New Insights into the 1998-2001"Serreta" Eruption (Azores) from a Marine Geological Investigation (Faivi Cruise, September 2011)

In September 2011 a marine geological survey, the FAIVI (Features of Acores and Italian Volcanic Islands) cruise, was carried offshore the Terceira Island aboard the French R/V L'Atalante, in the framework of the Eurofleet program. One of the targets of the FAIVI cruise was the study of the Terceira volcanic edifice, particularly the area offshore the NW sector of the island (Serreta Ridge). Here an intermediate-depth submarine eruption occurred in 1998-2001, producing a consistent amount of lava balloons, i.e. balloon-like structures consisting of a thin lava shell surrounding a closed hollow interior, normally corresponding to a single large vesicle, or a few large convoluted vesicles, floating on the sea (Gaspar et al., 2003; Kueppers et al., 2012). Lava balloons are fairly unusual products of submarine eruptions. To date, balloon-like products have only been documented at Pantelleria, Italy in 1891 (Riccò, 1982; Washington, 1909), Mauna Loa, Hawaii in 1877 (Moore et al., 1985); Socorro Island, Mexico in 1993 (Siebe et al., 1995); at the 1998-2001 Serreta eruption offshore Terceira island (Azores) (Gaspar & Wallenstein, 1999) and, very recently offshore El Hierro, Canary Islands in October 2011 (Troll et al., 2012). Their genesis, in the framework of shallow-intermediate-depth submarine eruptions, is still poorly understood although several efforts have been made to improve the knowledge of the mechanisms producing such basaltic floating scorias (Kueppers et al., 2012). In order to reconstruct the style of eruption producing lava balloons and associated volcaniclastic deposits at Serreta, the investigations focused on: i) the identification and characterization of the emission vent(s) associated to the 1998-2001 submarine eruption; ii) the characterization of related deposits dispersal and grain size and, iii) the quantification of physical parameters of the eruptive event from the characteristics of sampled products (e.g., volatile contents, cooling rate). Here, we present preliminary results of the FAIVI cruise and attempt a first comparison with the 1891 eruption at Pantelleria, object of recent marine geological studies (Conte et al., submitted). As a first result, the FAIVI survey allowed to reconstruct with unