin, between Ananes and Antimilos (west of Milos) to the south of Milos, and the Myrtoon Basin (Anastasakis and Piper, 2005; Anastasakis et al., 2006). The NW-SE to N100 faults that border the southern Myrtoon Basin show a clear morpho-tectonic expression and post-MIS12 Quaternary activity (e.g., Fig. 5 in Anastasakis and Piper, 2005). The N-S faults that control the eastern margin of the Myrtoon Basin, east of Sifnos, show significant Pliocene activity (fault and fault-propagation fold) and clear Quaternary reactivation (e.g., Fig. 4 dans Anastasakis and Piper, 2005). The offshore N-S to NE-SW fault systems to the south of Milos also show significant post-MIS-12 activity, accompanied by a clear morpho-tectonic expression, both indicative of their current activity (Fig. 8 in Anastasakis and Piper, 2005).

4.1.5 Volcanism

The volcanic centres of the Aegean Arc are located ∼200 km from the Hellenic Trough and are separated by ∼80 to 100 km (Fig. 2B; e.g., Sakellariou and Tsampouraki-Kraounaki, 2019). The spatial distribution of present or recent volcanic edifices within the volcanic centres of Nisyros and Santorini shows an elongation sub-parallel to the mapped NE-SW tectonic lineaments, characteristic of volcanic rifts (e.g., Preine et al., 2022). On a regional scale, the volcanoes of the Milos Archipelago, Poros, Methana, Aegina and Sousaki align on a NE-SW direction, sub-parallel to regional tectonic lineaments (Fig. 2B).

4.2 Milos archipelago

Figure 4A shows a topo-bathymetric map of the Milos archipelago, while Figure 4B shows a slope map. The combined physiographic analysis of these two figures shows that emerged and submerged landforms of the archipelago can be categorized as follows:

Structural high areas: Areas of relatively high altitude (> 100 m) with angular and jagged contours, covering several km2, such as the western areas of the island of Milos (Profitis Elias, Fig. 3) and the Tripiti area (> 200 m), the high N-S band running from Pollonia to Paliochori at the eastern end of Milos, most of the island of Kimolos and the island of Polyegos, generally correspond to tectonic blocks uplifted (relative horsts) by parallel or secant normal faults (Figs. 4A and 5). These normal faults create escarpments or cliffs, which can be identified by the linear slope breaks in Figure 4B (e.g., Thiorychia, old sulfur mines, Figs. 3 and 5). These structural highs are generally eroded by talwegs and canyons exhibiting regressive erosion morphologies in the emerged areas, which are located along second-order linear tectonic structures. Valleys and talwegs can either be perpendicular on a small scale (10–100 m), or parallel on a large scale (100 m–1 km) to slope failures, as commonly described for landforms associated with the dismantling of normal fault escarpments (Cotton, 1950). The structural highs are dissected by networks of fracture corridors and second-order faults, which are also underlain by erosion (e.g., Ksylokeratia area). Along escarpments or propagation-folds of active normal faults (e.g., Thiorychia, old sulphur mines), steep slopes of tectonic origin are readjusted by rockfalls and rotational landslides (e.g., Hesthammer and Fossen, 1999). Regressive erosion is particularly active along the tectonic lineaments to the east of Spathi (Fig. 3), where the Green Lahar (27 ka, Principe et al., 2002) is clearly affected by these gravitational destabilizations.

At sea, especially north of Antimilos and the Myrtoon Basin, areas of low relative bathymetry and large surface area (km2) also correspond to structural highs with angular contours (horsts or tilted blocks), whose termination can be explained by intersecting faults or fault segment relays (Fig. 4A). The straight slope breaks can also be explained by normal fault escarpments or normal fault propagation folds creating a flexure (Florinsky, 1996; Bouroullec et al., 2004). Due to the low resolution of the bathymetric data and their submerged position, these zones do not show fractured corridors or faults of scales smaller than multi-decametre displacement, nor clear erosional figures with the exception of the major canyons previously identified by Nomikou et al. (2013) (e.g., curvilinear canyon developed at the southwest margin of Serifos Island, Fig. 4B).

Structural low areas: Elongated areas of relative depression expressed by their flatness and adjoined by sub-linear structures show clear slope breaks that have already been interpreted as more or less rectilinear boundary faults (e.g., graben de Zephyria, Fytikas et al., 1989). The flatness of these depressions, particularly on land, is explained by the accumulation of sediments within these grabens, with no significant export zone towards the Gulf of Milos (i.e., airport zone and nearby salt marshes). However, the Quaternary filling of the Zephyria Graben remains limited, between 15 and 60 m according to borehole data analysis (Fytikas, 1989).

Offshore, the Myrtoon basin (−1080 m, Fig. 5) is an asymmetric WNW-ESE tectonic graben characterized by very steep slopes (> 30°) on its boundary faults. Its southern tectonic shoulder corresponds to a normal boundary fault with an offset of more than 1500 m (Nomikou et al., 2013). The shallow or relatively flat areas offshore correspond to the hanging-wall blocks of normal faults dipping towards the listric boundary fault (Fig. 5).

The volcanic edification of the archipelago is expressed by cones and craters recognizable by their circularity, generating crown-shaped reliefs that are sometimes open and occupied by quaternary pyroclastic flows and sequences (e.g., Fyriplaka volcano, Trachilas volcano, Fig. 4). Volcanic edifices (e.g., Antimilos (320 000 yr), Nomikou et al. (2013), Fig. 4B) can cause significant and steep reliefs (700 m of elevation with slope exceeding 30%) creating highly unstable cliffs (e.g., 1810, coastal landslide following a Hellenic arc earthquake, Ambraseys, 2009). The rugged terrain surrounding Antimilos Island is interpreted to be associated with submarine volcanic debris, flows, domes or dykes, analogous to outcropping volcanic features on land. The sub-conical shape of the submarine hills to the east of Antimilos suggests the existence of recent volcanic domes (Nomikou et al., 2013). Nearby fault systems appear to be disturbed and reoriented in the vicinity of the Antimilos volcanic edifices, shifting from a general NW-SE direction to N-S or E-W (Figs. 4 and 5).

On land, the differential erodibility of the exposed rocks (dome vs. volcanic-sedimentary cover) explains the differences in elevation that is not reflected in the linearity of the structural lineaments (e.g., dacitic dome of the trachylas).

A field of phreatic craters was mapped by Fytikas (1977) in the eastern part of the island (structural high) and is reported on Figure 5A (Langada area in Fig. 3). There are hundreds of craters, each measuring 15–20 metres in diameter and 3–4 metres deep, covering an area of 1–2 km2. Their clustering and rim overlap suggest successive local phreatic eruptions over a short period of time (Fontaine et al., 2003; Chiodini et al., 2023). This crater field extends in a north-west to south-east direction, sub-parallel to the Fyriplaka graben and the structures lining it.

4.3 Architecture of the Milos Gulf

At the scale of the island, morphotectonic analysis allows us to draw a detailed structural sketch in which the angularity of the Gulf of Milos appears to be structurally controlled:

  • NW-SE faults, fractures and fracture corridors have created and maintained the elongation of the Gulf of Milos through the activity of the Fyriplaka Graben (Fig. 5). The southern part of this graben emerged in response to the presence of the Fyriplaka quaternary volcano (tuff ring, Fig. 4). The northern part of the gulf is deepened by boundary faults that create unstable cliffs in the area of Cape Vani (i.e., the Vromolini-Kondaro fault) and south-west of Tripiti. These structures continue to the NW and connect with the network of normal faults in the offshore Myrtoon Basin. In the Antimilos area, these faults appear to intersect some of the submarine volcanic edifices identified in the work of Nomikou et al. (2013). These NW-SE normal faults also control the graben to the south-east of Kimolos and very probably the tectonic architecture of the Kimolos Basin (-743 m, Karageorgis et al., 2016; Fig. 5).

  • The N-S structures are particularly visible due to the geomorphological expression of the Zephyria Graben (N-S to N010), which creates an elongated flat area bounded by two subparallel N-S faults 1.5 km apart. The Zephyria Graben appears to be the lowest point of a 5 km wide N-S graben whose active eastern boundary fault lowers the Sarakiniko − Zephyria block along an axis passing through Mandrakia, Adamantas and Provatas Bay (Figs. 3 and 5). These faults control the elongation of the coastal coves, the linear morphology of the canyons and their Quaternary fillings in the Sarakiniko region (Fytikas et al., 1976; Leroy et al., 2023). These N-S faults also control the coastline to the east of Kimolos and the submarine slope rupture 4 km east of Milos, probably the expression of a normal fault with a collapsed western block.

  • NE-SW trending active faults are visible to the south and east of Milos, defining grabens that are well represented in the bathymetric surveys of Nomikou et al. (2024). This fracture set is well expressed geomorphologically on the islands of Polyegos and Kimolos, and on the western elevated part of the island of Milos where they also locate Upper Pleistocene systems of gold-bearing fractures (Kilias et al., 2001). This is also the most recent extensional fracture direction (i.e., Quaternary − present) reported by various authors following their field studies (Jarrige, 1978; Leroy et al., 2023).

  • The E-W trending structural lineaments are visible in the physiography of Milos on the northern edge of the island, where they control an identifiable fault break a few hundred metres upstream of the north-south trending canyon of Sarakiniko (Fig. 5). This structural lineament controls the E-W coastline near the Kodarou Bay by collapsing the northern block (on the Cap Vani side, Fig. 3). This E-W structure continues to the south of Antimilos in a network of normal faults that are poorly imaged but can be inferred from the seismic profiles published by Anastasakis et al. (2006). They extend westwards to connect with the N100 shoulder of the Myrtoon basin, near the Falkonera islet horst. On land, south-dipping normal faults cut the Lower Pliocene deposits east of Provatas Bay and the metamorphic basement at Spathi Cap, west of Paleochori (Fig. 3). These faults appear to be covered by the 27 000 yr old Green Lahar at this point (Principe et al., 2002).

4.4 Spatial relationships between structures and volcanism

Based on the detailed physiographic analysis in Figure 4, a neotectonic structural sketch is proposed in Figure 5A, with the superimposed spatial distributions of (i) volcanic centres and (ii) phreatic eruption craters with recent morphological expressions, (iii) recent hydrothermal alterations and (iv) current hydrothermal expressions (water, gas) listed in the scientific literature (Fytikas et al., 1977; Fytikas and Vugiukalakis, 1993; Dando et al., 1995, 2000; Mendrinos et al., 2010; Nomikou et al., 2013; Daskalopoulou et al., 2018; Megalovasalis, 2020; Burhing et al., 2023; Cavailhes, 2025a). The main Quaternary subaerial rhyolitic volcanic centres are Trachilas (north) and Fyriplaka (south), the latter closing off the bay of Milos and lying at the intersection of the Zephyria and Fyriplaka grabens (Figs. 4 and 5A). The Fyriplaka crater is a tuff ring measuring ∼1700 m in diameter and 220 m high, the north-western part of which is ripped open towards the Gulf of Milos (Fig. 5A).

Phreatic eruption craters are mainly located in the structural high (alt. ≃200 m, Langada area, Fig. 3) east of the N-S Zephyria graben, and are aligned on NW-SE oriented fracture zones, i.e., subparallel to the Fyriplaka graben (Fig. 5A). This sequence of phreatic eruption craters ends south of Thiorychia (old sulphur mines, Fig. 3), where their distribution suggests a lateral connection with the landslides observed on the cliffs of the area and the emerged and immerged hydrothermal seeps (Fig. 5A). Phreatic craters align along the same NW-SE direction in the Voudia area, notably aligned with the islet of Glaronissia made up of Lower Pleistocene lavas (Fytikas, 1977).

The Milos geothermal field includes active and extinct hot springs, hot soils, solfataras and submarine gas seeps, the latter only partially mapped (Fytikas et al., 1976; Fig. 5A). These geothermal or hydrothermal expressions are located only to the east of the western boundary fault of the Gulf of Milos (named the Achivadolimni fault in Kokkalas and Aydin, 2013), near phreatic eruption craters or even within them (e.g., into Fyriplaka ring; around the well MA-1 north to Adamantas). Disparate alignments can be interpreted along tectonic directions NW-SE, NE-SW (e.g., Kimolos, Tsokas et al., 1995) and to a lesser extent N-S (Fig. 5A). These surface manifestations are distributed preferentially in the metamorphic bedrock of the island, in the overlying cover, especially in areas where this cover is thin or missing (Fytikas and Marinelli, 1976; Fig. 3). With a few exceptions (e.g., the Fyriplaka antennae, ≃160 m, near Kamalos cap, alt.), most of the surface hydrothermal features are close to present sea level. Below the sea level, around fifty isolated or coalescing submarine depressions at depths between 125 and 220 meters in the outer part of the Gulf of Milos, 2 km north of Cape Vani and to the east of the N-S trending graben of the outer Gulf of Milos (Cavailhes, 2025a). Their diameters range from 10 to 200 m with negative relief of up to 70 m. These depressions are located directly above and/or within the hanging walls of the active normal faults, as indicated by their escarpments, and are aligned along the associated NW-SE and N-S structural directions. Burhing (2023) observed numerous areas of CO2 degassing from open fractures (100 °C), near the active faults that divert these depressions, as well as white/yellow chimneys structures made of massive cristalline sulphurs, carbonate crusts and bacterial mats. In the south of Milos, 3 km south of Aghia Kiriaki Bay, submarine depressions (10–100 m in diameter) that also degas CO2 are observed in a 1.8 km long, 1 km wide, 120 m deep, structurally controlled graben (Burhing, 2023). These depressions are preferentially distributed along the P and T structural planes (Nicolas, 1989). This structural architecture highlights a dextral, transtensional, strike-slip fault system that is expanding on the western edge of the basin. This fault architecture is consistent with the 1992 earthquake kinematics inferred from the focal mechanism (Delibasis and Drakopoulos, 1993).

The hydrothermalized zones mapped on the Fytikas (1977) geological map correspond either to the location of phreatic eruption craters, or to active hydrothermal vents (Fig. 5A). They may be elongated along major tectonic directions, either NE-SW in the Voudia area, or NW-SE in the Paleochori area (Fig. 5A).

Thumbnail: Fig. 4 Refer to the following caption and surrounding text. Fig. 4

(A) Topo-bathymetric and relief map of the Milos Archipelago derived from GEBCO data and completed with the bathymetric map of Nomikou et al. (2014). (B) Associated slope map derived from GEBCO data and including Antimilos, Kimolos and Polyegos. Both maps were used in addition to the cited literature to construct the synoptic structural sketch in Figure 5.

(A) Carte topo-bathymétrique de l’archipel de Milos dérivée des données GEBCO et complétée par la carte bathymétrique de Nomikou et al. (2014). (B) Carte des pentes associée dérivée des données GEBCO et incluant Antimilos, Kimolos et Polyegos. Les deux cartes ont été utilisées en complément de la littérature citée pour construire le schéma structural synoptique de la Figure 5.
Thumbnail: Fig. 5 Refer to the following caption and surrounding text. Fig. 5

 Structural sketch of the Milos archipelago derived from the physiographic analysis of Figure 4, showing the main structural lineaments (see the equal-area rose diagram for the strikes of the lineaments) and highlighting faults, fault propagation folds and fracture corridors. The Milos archipelago is located at the intersection of neotectonic volcanic grabens that express multidirectional extension. (A) The structural sketch is overlaid with the location of Quaternary volcanic centres, phreatic craters, hydrothermalized rocks (Fytikas, 1977) and hydrothermal vents (Fytikas and Vugiukalakis, 1993; Dando et al., 1995; Khimasia et al., 2021; Mendrino et al., 2010; Louis et al., 2003). (B) The structural sketch is crossplotted with the locations of the earthquake epicenters. References are on Figure 5B and comments are in the text. The inset on the left, taken from Ganas et al. (2022), shows the aligned epicentres along the NW-SE and N-S lineaments. The focal point mechanism shown in the figure is related to the earthquake that occurred the 20th of March 1992 (Delibasis, N. D., and Drakopoulos, J. C., 1993). The focal depth distribution of local earthquake is taken from Ochmann et al., (1989).

Schéma structural de l’archipel de Milos dérivée de l’analyse physiographique de la Figure 4, montrant les principaux linéaments structuraux (cf., projection stéréographique pour les orientations de linéaments) et mettant en évidence les failles, les plis de propagation des failles et les couloirs fracturés. L’archipel de Milos est situé à l’intersection de grabens volcaniques néotectoniques exprimant une extension multidirectionnelle. (A) Le schéma structural est superposé à l’emplacement des centres volcaniques quaternaires, des cratères phréatiques, des roches hydrothermales (Fytikas, 1977) et des évents hydrothermaux (Fytikas et Vugiukalakis, 1993; Dando et al., 1995; Khimasia et al., 2021; Mendrino et al., 2010; Louis et al., 2003). (B) Le schéma structural est superposé aux localisations des épicentres de séismes. Les références figurent sur la Figure 5B et les commentaires sont dans le texte. L’encart à gauche est issu de Ganas et al. (2022) et montrent les épicentres de séismes alignés sur les linéaments NW-SE et N-S. Le mécanisme du foyer représenté sur la figure est lié au séisme qui s’est produit le 20 mars 1992 (Delibasis and Drakopoulos 1993). La distribution des hypocentres de séismes est issue de Ochmann et al., (1989).

4.5 Spatial relationships between structures and earthquakes

The island of Milos shows variable seismicity in time and clustered in space, primarily attributed to tectonic earthquakes and swarms of microearthquakes (Valsami and Baier, 2001). Figure 5B shows the location of the 1992 tectonic earthquake (Mw 5.3) and the distribution of microseismicity (Mw < 2) recorded on the archipelago since 1986 (Ochmann et al., 1989). Based on published scaling laws, the lengths of ruptured fault segments are about 4.9 km for the 1992 earthquake (Mw 5.3) and multi-metric (5 m–100 m) for microseismicity (Mw −0.5 to Mw 2) (Kanamori and Anderson, 1975; Wells and Coppersmith, 1994).

The epicentres of the micro-earthquakes are mainly offshore, particularly at the southern end of the Gulf of Milos, south of the town of Adamantas (Figs. 3 and 5B). These earthquakes are located at the intersection of the N-S and NW-SE grabens. Within a single microseismogenic event, the alignment of the epicentres in two distinct directions suggests that the NW-SE and N-S faults rupture simultaneously (Ganas et al., 2022). The seismological and InSAR data from the 2021–2022 microseismicity crisis suggest that the Achivadolimni fault hosts microseismicity under the Gulf of Milos (5–6 km deep), dips 60° to the east and accommodates 1 cm/yr of differential movement between its blocks (Ganas et al., 2022). In this same area, during the 1986 seismic crisis, the alignment of epicentres along NW-SE tectonic lineaments was also highlighted (Ochmann et al., 1989, Fig. 5B). These swarms of micro-earthquakes are mainly located 5–7 km below the surface and in the lowered block of the N-S graben. Scattered earthquakes, also offshore, mark the Fyriplaka Graben towards NW, at the exit of the Gulf of Milos, at the transition to the Myrtoon Basin. (e.g., Ochmann et al., 1989; Ganas et al., 2022). In this zone, 6 km east of Antimilos, normal faults oriented E-W and N-S, with a clear submarine morphotectonic expression, have been mapped and could locate these earthquakes (Fig. 5B). Alignment of NW-SE trending earthquakes can also be identified southeast of Voudia and in the Thiorychia area (old sulphur mines), just below the mapped phreatic craters and extending in a NW-SE direction (Figs. 5A and 5B).

South of Milos, earthquake epicentres have been mapped in the submerged part of the Fyriplaka graben, SE of the active volcano, at the structural intersection with the boundary fault east of the N-S Zephyria graben and particularly off Paleochori (Fig. 5B). In this area, the 1992 dextral transtensional 9.6 km-deep earthquake was located on a NW-SE fault, probably the Achivadolimni fault, which controls the subsidence of the Gulf of Milos (Delibasis and Drakopoulos, 1993).

A swarm of microearthquakes is also located in the southern part of the Zephyria graben, at the contact with the Fyriplaka volcanic edifice (Fig. 5B). Instrumental microseismicity appears to be absent beneath the Quaternary volcanic edifice of Tachylas, but a group of microearthquakes (1986 and 1987) was located below the structural high near Tripiti, separating the Fyriplaka grabens and the N-S Zephyria graben (Fig. 5B).

5 Discussion

5.1 Milos archipelago and the regional tectonics

The present physiography of the Aegean Sea can be explained by Plio-Quaternary extensional deformation, expressed by zones of preferential subsidence, corresponding to grabens where sediments accumulate, alternating with normal faults with raised structural highs, submerged (shallow banks) or emerged (islands), corresponding to tectonic horsts (Sakellariou and Tsampouraki-Kraounaki, 2019, Fig. 2). The volcanic archipelago of Milos lies at the intersection of three directional sets of tectonic lineaments, currently segmenting the Aegean Sea into a tectonic jigsaw puzzle (Figs. 2 and 6A). The fault directions are inherited from earlier tectonic-metamorphic pre-structuring, mainly alpine (e.g., Eocene compressive or transpressive phases, Aegean crustal shear zones described in Mascle and Martin (1990)) or post-orogenic (e.g., middle Miocene extension; Ducoux et al., 2017) and reactivated in the area by Plio-Quaternary extensive stresses. Based on the tectonic map shown in Figure 2 and following the work of Agostini et al. (2009), Plio-Quaternary extensional deformations are not confined to a back-arc basin to the north of the volcanic arc (i.e., Methana, Milos, Santorini, Nysiros). Instead, these volcanic edifices have settled in a multi-directional, pre-structured Aegean extensional region. In detail, the NE-SW trending faults appear to be the consequences of extensional deformation developed along the Mid-Cycladic Lineament or the Myrthes-Ikaria Fault (Fig. 6). These faults are active in the Milos area, where they show a clear morpho-bathymetric expression (Figs. 2 and 6) and Quaternary deformations observed on land and at sea (Anastasakis and Piper, 2005; Leroy et al., 2023; Belka et al., 2024). The NW-SE faults are associated with the opening of the Myrtoon Basin and the Gulf of Milos, the latter being partially emerged in the south of the island due to the presence of the Fyriplaka volcano. These faults are active in the Milos area, where they have a clear morphotectonic expression both on land and at sea, and localize seismicity, particularly along the Achivadolimni fault (Figs. 3 and 5, Ganas et al., 2022). On the regional scale of the Aegean, the E-W oriented faults observed in the Milos archipelago are associated with the opening of the Plio-Quaternary basins of the Milos, Cretan and Christiana basins, probably along post-orogenic structures sub-parallel to the West Cycladic detachments (Ducoux et al., 2017). The N-S fault systems responsible for the Zephyria tectonic graben on Milos, the offshore basins east of Serifos and Kythnos (Myrtoon Basin) extend as far as Crete, ruling out an origin solely associated with the Eocene syn-orogenic trans-cycladic thrust (e.g., Grasemann et al., 2018, Fig. 6A). These orientations are sub-parallel to magnetic anomalies measured in the subducted Tethyan oceanic crust (Granot, 2016) and may be their partial expression in the upper plate. This last family of active faults is probably the least studied and least understood of the Aegean system. To facilitate the following discussion, active deformation, seismicity and hydraulic drainage by hydrothermal fluids (water, gas) are located along the three main fault directions described above, oriented NW-SE (Fyriplaka Graben), N-S (Zephyria Graben) and NE-SW (MCL direction) (Figs. 5 and 6).

Thumbnail: Fig. 6 Refer to the following caption and surrounding text. Fig. 6

Synoptic sketches showing (A) the current location of the volcanic center of Milos at the starry-shape intersection of 3 main neotectonic grabens of the Aegean Sea. (B) Hydrothermalism and phreatic eruptions can be found along all fault lines east of the Achivadolimni fault, particularly near the Fyriplaka volcanic centre. Tectonic earthquakes are mainly located along the N-S and NW-SE grabens.

Schémas synoptiques montrant (A) l’emplacement actuel du centre volcanique de Milos à l’intersection des trois principaux grabens néotectoniques de la mer Égée. (B) L’hydrothermalisme et les éruptions phréatiques peuvent être localisés le long de toutes les directions de failles à l’est de la faille de Achivadolimni, notamment à proximité du centre volcanique de Fyriplaka. Les séismes tectoniques sont principalement localisés le long des grabens N-S et NW-SE.

5.2 Seismo-tectonic architecture and earthquake hazards

The maximum expected magnitude of an earthquake is a function of the length of the fault zones that are likely to rupture (Wyss, 1979; Wells and Coppersmith, 1994). Precise mapping of active faults and their possible segmentation is thus essential for seismic hazard evaluation/quantification. In the Milos area, the spatial distribution of seismic swarms along the preferential directions of faults recognized at sea and on land materializes active faults (e.g., Liu et al., 2024). Historical and instrumental seismic activity is mainly associated with the NW-SE Fyriplaka graben and in particular the NE dipping Achivadolimni fault (Ochmann et al., 1989; Ganas et al., 2022), although both morphotectonic analysis and historical records suggest that other active fault directions may be seismogenic. (e.g., 1738, Zephyria earthquake, Ambraseys, 2009). The average lengths of fault segments mapped by morphotectonic analysis are often >1 km with normal to transtensional kinematics (Fig. 5B) and can exceed 10 km in length, especially the Achivadolimni fault zone (Fig. 6B). Published scaling laws (e.g., Wells and Coppersmith, 1994) show that a 10 km long rupture may produce a Mw 5.8 earthquake, while ∼1 km rupture lengths may produce earthquakes of Mw 5, consistent with the historical seismicity of the area (Papanikolaou et al., 1993). Note that the multidirectional and synchronous extensions in the study area generate a highly connected fault network, i.e., few fault tips are mapped, only connected, relay or intersecting zones (Fig. 5B, e.g., Nixon et al., 2020). This connectivity can lead to effective stress transfer from a given fault to its neighbour, a mechanism described for sequence seismicity on fault systems (Hauksson et al., 1993). In addition, the rupture of multiple faults in a single earthquake generally leads to an underestimation of the predicted magnitude based on the length of a mapped fault alone (Sieh et al., 1993). Given the tectonic architecture described in Figure 5 and published seismicity data showing microearthquakes on at least two fault directions (i.e., NW-SE and N-S) during the same seismic crisis (Ochmann et al., 1989), we propose that the earthquakes on Milos may be larger than the 1992 earthquake. In addition, the Peak Ground Acceleration (PGA) during an earthquake depends mainly on the magnitude and the distance between the hypocenter (a function of its depth) and the location of interest (e.g., Adamas, Fig. 2); the role of site effects seems statistically insignificant if the distance is less than 23 km (Masoumi et al., 2024). Historical seismicity on Milos shows that the hypocenters of crustal earthquakes are located at relatively shallow depths (5 km–10 km, Ochmann et al., 1989) and are therefore likely to generate high and destructive PGAs.

5.3 Recent thermal anomalies and magma chambers

Figures 5A and 5B show that, at the archipelago scale, the map coverage of microseismic epicenters is the same in area and location as the map coverage of hydrothermal events, suggesting that they may be linked. Microseismicity and hydrothermal manifestations are exclusively to the east of the Achivadolimni fault (i.e., in its hanging wall, where preferential fracturing is located, see Withjack et al., 1995), bounding the western border of the Gulf of Milos and the Fyriplaka Graben (Figs. 5A, 5B, and 7). In the context of volcanic arcs, hydrothermal manifestations are generally located above or in the immediate vicinity of magma chambers undergoing crystallization, particularly in the case of fractured/faulted systems where sub-vertical structures effectively drain the underlying liquid or gas pressure gradients and are likely to generate microseismogenic fault movements (e.g.; Weis, 2012; Doke et al., 2018; Nowacki et al., 2018; Okamoto et al., 2022; Fig. 7). This configuration would explain why various thermal expressions such as the Fyriplaka volcano, perennial hydrothermal springs, recent phreatic eruptions, microseismicity and recently hydrothermalized rocks are located directly above the magma chamber (Fig. 5B, Fig. 9 in Fytikas et al. (1989)). However, in detail, topography of volcanic or tectonic origin is likely to modify and shift hydrothermal fluid escape zones by modifying surface water entry points (e.g., seawater and meteoric water) and hydraulic load gradients in the hydrothermal system (Weiss, 2012). Self-sealing of faults and volcanic-sedimentary series can also generate temporal and spatial migrations of fluid circulation by cyclic reduction of porosity and permeability of rocks (Sibson, 2001; Spagnoli et al., 2024). The current mineralizing epithermal system is located in the hanging-wall of the Achivadolimni fault, while the western part of the island corresponds to a relict submarine epithermal system of upper Pliocene to lower Quaternary age (Liakopoulos et al., 2001), tectonically exhumed in the footwall of the same fault (Stewart, 2003; e.g., Au, Mn, Ba, Ag, Pb, Zn). Ongoing microseismic deformations in the hanging wall of the Achivadolimni Fault probably generate the opening increments required for the present stockwork-type epithermal mineralization a few hundred meters below the surface above the magma chamber (Fig. 5; Sibson, 1987, 2001; Liakopoulos et al., 2001).

Thumbnail: Fig. 7 Refer to the following caption and surrounding text. Fig. 7

Synoptic cross-section showing the horst and graben architecture of the island of Milos along an E-W transect. The lower part of the section is stylized from the work of Jolivet et al. (2021), showing the crustal-scale relationship between plutons and normal faults in a stretching continental context. The current hydrothermal system is located exclusively in the hanging-wall (to the east) of the Achivadolimni fault. The 370 °C isotherm probably limits the self-sealed hydrothermal cap that could correspond to a local “thermal BDT” (Fournier, 1999) in which matrix and structural permeabilities are very low. This zone represents the lower limit of convective circulation in the faults and fractures of the first 2–3 kilometres under hydrostatic pore pressure (R1, shallow reservoir, R2 deep reservoir, Naden et al., 2005). Lenses of fluids (gases, magmatic waters, meteoric waters and brines) under lithostatic pore pressure are trapped beneath the very low permeability hydrothermal self-sealing cap. During a significant seismic rupture (> Mw 5.5?), the structural drains open up, allowing significant hydraulic communication between the reservoir under lithostatic pressure and the surface reservoirs. The result is sudden destabilization due to temperature rise and decompression of surface reservoirs, which can lead to phreatic eruptions such as those that created the historic Thiorychia craters (Langada area, Fig. 3). On the right of the figure is the average geothermal gradient, 250-300 °C/km, recorded in the 5 boreholes of the Zephyria Graben. The toponymy is projected onto the section. The distinctions between the different lithologies (e.g., basement, metamorphic basement, Miocene cover, Miocene-Pliocene volcanism, domes and flows, Fytikas et al. (1989)) have not been included to make the figure easier to read. In the geophysical sense, the brittle-ductile transition (BDT) is defined as the depth above which most earthquake hypocentres are located (Maggini and Caputo, 2021). It has also been reported on the figure.

Schéma synoptique montrant l’architecture en horst et graben de l’île de Milos suivant un transect E-W. La partie inférieure de la coupe est stylisée à partir des travaux de Jolivet et al. (2021), montrant la relation à l’échelle de la croute entre les plutons et les failles normales en contexte d’étirement. Le système hydrothermal actuel se situe exclusivement dans le bloc de toit (à l’est) de la faille de Achivadolimni. L’isotherme 370 °C limite probablement le « hydrothermal self-sealed cap » pouvant correspondre localement à la « brittle-ductile transition zone » au sens thermique du terme (Fournier, 1999) dans laquelle les perméabilités matricielle et structurale sont très faibles. Cette zone limite par le bas une circulation convective dans les failles et fractures des 2-3 premiers kilomètres sous pression de pore hydrostatique. Sous le « hydrothermal self-sealed cap » de très faible perméabilité, sont piégés des lentilles de fluides (gaz, eaux magmatiques, météoriques et saumures magmatiques) sous pression de pore lithostatique. La dilation de ces fluides par décompression et refroidissement, notamment à l’aplomb du golfe de Milos, génère probablement la micro-sismicité dans les core-zones et damage zones de la faille artérielle de Achivadolimni. Au cours d’une rupture sismique significative (> Mw 5.5 ?), les drains structuraux s’ouvrent, autorisent une communication hydraulique significative entre le réservoir sous pression lithostatique et les réservoirs superficiels. Il en résulte une déstabilisation soudaine par élévation de température et décompression des réservoirs superficiels pouvant engendrer des éruptions phréatiques, telles que celles à l’origine des cratères historiques de Thiorychia (Langada area, Fig. 3). A droite de la figure, le gradient géothermique moyen, 300 °C / km, compilé dans les 5 puits du graben de Zephyria, est figuré. La toponymie est projetée sur la coupe. Les distinctions entre les différentes lithologies (e.g., socle, socle métamorphique, couverture Miocène, volcanisme mio-pliocène, dômes et coulées, Fytikas et al. (1989)) n’ont pas été reportées pour faciliter la lecture ­­de la figure. La Brittle-Ductile Transition (BDT) au sens géophysique est définie comme la profondeur au-dessus de laquelle l’essentiel des hypocentres de séismes est localisé (Maggini and Caputo, 2021). Elle est également indiquée sur la figure.

5.3.1 Macroseismicity and arterial fault

Macroseismicity at Milos has been interpreted as tectonic in origin and associated with regional normal faults (e.g., Achivadolimni fault; Papadopoulos, 1993, Mw 5.3) while microseismicity is likely associated with hydrothermal fluid circulation along active fault systems (Valsami-Jones and Baier, 2001; Ganas et al., 2022). Long-term Mio-Plio-Quaternary tectonic loading in the Aegean Sea generates extensional stresses that activate ductile listric faults with detachment geometry, locally associated with synkinematic plutons at the origin of surface arc volcanism (Jolivet et al., 2021, lower part of the Fig. 7). In the upper structural level or brittle domain (Mattauer, 1973), this extension is expressed by systems of horsts and grabens with Andersonian fault dips (i.e., 60–70° Anderson, 1905; Ganas et al., 2022), where normal faults and subvertical fractures dominate and where hydrothermal circulations are generally under near-hydrostatic pressure (epithermal system, Fytikas and Marinelli, 1976; Reynold and Lister, 1987; Fytikas, 1989; Fournier, 1999, Fig. 7). High-enthalpy hydrothermal alteration significantly modifies the initial chemical, mineralogical and textural compositions of rocks, resulting in changes in their petrophysical (i.e., porosity, permeability, density, elastic wave velocities, thermal conductivity) and mechanical (e.g., cohesion, strength, friction) properties (e.g., Mayer et al., 2016; Pirajno, 2012). In systems with strong geothermal gradients (greater than 50 °C/km), quartz solubility and precipitation are particularly effective between the isotherms 300–450 °C (Saishu et al., 2014) implying the hydrological compartmentalization of the crust between the 370 °C and 450 °C isotherms, depending on the authors (Fournier, 1999; Uno et al., 2023). This alteration zone of low matrix and structural permeability probably assumes the geometry of an antiform at the top of the crystallizing granite as a result of the heat flux emanating from the latter (Fig. 7). This hydrothermally self-sealing cap generally corresponds to the maximum depth of the hydrothermal convection zone (Utkin and Afanasyev, 2021), which is probably limited to the first 2–3 km of depth in the case of Milos, approximately based on the geothermal gradients measured in the wells (Fytikas, 1989). Analysis of geochemical data on epithermal mineralization on the island of Milos has shown that this upper hydrological system is composed of a mixture of seawater/meteoric water/volcanic gases and structured into two types of reservoirs partially connected by fault and fracture systems (Naden et al., 2005). The first type is associated with lateral saline intrusions heated at 100–175 °C (0–200 m), and currently producing submarine brine seeps. The second type (2 km) is fed by seawater and heated at 205 °C–350 °C (Naden et al., 2005). In a third type of reservoirs located underneath the hydrothermal self-sealing cap of low permeability, magmatic liquids, gases and brines under lithostatic pressure may accumulate in connection with the latent crystallization of the magma chamber (Saishu et al., 2014). Occasionally, these fluids can sporadically inject themselves into the overlying hydrostatic system, destabilizing it in terms of pressure, temperature, flow rate and chemical composition (Muraoka et al., 1998; Fournier, 1999, Fig. 7). Liquids from the upper system may also descend underneath the hydrothermal self-cap during the seismic rupture (Geoffroy and Dorbath, 2008). Given field observations of fluid overpressures associated with the 1992 seismic rupture (Papanikolaou et al., 1993), it is likely that the Achivadolimni fault is an arterial fault in the sense of Sibson (2019), i.e., rooted in basal seismogenic zone reservoirs located around the BDT plumb (with the magma chamber and likely to be drained during episodic fault movements (i.e., hypocenter depth 9.6 km, Fig. 7). In fact, during seismic ruptures, the high strain rates imply a rheological brittle behaviour of fault rocks (e.g., 10−11, Sibson, 1982), although they are subjected to thermal conditions implying long-term ductility at low strain rates (Fournier et al., 1999; Fridleifsson and Elders, 2005; lower part of the Fig. 7).

5.3.2 Microseismicity and arterial fault

In tectonic-volcanic contexts, swarms of microseismicity are commonly explained by fluid circulation (e.g., Scholz, 2019; Vidale and Shearer, 2006), either per descensum during co-seismic rupture (e.g., Geoffroy and Dorbath, 2008), per ascensum (e.g., Geoffroy et al., 2022), or by lateral flow (Okamoto et al., 2022). They are often related to liquid to gas transition (i.e., boiling) in a closed system by inducing rapid decompression of fluid-filled fault-zones (Benson et al., 2014). This boiling depth may be associated with high fluid pressures (near-lithostatic pore pressures), which are located beneath a permeability barrier either below a hydrothermal self-sealing cap and / or at the base of a brittle crust (Fournier et al., 1999; Zencher et al., 2006). Mechanically, the increase in fluid pressure in the fault zone (fractures and pores of the fault rocks) reduces effective stresses and generates aseismic slip and/or microseismicity (Sherburn et al., 2015; Cox, 2016; Bentz et al., 2019). The aseismic slip in turn perturbs the surrounding stresses, resulting in microseismicity or even macroseismicity that migrates in time and space from the point of fluid overpressure toward the tip of the fault (Liu et al., 2024). Opening increments on active faults result in episodic structural permeability conducive to mineral neoformation (Sibson, 2019). Due to their abnormal geomechanical properties, these newly formed minerals cause frictional softening, which can also cause subsidence and associated microseismicity (Valsami and Baier, 2001; Bromley, 2014).

The Gulf of Milos was fully emerged during the last glacial maximum (bathymetry < 120 m, Karageorgis et al., 1998), which means that the hydrothermal system has been receiving seawater from the surface for 20,000 yr. This inflow of seawater in disequilibrium with the underlying systems has probably changed the hydrothermal dynamics of the system (temperature, salinity, Coulibaly et al., 2006, Fig. 7). For instance, the hypocentres of the seismic swarms located 1 to 2 km deeper beneath the gulf than beneath the volcanic emerged part of the island (Ochmann et al., 1989) suggest that the presence of seawater above the system could cool down and lower both the isotherms and the boiling depths (responsible for fluid overpressure in the faults) in this area compared to the surrounding area (Fig. 7; Tanaka, 2004; Fontaine et al., 2003). The spatio-temporal distribution of micro-earthquakes allows the mapping of the propagation front of hydrothermal fluid movements in the subsurface and the discussion of their origin and geometry (e.g., Jeanne et al., 2014; Okamoto et al., 2022). The relative spatial stability of the location of microearthquake hypocentres under the Gulf of Milos over time corroborates the fact that microseismicity is probably due to fluid circulation and liquid to gas transition into faults (Benson et al., 2014; Danrée et al., 2024), notably at the intersecting Fyriplaka and Zephyria grabens (Fig. 5). The perennial and notable microseismicity at Milos is probably associated with the large volume of fluids (partly boiling) trapped below the hydrothermal self-sealing cap, right above the magmatic chamber, and sporadically injected into the shallow hydrothermal loop during tectonic earthquakes. This sudden natural fluid injection (per ascensum) is permitted by active and arterial faults (Sibson, 2019; e.g., Achivadolimni fault) generating opening increments and structural permeability in the hydrothermal self-sealing cap during co-seismic ruptures beneath the active secant grabens of the archipelago.

6 Conclusion

This work shows the spatial relationships between the structural architecture at the scale of a volcanic arc archipelago shaped by multidirectional extension, the 1992 macroseismicity, the instrumental microseismicity, the perennial hydrothermal escapes and the sporadic overpressures observed at the surface (i.e., historical phreatic eruptions). In particular, we put in perspective the role of the seismogenic and arterial Achivadolimni fault potentially rupturing the self-sealed hydrothermal cap (variable in space and time), this latter generally compartmentalizing the shallow and deep hydrothermal convection loops. In terms of geological hazards, these observations lead us to reconsider the tectonic origins of the phreatic eruptions (Thiorychia crater field?) that led to the destruction of 200–300 yr B.C. −aged Aghia-Kyriaki roman city (Traineau and Dalabakis, 1989; Photos-Jones et al., 1999; Cavailhes et al., 2025b). Indeed, our work supports that the occurrence of a large earthquake (Mw > 5.5?) on Milos could rupture the 2–3 km deep-located hydrothermally self-sealed cap system of fractures, depressurize and chaotically drain the fluids (water, gases) probably trapped near the BDT and currently explaining the microseismicity hypocenters. The fact that the Mw 5.3 earthquake in 1992 did not produce any significant phreatic eruptions suggests that the earthquake, which produced a crater field of about 4 km2, must have been much larger to mobilize such amount of fluids (King and Wood, 1994). This chain of events seems to have happened about 2,300 yr ago and could happen again in the future. Detailed work is also needed to constrain the links between current shallow and deep hydrothermal convection loops, and natural hydrogeologic/hydraulic fluctuations at different time-scales (e.g., precipitation, eustatism, tides, sealing processes, etc.).

Acknowledgments

This research was partially supported by INSU funding from the CNRS (Centre National de la Recherche Scientifique). We extend our sincere thanks to Hervé Gillet and Vincent Hanquiez for their work in providing and refining the bathymetric and topographic maps in 2023 and 2024. We are also grateful to Margaux Saint-George for preparing the thin-sections used in this research project, the University of Bordeaux and the Research Department EPOC for their support. Our appreciation goes to Lydia Maigne, Vincent Breton, Lucas Terray, and Philippe Chardon for their valuable scientific discussions and field assistance in 2023. We warmly thank Pascal Allemand for his support. We are grateful to IFREMER and GENAVIR for making the 2025 “Physiography and Geology of Milos” cruise possible, as well as to the crews on board. Finally, we thank Olivier Vanderhaeghe, Yannick Branquet, Antonin Richard, and an anonymous reviewer for their constructive comments on the initial version of this manuscript.

References

  • Agostini S, Doglioni C, Innocenti F, Manetti P, Tonarini S. 2010. On the geodynamics of the Aegean rift. Tectonophysics 488 (1-4): 7–21. [Google Scholar]
  • Ambraseys N. 2009. Earthquakes in the Mediterranean and Middle East: a multidisciplinary study of seismicity up to 1900. Cambridge University Press. [Google Scholar]
  • Anastasakis G, Piper DJ. 2005. Late Neogene evolution of the western South Aegean volcanic arc: sedimentary imprint of volcanicity around Milos. Mar Geol 215 (3-4): 135–158. [Google Scholar]
  • Anastasakis G, Piper DJ, Dermitzakis MD, Karakitsios V. 2006. Upper Cenozoic stratigraphy and paleogeographic evolution of Myrtoon and adjacent basins, Aegean Sea, Greece. Mar Pet Geol 23 (3): 353–369. [Google Scholar]
  • Anderson EM. 1905. The dynamics of faulting. Trans Edinburgh Geol Soc 8 (3): 387–402. [Google Scholar]
  • Angelier J. 1977. Sur l’évolution tectonique depuis le Miocène supérieur d’un arc insulaire méditerranéen : l’arc Ogden. Rev Geogr Phys Gand01 Dyn (2), 19 (3): 271–274. [Google Scholar]
  • Angelier J. 1979. Recent Quaternary tectonics in the Hellenic Arc: examples of geological observations on land. In Whitten CA, Green R, Meade BK, eds. Recent crustal movements, 1977. Tectonophysics, 52: 267–275. [Google Scholar]
  • Belka A, Nomikou P, Bühring SI, Bejelou K, Lampidou D, Kothri S, et al. 2024. Submarine volcanic structures from high resolution swath data around Antimilos Aegean Sea, Greece. 2nd International Conference on Seafloor Landforms, Processes and Evolution Abstracts Volume p.39 [Google Scholar]
  • Benson PM, Vinciguerra S, Nasseri MH, Young RP. 2014. Laboratory simulations of fluid/gas induced micro-earthquakes: application to volcano seismology. Front Earth Sci 2: 32. [Google Scholar]
  • Bentz S, Martínez‐Garzón P, Kwiatek G, Dresen G, Bohnhoff M. 2019. Analysis of microseismicity framing ML> 2.5 earthquakes at the Geysers geothermal field, California. J Geophys Res: Solid Earth 124 (8): 8823–8843. [Google Scholar]
  • Bini G, Chiodini G, Lucchetti C, Moschini P, Caliro S, Mollo S, et al. 2020. Deep versus shallow sources of CO2 and Rn from a multi-parametric approach: the case of the Nisyros caldera (Aegean Arc, Greece). Sci Rep 10 (1): 13782. [Google Scholar]
  • Bohnhoff M, Rische M, Meier T, Becker D, Stavrakakis G, Harjes HP. 2006. Microseismic activity in the Hellenic Volcanic Arc, Greece, with emphasis on the seismotectonic setting of the Santorini-Amorgos zone. Tectonophysics 423 (1-4): 17–33. [Google Scholar]
  • Bonini M, Rudolph ML, Manga M. 2016. Long-and short-term triggering and modulation of mud volcano eruptions by earthquakes. Tectonophysics 672: 190–211. [Google Scholar]
  • Bouroullec R, Cartwright JA, Johnson HD, Lansigu C, Quémener JM, Savanier D. 2004. Syndepositional faulting in the Grès d’Annot Formation, SE France: high-resolution kinematic analysis and stratigraphic response to growth faulting. Geol Soc London Spec Publ 221 (1): 241–265. [Google Scholar]
  • Bromley CJ. 2014. Seismicity and subsidence: Examples of observed geothermal deformation synergies from New Zealand. In Proceedings. [Google Scholar]
  • Burhing SI. 2023. Bridging hydrothermal sites along the Hellenic Arc off Milos from shallow to deep. 2023. METEOR-Berichte, Cruise M192, Piraeus − Limassol, 08 August − 05 September 2023, 17 p. [Google Scholar]
  • Brun, J.-P., Faccenna, C., Gueydan, F., Sokoutis, D., Philippon, M., Kydonakis, K. & Gorini, C. (2016). The two-stage Aegean extension, from localized to distributed, a result of slab rollback acceleration. Canadian Journal of Earth Sciences, 53(11): 1142–1157. [Google Scholar]
  • Camerlenghi A, Urgeles R, Fantoni L. 2010. A database on submarine landslides of the Mediterranean Sea. Submarine Mass Movements Consequences 503–513. [Google Scholar]
  • Coulibaly, A. S., Anschutz, P., Blanc, G., Malaize, B., Pujol, C., & Fontanier, C. (2006). The effect of paleo-oceanographic changes on the sedimentary recording of hydrothermal activity in the Red Sea during the last 30,000 years. Marine geology, 226(1-2), 51-64. [Google Scholar]
  • Cavailhes T, Soliva R, Labaume P, Wibberley C, Sizun JP, Gout C, et al. 2013. Phyllosilicates formation in faults rocks: Implications for dormant fault‐sealing potential and fault strength in the upper crust. Geophys Res Lett 40 (16): 4272–4278. [Google Scholar]
  • Cavailhes T. 2025a. Physiography and Geology of the Gulf of Milos, PGG MILOS cruise, RV L’Europe, https://doi.org/10.17600/18002999. [Google Scholar]
  • Cavailhes T, Martelat J-E., Grandjean P, Augier S, Passot S, Allemand P, et al. 2025b. On the historical phreatic eruptions of Milos (Greece): an intertwined seismic/volcanic hazard? IAVCEI 2025 conference in Geneva from 29thJune to 4th July 2025. [Google Scholar]
  • Chiarabba C, De Gori P, Valoroso L, Petitta M, Carminati E. 2022. Large extensional earthquakes push-up terrific amount of fluids. Sci Rep 12 (1): 14597. [Google Scholar]
  • Chiodini G, Bini G, Massaro S, Caliro S, Kanellopoulos C, Tassi F, et al. 2023. Ascent and decompressional boiling of geothermal liquids tracked by solute mass balances: a key to understand the hydrothermal explosions of Milos (Greece). Front Earth Sci 11: 1254547. [Google Scholar]
  • Cotton CA. 1950. Tectonic scarps and fault valleys. Geol Soc Am Bull 61 (7): 717–758. [Google Scholar]
  • Cox SF. 2016. Injection-driven swarm seismicity and permeability enhancement: Implications for the dynamics of hydrothermal ore systems in high fluid-flux, overpressured faulting regimes—An invited paper. Econ Geol 111 (3): 559–587. [Google Scholar]
  • Daskalopoulou K, Gagliano AL, Calabrese S, Longo M, Hantzis K, Kyriakopoulos K, et al. 2018. Gas geochemistry and CO2 output estimation at the island of Milos, Greece. J Volcanol Geotherm Res 365: 13–22. [Google Scholar]
  • Dando PR, Hughes JA, Leahy Y, Niven SJ, Taylor LJ, Smith C. 1995. Gas venting rates from submarine hydrothermal areas around the island of Milos, Hellenic Volcanic Arc. Cont Shelf Res 15 (8): 913–929. [Google Scholar]
  • Dando PR, Aliani S, Arab H, Bianchi CN, Brehmer M, Cocito S, et al. 2000. Hydrothermal studies in the Aegean Sea. Phys Chem Earth Part B: Hydrol Oceans Atmos 25 (1): 1–8. [Google Scholar]
  • Danré P, De Barros L, Cappa F, Passarelli L. 2024. Parallel dynamics of slow slips and fluid-induced seismic swarms. Nat Commun 15 (1): 8943. [Google Scholar]
  • Delibasis ND, Drakopoulos JC. 1993. The Milos island earthquake of March 20, 1992 and its tectonic significance. Pure Appl Geophys 141 (1): 43–58. [Google Scholar]
  • Doke R, Harada M, Miyaoka K. 2018. GNSS Observation and Monitoring of the Hakone Volcano and the 2015 Unrest. J Disaster Res 13 (3): 526–534. [Google Scholar]
  • Drakopoulos J, Delibasis N. 1975. Volcanic type microearthquake activity in Melos, Greece. Anali di Geofisica 28: 131–153 [Google Scholar]
  • Ducoux, M., Branquet, Y., Jolivet, L., Arbaret, L., Grasemann, B., Rabillard, A., ... & Drufin, S. (2017). Synkinematic skarns and fluid drainage along detachments: The West Cycladic Detachment System on Serifos Island (Cyclades, Greece) and its related mineralization. Tectonophysics, 695, 1–26. [Google Scholar]
  • Faccenna C, Gueydan F, Sokoutis D, Philippon M, Kydonakis K, Gorini G, et al. 2016. The two-stage aegean extension, from localized to distributed, a result of slab rollback acceleration. Can J Earth Sci 53 (11): 1142–1157. https://doi.org/10.1139/cjes-2015-0203.insu-01271296. [CrossRef] [Google Scholar]
  • Feuillet N. 2013. The 2011–2012 unrest at Santorini rift: stress interaction between active faulting and volcanism. Geophys Res Lett 40 (14): 3532–3537. [Google Scholar]
  • Fitzsimons MF, Dando PR, Hughes JA, Thiermann F, Akoumianaki I, Pratt SM. 1997. Submarine hydrothermal brine seeps off Milos, Greece. Observations and geochemistry. Mar Chem 57 (3-4): 325–340. [Google Scholar]
  • Florinsky IV. 1996. Quantitative topographic method of fault morphology recognition. Geomorphology 16 (2): 103–119. [Google Scholar]
  • Fontaine FJ, Rabinowicz M, Boulegue J. 2003. Hydrothermal processes at Milos Island (Greek Cyclades) and the mechanisms of compaction-induced phreatic eruptions. Earth Planet Sci Lett 210 (1-2): 17–33. [Google Scholar]
  • Fournier, R. O. (1989). Geochemistry and dynamics of the Yellowstone National Park hydrothermal system. Annual Review of Earth and Planetary Sciences, Vol. 17, p. 13, 17, 13. [Google Scholar]
  • Fournier RO. 1999. Hydrothermal processes related to movement of fluid from plastic into brittle rock in the magmatic-epithermal environment. Econ Geol 94 (8): 1193–1211. [Google Scholar]
  • Frank FC. 1965. On dilatancy in relation to seismic sources. Rev Geophys 3: 485–503. [Google Scholar]
  • Fridleifsson GO, Elders WA. 2005. The Iceland Deep Drilling Project: a search for deep unconventional geothermal resources. Geothermics 34 (3): 269–285. [Google Scholar]
  • Fytikas M, Giuliani O, Innocenti F, Marinelli G, Mazzuoli R., 1976. Geochronological data on recent magmatism of the Aegean Sea. Tectonophysics 31 (1-2): T29–T34. https://doi.org/10.1016/0040-1951(76)90161-X. [Google Scholar]
  • Fytikas M. 1977. Geological Map of Greece: Milos Island: Athens, Greece, Institute of Geology and Mining Research (IGMR), Mapped during the Years 1971–1973. Scale 1: 25,000. [Google Scholar]
  • Fytikas M. 1989. Updating of the geological and geothermal research on Milos Island. Geothermics 18 (4): 485–496. https://doi.org/10.1016/0375-6505(89)90051-5. [Google Scholar]
  • Fytikas M, Garnish JD, Hutton VRS, Staroste E, Wohlenberg J. 1989. An integrated model for the geothermal field of Milos from geophysical experiments. Geothermics 18 (4): 611–621. [Google Scholar]
  • Fytikas M, Innocenti F, Kolios N, Manetti P, Mazzuoli R, Poli G, et al. 1986. Volcanology and petrology of volcanic products from the island of Milos and neighbouring islets. J Volcanol Geotherm Res 28 (3-4): 297–317. [Google Scholar]
  • Fytikas M, Vugiukalakis G. 1993, Volcanic structure and evolution of Kimolos and Polyegos Islands. Bull Greek Geol Soc 28/2: 221–237. [Google Scholar]
  • Ganas A, Karakonstantis A, Kapetanidis V, Tsironi V, Karamitros I, Efstathiou E, et al. 2022. Ground deformation and microseismicity patterns onshore Milos, Cyclades, Greece. Bull Geol Soc Greece. [Google Scholar]
  • Gautier P, Brun JP, Moriceau R, Sokoutis D, Martinod J, Jolivet L. 1999. Timing, kinematics and cause of Aegean extension: a scenario based on a comparison with simple analogue experiments. Tectonophysics 315: 31–72. [CrossRef] [Google Scholar]
  • Geoffroy, L., & Dorbath, C. (2008). Deep downward fluid percolation driven by localized crust dilatation in Iceland. Geophysical research letters, 35(17). [Google Scholar]
  • Geoffroy L, Dorbath C, Ágústsson K, Kristjánsdóttir S, Flóvenz ÓG, Doubre C, et al. 2022. Hydrothermal fluid flow triggered by an earthquake in Iceland. Commun Earth Environ 3 (1): 54. [Google Scholar]
  • Glodny J, Ring U. 2022. The Cycladic Blueschist Unit of the Hellenic subduction orogen: Protracted high-pressure metamorphism, decompression and reimbrication of a diachronous nappe stack. Earth-Sci Rev 224: 103883. [Google Scholar]
  • Gold T, Soter S. 1984. Fluid ascent through the solid lithosphere and its relation to earthquakes. Pure Appl Geophys 122: 492–530. [Google Scholar]
  • Granot R. 2016. Palaeozoic oceanic crust preserved beneath the eastern Mediterranean. Nat Geosci 9 (9): 701–705. [Google Scholar]
  • Grasemann B, Huet B, Schneider DA, Rice AHN, Lemonnier N, Tschegg C. 2018, Miocene postorogenic extension of the Eocene synorogenic imbricated Hellenic subduction channel : New constraints from Milos (Cyclades, Greece). GSA Bull 130 (1/2): 238–262. https://doi.org/10.1130/B31731.1. [Google Scholar]
  • Grosche A, Klemd R, Denkel K, Keith M, Haase KM, Voudouris PC, et al. 2023. Sources, transport, and deposition of metal (loid) s recorded by sulfide and rock geochemistry: constraints from a vertical profile through the epithermal Profitis Ilias Au prospect, Milos Island, Greece. Miner Depos 58 (6): 1101–1122. [Google Scholar]
  • Gueydan F, Sakellariou D, Paquette JL, Roger F, Alsaif M, Faucher A, et al. 2025. Late Miocene high-angle faulting in the Cyclades: offshore-onshore tectonic studies and U-Pb calcite dating. BSGF-Earth Sci Bull 196: 1. [Google Scholar]
  • Hall AJ, Photos‐Jones E, McNulty A, Turner D, McRobb A. 2003. Geothermal activity at the archaeological site of Aghia Kyriaki and its significance to Roman industrial mineral exploitation on Melos, Greece. Geoarchaeology 18 (3): 333–357. [Google Scholar]
  • Hauksson E, Jones LM, Hutton K, Eberhart‐Phillips D. 1993. The 1992 Landers earthquake sequence: Seismological observations. J Geophys Res: Solid Earth 98(B11): 19835–19858. [Google Scholar]
  • Hesthammer J, Fossen H. 1999. Evolution and geometries of gravitational collapse structures with examples from the Statfjord Field, northern North Sea. Mar Pet Geol 16 (3): 259–281. [Google Scholar]
  • Jarigue J-J. 1978, Etudes néotectoniques dans l’arc volcanique Egéen : Les îles de Kos, Santorin et Milos. [Google Scholar]
  • Jeanne P, Rutqvist J, Dobson PF, Walters M, Hartline C, Garcia J. 2014. The impacts of mechanical stress transfers caused by hydromechanical and thermal processes on fault stability during hydraulic stimulation in a deep geothermal reservoir. Int J Rock Mech Min Sci 72: 149–163. [Google Scholar]
  • Jolie E, Scott S, Faulds J, Chambefort I, Axelsson G, Gutiérrez-Negrín LC, et al. 2021. Geological controls on geothermal resources for power generation. Nat Rev Earth Environ 2 (5): 324–339. [Google Scholar]
  • Jolivet L, Brun JP. 2010. Cenozoic geodynamic evolution of the Aegean. Int J Earth Sci 99 (1): 109–138. [CrossRef] [Google Scholar]
  • Jolivet L, Gorini C, Smit J, Leroy S. 2015. Continental breakup and the dynamics of rifting in back‐arc basins: The Gulf of Lion margin. Tectonics 34 (4): 662–679. [CrossRef] [Google Scholar]
  • Jolivet L, Menant A, Roche V, Le Pourhiet L, Maillard A, Augier R, et al. 2021. Transfer zones in Mediterranean back-arc regions and tear faults. Zones de transfert dans les domaines arrière-arc méditerranéens et déchirures des panneaux plongeants. Bull Soc Géol Fr 192 (1). [Google Scholar]
  • Kanamori H, Anderson DL. 1975. Theoretical basis of some empirical relations in seismology. Bull Seismol Soc Am 65 (5): 1073–1095. [Google Scholar]
  • Karageorgis, A., Anagnostou, C., Sioulas, A., Chronis, G., & Papathanassiou, E. (1998). Sediment geochemistry and mineralogy in Milos bay, SW Kyklades, Aegean Sea, Greece. Journal of Marine Systems, 16 (3-4): 269–281. [Google Scholar]
  • Karageorgis AP, Ioakim C, Rousakis G, Sakellariou D, Vougioukalakis G, Panagiotopoulos IP, et al. 2016. Geomorphology, sedimentology and geochemistry in the marine area between Sifnos and Kimolos Islands, Greece. Bull Geol Soc Greece 50 (1): 334–344. [Google Scholar]
  • Kassaras I, Kapetanidis V, Ganas A, Tzanis A, Kosma C, Karakonstantis A, et al. 2020. The new seismotectonic atlasof Greece (v1. 0) and its implementation. Geosciences 10 (11): 447. [Google Scholar]
  • Khimasia A, Renshaw CE, Price RE, Pichler T. 2021. Hydrothermal flux and pore water geochemistry in Paleochori Bay, Milos, Greece. Chem Geol 571: 120188. [Google Scholar]
  • Kilias SP, Naden J, Cheliotis I, Shepherd TJ, Constandinidou H, Crossing J, et al. 2001. Epithermal gold mineralisation in the active Aegean volcanic arc: The Profitis Ilias deposit, Milos Island, Greece. Miner Depos 36: 32–44. [Google Scholar]
  • Kilias SP, Nomikou P, Papanikolaou D, Polymenakou PN, Godelitsas A, Argyraki A, et al. 2013. New insights into hydrothermal vent processes in the unique shallow-submarine arc-volcano, Kolumbo (Santorini), Greece. Sci Rep 3 (1): 2421. [Google Scholar]
  • King GC, Wood RM. 1994. The impact of earthquakes on fluids in the crust. Annals Geophys 37 (6). [Google Scholar]
  • Kokkalas S, Aydin A. 2013. Is there a link between faulting and magmatism in the south-central Aegean Sea? Geol Mag 150 (2): 193–224. [Google Scholar]
  • Kutterolf S, Freundt A, Druitt TH, McPhie J, Nomikou P, Pank K, et al. 2021. The medial offshore record of explosive volcanism along the central to eastern Aegean Volcanic Arc: 2. Tephra ages and volumes, eruption magnitudes and marine sedimentation rate variations. Geochem Geophys Geosyst 22: e2021GC010011. https://doi.org/10.1029/2021GC010011. [Google Scholar]
  • Leroy E, Cavailhes T, Anguy Y, Soliva R, Rotevatn A, Gaborieau C. 2023. Deformation bands and alteration in porous glass-rich volcaniclastics: Insights from Milos, Greece. J Struct Geol 177: 104982. [Google Scholar]
  • Le Pichon X, Angelier J., 1979. The Hellenic arc and trench system: a key to the neotectonic evolution of the eastern Mediterranean area. Tectonophysics 60: 1–42 [CrossRef] [Google Scholar]
  • Le Pichon X, Angelier J. 1981. The Aegean Sea. Philos Trans Royal Soc London Ser A, Math Phys Sci 300 (1454): 357–372. [Google Scholar]
  • Le Pichon X, Celâl Şengör AM, İmren C. 2019. A new approach to the opening of the eastern Mediterranean Sea and the origin of the Hellenic Subduction Zone. Part 1: The eastern Mediterranean Sea. Can J Earth Sci 56 (11): 1119–1143. https://doi.org/10.1139/cjes-2018-0128. [Google Scholar]
  • Liakopoulos A, Glasby GP, Papavassiliou CT, Boulegue J. 2001. Nature and origin of the Vani manganese deposit, Milos, Greece: an overview. Ore Geol Rev 18 (3-4): 181–209. [Google Scholar]
  • Liu M, Tan YJ, Lei X, Li H, Zhang Y, Wang W. 2024. Intersection between tectonic faults and magmatic systems promotes swarms with large-magnitude earthquakes around the Tengchong volcanic field, southeastern Tibetan Plateau. Geology 52 (4): 302–307. [Google Scholar]
  • Loreto MF, Düşünür-Doğan D, Üner S, İşcan-Alp Y, Ocakoğlu N, Cocchi L, et al. 2019. Fault-controlled deep hydrothermal flow in a back-arc tectonic setting, SE Tyrrhenian Sea. Sci Rep 9 (1): 17724. [Google Scholar]
  • Maggini M, Caputo R. 2021. Seismological data versus rheological modelling: Comparisons across the Aegean Region for improving the seismic hazard assessment. J Struct Geol 145: 104312. [Google Scholar]
  • Megalovasilis P. 2020. Geochemistry of hydrothermal particles in 1515 shallow submarine hydrothermal vents on Milos Island, Aegean 1516 Sea EastMediterranean, ISSN 0016-7029 Geochem Int 58 (2): 151- 1517 181. © Pleiades Publishing, Ltd., 2020. 1518 Mendrinos D, Choropanitis I, Polyzou O [Google Scholar]
  • Mayer K, Scheu B, Montanaro C, Yilmaz TI, Isaia R, Aßbichler D, et al. 2016. Hydrothermal alteration of surficial rocks at Solfatara (Campi Flegrei): Petrophysical properties and implications for phreatic eruption processes. J Volcanol Geotherm Res 320: 128–143. https://doi.org/10.1016/j.jvolgeores.2016.04.020. [Google Scholar]
  • Marguin V, Simpson G. 2023. Influence of fluids on earthquakes based on numerical modeling. J Geophys Res: Solid Earth 128 (2): e2022JB025132. [Google Scholar]
  • Marsellos AE, Kidd WSF, Garver JI, Kyriakopoulos KG. 2012, Exhumation of HP-rocks accompanied by low-angle normal faulting and associated detachment fault of Milos Island—Evidence from zircon fissiontrack thermochronology. Geochem Miner Petrol 49: 49–46. [Google Scholar]
  • Martelat JE, Escartín J, Barreyre T. 2020. Terrestrial shallow water hydrothermal outflow characterized from out of space. Mar Geol 422: 106119. [Google Scholar]
  • Mascle J, Martin L. 1990. Shallow structure and recent evolution of the Aegean Sea: a synthesis based on continuous reflection profiles. Mar Geol 94 (4): 271–299. [Google Scholar]
  • Masoumi N, Sadeghi H, Sarmad M. 2024. Dependence of horizontal PGA prediction on site conditions at critically short hypocentral distances: an analysis based on Japanese strong motion data, Annals Geophys 67 (3): S323. https://doi.org/10.4401/ag-9049. [Google Scholar]
  • Megalovasilis P. 2020. Geochemistry of hydrothermal particles in shallow submarine hydrothermal vents on Milos Island, Aegean Sea East Mediterranean, ISSN 0016-7029 Geochem Int 58 (2): 151–181. © Pleiades Publishing, Ltd., 2020. [Google Scholar]
  • Mendrinos D, Choropanitis I, Polyzou O, Karytsas C. 2010. Exploring for geothermal resources in Greece. Geothermics 39 (1): 124–137. [Google Scholar]
  • Mercier JL. 1981. Extensional-compressional tectonics associated with the Aegean Arc: comparison with the Andean Cordillera of south Peru-north Bolivia. Philos Trans Royal Soc London Ser A, Math Phys Sci 300 (1454): 337–355. [Google Scholar]
  • Montanaro C, Mick E, Salas-Navarro J, Caudron C, Cronin SJ, de Moor JM, et al. 2022. Phreatic and hydrothermal eruptions: from overlooked to looking over. Bull Volcanol 84 (6): 64. [Google Scholar]
  • Mattauer, M. (1973). Les déformations des matériaux de l'écorce terrestre. [Google Scholar]
  • Mouslopoulou V, Moraetis D, Benedetti L, Guillou V, Bellier O, Hristopulos D. 2014. Normal faulting in the forearc of the Hellenic subduction margin: Paleoearthquake history and kinematics of the Spili Fault, Crete, Greece. J Struct Geol 66: 298–308. [Google Scholar]
  • Muir Wood R. 1994. Earthquakes, strain-cycling and the mobilization of fluids. Geol Soc London Spec Publ 78 (1): 85–98. [Google Scholar]
  • Muraoka H, Uchida T, Sasada M, Yagi M, Akaku K, Sasaki M, et al. 1998. Deep geothermal resources survey program: igneous, metamorphic and hydrothermal processes in a well encountering 500C at 3729 m depth, Kakkonda, Japan. Geothermics 27 (5-6): 507–534. [Google Scholar]
  • Naden J, Kilias SP, Leng MJ, Cheliotis I, Shepherd TJ. 2003. Do fluid inclusions preserve δ18O values of hydrothermal fluids in epithermal systems over geological time? Evidence from paleo-and modern geothermal systems, Milos Island, Aegean Sea. Chem Geol 197 (1-4): 143–159. [Google Scholar]
  • Naden J, Kilias SP, Darbyshire DF. 2005. Active geothermal systems with entrained seawater as modern analogs for transitional volcanic-hosted massive sulfide and continental magmato-hydrothermal mineralization: The example of Milos Island, Greece. Geology 33 (7): 541–544. [Google Scholar]
  • Nicolas A. 1989, Principes de tectonique. Paris: Masson. [Google Scholar]
  • Nixon CW, Nærland K, Rotevatn A, Dimmen V, Sanderson DJ, Kristensen TB. 2020. Connectivity and network development of carbonate-hosted fault damage zones from western Malta. J Struct Geol 141: 104212. [Google Scholar]
  • Nomikou P, Papanikolaou D, Alexandri M, Sakellariou D, Rousakis G. 2013. Submarine volcanoes along the Aegean volcanic arc. Tectonophysics 597: 123–146. [Google Scholar]
  • Nomikou P, Papanikolaou D, Tibaldi A, Carey S, Livanos I, Bell KLC, et al. 2014. The detection of volcanic debris avalanches (VDAs) along the Hellenic Volcanic Arc, through marine geophysical techniques. In: Submarine Mass Movements and Their Consequences. Cham: Springer, pp. 339–349. [Google Scholar]
  • Nomikou P, Bühring SI, Bejelou K, Lampidou D, Ferreira C, Kothri S, et al. 2024. New insights into the morphology, extent and distribution of hydrothermal vents in the unique shallow-submarine volcanic seabed around Milos, Aegean Sea, Greece. In: 2nd International Conference on Seafloor Landforms, Processes and Evolution Abstracts Volume, p. 107 [Google Scholar]
  • Nowacki A, Wilks M, Kendall JM, Biggs J, Ayele A. 2018. Characterising hydrothermal fluid pathways beneath Aluto volcano, Main Ethiopian Rift, using shear wave splitting. J Volcanol Geotherm Res 356: 331–341. [Google Scholar]
  • Okamoto K, Imanishi K, Asanuma H. 2022. Structures and fluid flows inferred from the microseismic events around a low-resistivity anomaly in the Kakkonda geothermal field, Northeast Japan. Geothermics 100: 102320. [Google Scholar]
  • O’leary DW, Friedman JD, Pohn HA. 1976. Lineament, linear, lineation: some proposed new standards for old terms. Geol Soc Am Bull 87 (10): 1463–1469. [Google Scholar]
  • Ochmann N, Hollnack D, Wohlenberg J. 1989. Seismological exploration of the Milos geothermal reservoir, Greece. Geothermics 18 (4): 563–577. [Google Scholar]
  • Papachristou M, Voudouris K, Karakatsanis S, D’Alessandro W, Kyriakopoulos K, Baba A, et al. 2014. Geological setting, geothermal conditions and hydrochemistry of south and southeastern Aegean geothermal systems. Geotherm Syst Energy Resour: Turkey Greece 7: 47–75. [Google Scholar]
  • Papadopoulos GA, Pavlides SB. 1992. The large 1956 earthquake in the South Aegean: Macroseismic field configuration, faulting, and neotectonics of Amorgos Island. Earth Planet Sci Lett 113 (3): 383–396. [Google Scholar]
  • Papadopoulos GA. 1993. The 20 March 1992 South Aegean, Greece, earthquake (Ms= 5.3): possible anomalous effects. Terra Nova 5 (4): 399–404. [Google Scholar]
  • Papanikolaou −>Παπανικλάυ Δ, Λέκκας ΕΛ, Συσκάκης Δ, Αδαµπύλυ Ε. 1993. Correlation on neotectonic structures with the geodynamic activity in Milos during the earthquakes of March 1992. Δελτίν της Ελληνικής Γεωλγικής Εταιρίας 28 (3): 413. [Google Scholar]
  • Papanikolaou DJ, Royden LH. 2007. Disruption of the Hellenic arc: Late Miocene extensional detachment faults and steep Pliocene‐Quaternary normal faults—Or what happened at Corinth?. Tectonics 26 (5). [Google Scholar]
  • Parisio F, Vinciguerra S, Kolditz O, Nagel T. 2019. The brittle-ductile transition in active volcanoes. Sci Rep 9: 143 https://doi.org/10.1038/s41598-018-36505-x. [Google Scholar]
  • Petricca P, Babeyko AY. 2019. Tsunamigenic potential of crustal faults and subduction zones in the Mediterranean. Sci Rep 9 (1): 4326. [Google Scholar]
  • Pe GG, Piper DJW, 1972. Πη, Γ. 1972. Vulcanism at subduction zones: the Aegean area= Η ηϕαιστειóτης εις ζώνας καταδύσεως: η περιχή τυ Αιγαίυ. Δελτίν της Ελληνικής Γεωλγικής Εταιρίας 9 (2): 132–144. [Google Scholar]
  • Pe-Piper G, Piper DJW. 2005. The South Aegean active volcanic arc: relationships between magmatism and tectonics. In: Developments in Volcanology (Vol. 7, pp. 113–133). Elsevier. [Google Scholar]
  • Photos-Jones E, Hall AJ, Atkinson JA, Tompsett G, Cottier A, Sanders GDR. 1999. The Aghia Kyriaki, Melos survey: prospecting for the elusive earths in the Roman period in the Aegean1. Annual British School Athens 94: 377–413. [Google Scholar]
  • Piper DJW, Perissoratis C. 2003. Quaternary neotectonics of the South Aegean arc. Mar Geol 198 (3-4): 259–288. [Google Scholar]
  • Pirajno F. 2012. Hydrothermal mineral deposits: principles and fundamental concepts for the exploration geologist. Springer Science and Business Media. [Google Scholar]
  • Plimer. 2000. Milos Geological History: Athens, KOAN, 262 p. [Google Scholar]
  • Preine J, Hübscher C, Karstens J, Nomikou P. 2022. Volcanotectonic evolution of the Christiana Santorini-Kolumbo rift zone. Tectonics 41: e2022TC007524. [Google Scholar]
  • Principe C, Arias A, Zoppi U. 2002. Origin, transport and deposition of a debris avalanche deposit of phreatic origin on Milos Island (Greece), in: Montagne Pelee 1902-2002, Explosive Volcanism in Subduction Zones, Martinigue 12–16 May 2002, Abstracts p. 71. [Google Scholar]
  • Reilinger R, McClusky S, Paradissis D, Ergintav S, Vernant P. 2010. Geodetic constraints on the tectonic evolution of the Aegean region and strain accumulation along the Hellenic subduction zone. Tectonophysics 488 (1-4): 22–30 [CrossRef] [Google Scholar]
  • Reynolds SJ, Lister GS. 1987. Structural aspects of fluid-rock interactions in detachment zones. Geology 15 (4): 362–366. [CrossRef] [Google Scholar]
  • Ring U, Glodny J, Will T, Thomson S. 2010. The Hellenic subduction system: high pressure metamorphism, exhumation, normal faulting, and large-scale extension. Annual Rev Earth Planet Sci 38 (1): 45–76. [Google Scholar]
  • Roche V, Jolivet L, Papanikolaou D, Bozkurt E, Menant A, Rimmelé G. 2019. Slab fragmentation beneath the Aegean/Anatolia transition zone: Insights from the tectonic and metamorphic evolution of the Eastern Aegean region. Tectonophysics 754: 101–129. [Google Scholar]
  • Ross ZE, Cochran ES, Trugman DT, Smith JD. 2020. 3D fault architecture controls the dynamism of earthquake swarms. Science 368 (6497): 1357–1361. [Google Scholar]
  • Rowland J, Bardsley C, Downs D, Sepúlveda F, Simmons S, Scholz C. 2012. Tectonic controls on hydrothermal fluid flow in a rifting and migrating arc, Taupo Volcanic Zone, New Zealand. In: Proceedings of the 34th New Zealand geothermal workshop (Vol. 6) Auckland, New Zealand: University of Auckland. [Google Scholar]
  • Saishu H, Okamoto A, Tsuchiya N. 2014. The significance of silica precipitation on the formation of the permeable-impermeable boundary within Earth’s crust. Terra Nova 26 (4): 253–259. [Google Scholar]
  • Sakellariou D, Tsampouraki-Kraounaki K. 2019. Plio-Quaternary extension and strike slip tectonics in the Aegean. In: Transform plate boundaries and fracture zones, pp. 339–374. [Google Scholar]
  • Sander P. 2007. Lineaments in groundwater exploration: a review of applications and limitations. Hydrogeol J 15 (1): 71–74. [Google Scholar]
  • Schaarschimidt A, Haase KM, Klemd R, Keith M, Voudouris PC, Alfieris D, et al. 2021. Boiling effects on trace element and sulfur isotope compositions of sulfides in shallow-marine hydrothermal systems: Evidence from Milos Island, Greece. Chem Geol 583: 120457. [Google Scholar]
  • Scholz CH. 2019. The mechanics of earthquakes and faulting. Cambridge university press. [Google Scholar]
  • Searle MP, Lamont TN. 2022. Compressional origin of the Aegean orogeny, Greece. Geosci Front 13 (2): 101049. [Google Scholar]
  • Sherburn S, Sewell SM, Bourguignon S, Cumming W, Bannister S, Bardsley C, et al. 2015. Microseismicity at rotokawa geothermal field, New Zealand, 2008–2012. Geothermics 54: 23–34. [Google Scholar]
  • Sibson RH. 1981. Fluid flow accompanying faulting: field evidence and models. Earthquake Predict Int Rev 4: 593–603. [Google Scholar]
  • Sibson RH. 1982. Fault zone models, heat flow, and the depth distribution of earthquakes in the continental crust of the United States. Bull Seismol Soc Am 72 (1): 151–163. [Google Scholar]
  • Sibson RH. 1987. Earthquake rupturing as a mineralizing agent in hydrothermal systems. Geology 15 (8): 701–704. [Google Scholar]
  • Sibson RH. 1994. Crustal stress, faulting and fluid flow. Geol Soc London Spec Publ 78 (1): 69–84. [Google Scholar]
  • Sibson RH. 2000. Fluid involvement in normal faulting. J Geodyn 29 (3-5): 469–499. [CrossRef] [Google Scholar]
  • Sibson RH. 2001. Seismogenic framework for hydrothermal transport and ore deposition. [Google Scholar]
  • Sibson RH. 2019. Arterial faults and their role in mineralizing systems. Geosci Front 10 (6): 2093–2100. [Google Scholar]
  • Sieh K, Jones L, Hauksson E, Hudnut K, Eberhart-Phillips D, Heaton T, et al. 1993. Near-field investigations of the Landers earthquake sequence, April to July 1992. Science 260 (5105): 171–176. [Google Scholar]
  • Sonder RA. 1924. Zur Geologie and Petrographie der Inselgruppe yon Milos. Zeitschr Volc 8: 11–231. [Google Scholar]
  • Spagnoli F, Romeo T, Andaloro F, Canese S, Esposito V, Grassi M, et al. 2024. Seeps and Tectonic Structure of the Hydrothermal System of the Panarea Volcanic Complex (Aeolian Islands, Tyrrhenian Sea). Geosciences 14 (3): 60. [Google Scholar]
  • Stewart AL. 2003. Volcanic facies architecture and evolution of Milos, Greece. Doctoral dissertation, University of Tasmania. [Google Scholar]
  • Stewart AL, McPhie J. 2003. Internal structure and emplacement of an Upper Pliocene dacite cryptodome, Milos Island, Greece. J Volcanol Geotherm Res 124 (1-2): 129–148. [Google Scholar]
  • Tanaka A. 2004. Geothermal gradient and heat flow data in and around Japan (II) Crustal thermal structure and its relationship to seismogenic layer. Earth Planets Space 56 (12): 1195–1199. [Google Scholar]
  • Taylor B. (Ed.). 2013. Backarc basins: Tectonics and magmatism. Springer Science and Business Media. [Google Scholar]
  • Terakawa T, Zoporowski A, Galvan B, Miller SA. 2010. High-pressure fluid at hypocentral depths in the L’Aquila region inferred from earthquake focal mechanisms. Geology 38 (11): 995–998. [Google Scholar]
  • Traineau H, Dalabakis P. 1989. Mise en évidence d’une éruption phréatique historique sur l’ile de Milos (Grece). CR Acad Sci Paris 308(II): 247–252. [Google Scholar]
  • Tsokas GN, Hansen RO, Fytikas M, Vassilellis GD, Thanassoulas C. 1995. Geological and geophysical study of the island of Kimolos (Greece) and Geothermal implications. Geothermics 24 (5/6): 679–693. [Google Scholar]
  • Uno M, Okamoto A, Akatsuka T, Tsuchiya N. 2023. Continuous thermal structures of the present-day and contact-metamorphic geothermal systems revealed by drill cuttings in the Kakkonda geothermal field, Japan. Geothermics 115: 102806. [Google Scholar]
  • Utkin I, Afanasyev A. 2021. Decompaction weakening as a mechanism of fluid focusing in hydrothermal systems. J Geophys Res: Solid Earth 126 (9): e2021JB022397. [Google Scholar]
  • Uyeda S, Kanamori H. 1979. Back‐arc opening and the mode of subduction. J Geophys Res: Solid Earth 84(B3): 1049–1061. [Google Scholar]
  • Valsami-Jones E, Baier B. 2001. Relationship between hydrothermal fluids and microseismic activity on the south-east coast of Milos Island. Bull Geol Soc Greece 34 (4): 1441–1447. [Google Scholar]
  • Van Hinsbergen DJJ, Snel E, Garstman SA, Marunţeanu M, Langereis CG, Wortel MJR, et al. 2004, Vertical motions in the Aegean volcanic arc: Evidence for rapid subsidence preceding volcanic activity on Milos and Aegina. Mar Geol 209: 329–345, https://doi.org/10.1016/j.margeo.2004.06.006. [Google Scholar]
  • Vidale JE, Shearer PM. 2006. A survey of 71 earthquake bursts across southern California: Exploring the role of pore fluid pressure fluctuations and aseismic slip as drivers. J Geophys Res: Solid Earth 111(B5). [Google Scholar]
  • Voudouris P, Mavrogonatos C, Spry PG, Baker T, Melfos V, Klemd R, et al. 2019. Porphyry and epithermal deposits in Greece: An overview, new discoveries, and mineralogical constraints on their genesis. Ore Geol Rev 107: 654–691. [Google Scholar]
  • Walcott CR, White SH. 1998. Constraints on the kinematics of post-orogenic extension imposed by stretching lineations in the Aegean region. Tectonophysics 298 (1-3): 155–175. [Google Scholar]
  • Walsh JJ, Nicol A, Childs C. 2002. An alternative model for the growth of faults. J Struct Geol 24 (11): 1669–1675. [Google Scholar]
  • Wang, K., Chen, Q. F., Sun, S., & Wang, A. (2006). Predicting the 1975 Haicheng earthquake. Bulletin of the seismological society of America, 96(3): 757–795. [Google Scholar]
  • Weis P. 2012. The dynamic interplay between saline fluid flow and rock permeability in magmatic-hydrothermal systems. Crustal Permeability 373–392. [Google Scholar]
  • Wells DL, Coppersmith KJ. 1994. New empirical relationships among magnitude, rupture length, rupture width, rupture area, and surface displacement. Bull Seismol Soc Am 84 (4): 974–1002. [Google Scholar]
  • Withjack MO, Islam QT, La Pointe PR. 1995. Normal faults and their hanging-wall deformation: an experimental study. AAPG Bull 79 (1): 1–17. [Google Scholar]
  • Wyss M. 1979. Estimating maximum expectable magnitude of earthquakes from fault dimensions. Geology 7 (7): 336–340. [Google Scholar]
  • Zencher, F., Bonafede, M., & Stefansson, R. (2006). Near-lithostatic pore pressure at seismogenic depths: a thermoporoelastic model. Geophysical Journal International, 166(3): 13187–1334. [Google Scholar]
  • Zhou X, Kuiper K, Wijbrans J, Boehm K, Vroon P. 2021. Eruptive history and 40 Ar&z.urule; 39 Ar geochronology of the Milos volcanic field, Greece. Geochronology 3 (1): 273–297. [Google Scholar]

Cite this article as: Cavailhes T, Martelat J-E, Escartin J, Anguy Y, Augier S, Grandjean P, Nomikou P, Bejelou K. 2026. On the relations between morphotectonics, seismicity, perennial and sporadic hydrothermalism on the volcanic island complex of Milos (Greece), BSGF - Earth Sciences Bulletin 197: 6. https://doi.org/10.1051/bsgf/2025029

All Figures

Thumbnail: Fig. 1 Refer to the following caption and surrounding text. Fig. 1

Toponymic map showing the main localities and tectonic structures in the Aegean region mentioned in this study. See Section 4 (“Physiographic analysis”) and Figure 2 for more details. The studied Milos Archipelago forms part of the Aegean volcanic arc.

Carte toponymique indiquant les principales localités et structures tectoniques de la région égéenne mentionnées dans cette étude. Voir la section 4 (« Analyse physiographique ») et la Figure 2 pour plus de détails. L’archipel de Milos étudié fait partie de l’arc volcanique égéen.
In the text
Thumbnail: Fig. 2 Refer to the following caption and surrounding text. Fig. 2

(A) Topo-bathymetric and relief colored map of the Aegean Sea derived from SRTM (3 arc-second ca 30 m grid) and EMODNet Digital Bathymetry (1/16 arc-minute ca 115 m grid). (B) Associated morphotectonic sketch based on a detailed human-made lineaments mapping. Most of the lineaments are structural in origin and have been voluntarily overinterpreted to enhance discussion. The density of lineaments mostly depends on topo-bathymetric resolution and intensity of erosion, in particular on land. Milos appears to be located at the junction between the N110 trending southern part of the Myrtoon basin (My), the N-S trending structures dissecting the Aegean Sea, the Mid Cycladic Lineament (MCL) and­­ the N110 trending extensional faults leading to both the Cretan (C) and Christiana (Chr) basins. Hierarchy between neotectonics faults offsets can be positively correlated to the bathymetric relief. Am, Amorgos basin; An, Andros basin; Ar, Argolic basin; As, Astypalaea basin; C, Cretan basin; Chr, Christiana basin; Co, Corinth rift; CSK, Christiana-Santorini-Kolumbo rift; GoG, Gulf of Gokova; H, Heraklion Basin; Ka, Kamilonissi basin; Ma, Maleas basin; Mi-I, Mirthes-Ikaria fault; Myk, Mykonos; My, Myrtoon Basin; NI, North Ikaria Basin; NK, North Karpathos basin; SaG, Saronic Gulf; Si, Sikinos Basin; SK, South Karpathos basin.

(A) Carte topo-bathymétrique en couleurs de la mer Égée dérivée des données SRTM (grille de 3 secondes d’arc, soit environ 30 m) et de la bathymétrie numérique EMODNet (grille de 1/16 minute d’arc, soit environ 115 m). (B) Schéma structural associé basé sur une cartographie détaillée des linéaments. La plupart des linéaments sont d’origine tectonique et ont été volontairement surinterprétés afin d’enrichir la discussion. La densité des linéaments dépend principalement de la résolution topo-bathymétrique et de l’intensité de l’érosion, en particulier à terre. Milos semble être située à la jonction entre la partie sud du bassin de Myrtoon (My) orientée N110, les structures orientées N-S qui dissèquent la mer Égée, le Mid Cycladic Lineament (MCL) et les failles d’extension orientées N110 responsables des bassins crétois (C) et de Christiana (Chr). La hiérarchie entre les rejets des failles néotectoniques peut être positivement corrélée au relief bathymétrique. Am, Amorgos basin; An, Andros basin; Ar, Argolic basin; As, Astypalaea basin; C, Cretan basin; Chr, Christiana basin; Co, Corinth rift; CSK, Christiana-Santorini-Kolumbo rift; GoG, Gulf of Gokova; H, Heraklion Basin; Ka, Kamilonissi basin; Ma, Maleas basin; Mi-I, Mirthes-Ikaria fault; Myk, Mykonos; My, Myrtoon Basin; NI, North Ikaria Basin; NK, North Karpathos basin; SaG, Saronic Gulf; Si, Sikinos Basin; SK, South Karpathos basin.
In the text
Thumbnail: Fig. 3 Refer to the following caption and surrounding text. Fig. 3

Toponymy and synthetic geological map of Milos Island, based on the bibliography cited in the text and this study. The main tectonic structures (e.g., the Zephyria and Fyriplaka grabens), Quaternary volcanic centres (e.g., Trachylas, Fyriplaka), and mines (e.g., Cape Vani, Thiorychia) are also indicated.

Toponymie et carte géologique synthétique de l’île de Milos réalisée à partir de la bibliographie citée dans le texte et cette étude. Les principales structures tectoniques (e.g., le graben de Zephyria, le graben de Fyriplaka), les centres volcaniques du Quaternaire (e.g., Trachylas, Fyriplaka) et les mines (e.g., Cap Vani, Thiorychia) sont également indiqués.
In the text
Thumbnail: Fig. 4 Refer to the following caption and surrounding text. Fig. 4

(A) Topo-bathymetric and relief map of the Milos Archipelago derived from GEBCO data and completed with the bathymetric map of Nomikou et al. (2014). (B) Associated slope map derived from GEBCO data and including Antimilos, Kimolos and Polyegos. Both maps were used in addition to the cited literature to construct the synoptic structural sketch in Figure 5.

(A) Carte topo-bathymétrique de l’archipel de Milos dérivée des données GEBCO et complétée par la carte bathymétrique de Nomikou et al. (2014). (B) Carte des pentes associée dérivée des données GEBCO et incluant Antimilos, Kimolos et Polyegos. Les deux cartes ont été utilisées en complément de la littérature citée pour construire le schéma structural synoptique de la Figure 5.
In the text
Thumbnail: Fig. 5 Refer to the following caption and surrounding text. Fig. 5

 Structural sketch of the Milos archipelago derived from the physiographic analysis of Figure 4, showing the main structural lineaments (see the equal-area rose diagram for the strikes of the lineaments) and highlighting faults, fault propagation folds and fracture corridors. The Milos archipelago is located at the intersection of neotectonic volcanic grabens that express multidirectional extension. (A) The structural sketch is overlaid with the location of Quaternary volcanic centres, phreatic craters, hydrothermalized rocks (Fytikas, 1977) and hydrothermal vents (Fytikas and Vugiukalakis, 1993; Dando et al., 1995; Khimasia et al., 2021; Mendrino et al., 2010; Louis et al., 2003). (B) The structural sketch is crossplotted with the locations of the earthquake epicenters. References are on Figure 5B and comments are in the text. The inset on the left, taken from Ganas et al. (2022), shows the aligned epicentres along the NW-SE and N-S lineaments. The focal point mechanism shown in the figure is related to the earthquake that occurred the 20th of March 1992 (Delibasis, N. D., and Drakopoulos, J. C., 1993). The focal depth distribution of local earthquake is taken from Ochmann et al., (1989).

Schéma structural de l’archipel de Milos dérivée de l’analyse physiographique de la Figure 4, montrant les principaux linéaments structuraux (cf., projection stéréographique pour les orientations de linéaments) et mettant en évidence les failles, les plis de propagation des failles et les couloirs fracturés. L’archipel de Milos est situé à l’intersection de grabens volcaniques néotectoniques exprimant une extension multidirectionnelle. (A) Le schéma structural est superposé à l’emplacement des centres volcaniques quaternaires, des cratères phréatiques, des roches hydrothermales (Fytikas, 1977) et des évents hydrothermaux (Fytikas et Vugiukalakis, 1993; Dando et al., 1995; Khimasia et al., 2021; Mendrino et al., 2010; Louis et al., 2003). (B) Le schéma structural est superposé aux localisations des épicentres de séismes. Les références figurent sur la Figure 5B et les commentaires sont dans le texte. L’encart à gauche est issu de Ganas et al. (2022) et montrent les épicentres de séismes alignés sur les linéaments NW-SE et N-S. Le mécanisme du foyer représenté sur la figure est lié au séisme qui s’est produit le 20 mars 1992 (Delibasis and Drakopoulos 1993). La distribution des hypocentres de séismes est issue de Ochmann et al., (1989).
In the text
Thumbnail: Fig. 6 Refer to the following caption and surrounding text. Fig. 6

Synoptic sketches showing (A) the current location of the volcanic center of Milos at the starry-shape intersection of 3 main neotectonic grabens of the Aegean Sea. (B) Hydrothermalism and phreatic eruptions can be found along all fault lines east of the Achivadolimni fault, particularly near the Fyriplaka volcanic centre. Tectonic earthquakes are mainly located along the N-S and NW-SE grabens.

Schémas synoptiques montrant (A) l’emplacement actuel du centre volcanique de Milos à l’intersection des trois principaux grabens néotectoniques de la mer Égée. (B) L’hydrothermalisme et les éruptions phréatiques peuvent être localisés le long de toutes les directions de failles à l’est de la faille de Achivadolimni, notamment à proximité du centre volcanique de Fyriplaka. Les séismes tectoniques sont principalement localisés le long des grabens N-S et NW-SE.
In the text
Thumbnail: Fig. 7 Refer to the following caption and surrounding text. Fig. 7

Synoptic cross-section showing the horst and graben architecture of the island of Milos along an E-W transect. The lower part of the section is stylized from the work of Jolivet et al. (2021), showing the crustal-scale relationship between plutons and normal faults in a stretching continental context. The current hydrothermal system is located exclusively in the hanging-wall (to the east) of the Achivadolimni fault. The 370 °C isotherm probably limits the self-sealed hydrothermal cap that could correspond to a local “thermal BDT” (Fournier, 1999) in which matrix and structural permeabilities are very low. This zone represents the lower limit of convective circulation in the faults and fractures of the first 2–3 kilometres under hydrostatic pore pressure (R1, shallow reservoir, R2 deep reservoir, Naden et al., 2005). Lenses of fluids (gases, magmatic waters, meteoric waters and brines) under lithostatic pore pressure are trapped beneath the very low permeability hydrothermal self-sealing cap. During a significant seismic rupture (> Mw 5.5?), the structural drains open up, allowing significant hydraulic communication between the reservoir under lithostatic pressure and the surface reservoirs. The result is sudden destabilization due to temperature rise and decompression of surface reservoirs, which can lead to phreatic eruptions such as those that created the historic Thiorychia craters (Langada area, Fig. 3). On the right of the figure is the average geothermal gradient, 250-300 °C/km, recorded in the 5 boreholes of the Zephyria Graben. The toponymy is projected onto the section. The distinctions between the different lithologies (e.g., basement, metamorphic basement, Miocene cover, Miocene-Pliocene volcanism, domes and flows, Fytikas et al. (1989)) have not been included to make the figure easier to read. In the geophysical sense, the brittle-ductile transition (BDT) is defined as the depth above which most earthquake hypocentres are located (Maggini and Caputo, 2021). It has also been reported on the figure.

Schéma synoptique montrant l’architecture en horst et graben de l’île de Milos suivant un transect E-W. La partie inférieure de la coupe est stylisée à partir des travaux de Jolivet et al. (2021), montrant la relation à l’échelle de la croute entre les plutons et les failles normales en contexte d’étirement. Le système hydrothermal actuel se situe exclusivement dans le bloc de toit (à l’est) de la faille de Achivadolimni. L’isotherme 370 °C limite probablement le « hydrothermal self-sealed cap » pouvant correspondre localement à la « brittle-ductile transition zone » au sens thermique du terme (Fournier, 1999) dans laquelle les perméabilités matricielle et structurale sont très faibles. Cette zone limite par le bas une circulation convective dans les failles et fractures des 2-3 premiers kilomètres sous pression de pore hydrostatique. Sous le « hydrothermal self-sealed cap » de très faible perméabilité, sont piégés des lentilles de fluides (gaz, eaux magmatiques, météoriques et saumures magmatiques) sous pression de pore lithostatique. La dilation de ces fluides par décompression et refroidissement, notamment à l’aplomb du golfe de Milos, génère probablement la micro-sismicité dans les core-zones et damage zones de la faille artérielle de Achivadolimni. Au cours d’une rupture sismique significative (> Mw 5.5 ?), les drains structuraux s’ouvrent, autorisent une communication hydraulique significative entre le réservoir sous pression lithostatique et les réservoirs superficiels. Il en résulte une déstabilisation soudaine par élévation de température et décompression des réservoirs superficiels pouvant engendrer des éruptions phréatiques, telles que celles à l’origine des cratères historiques de Thiorychia (Langada area, Fig. 3). A droite de la figure, le gradient géothermique moyen, 300 °C / km, compilé dans les 5 puits du graben de Zephyria, est figuré. La toponymie est projetée sur la coupe. Les distinctions entre les différentes lithologies (e.g., socle, socle métamorphique, couverture Miocène, volcanisme mio-pliocène, dômes et coulées, Fytikas et al. (1989)) n’ont pas été reportées pour faciliter la lecture ­­de la figure. La Brittle-Ductile Transition (BDT) au sens géophysique est définie comme la profondeur au-dessus de laquelle l’essentiel des hypocentres de séismes est localisé (Maggini and Caputo, 2021). Elle est également indiquée sur la figure.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.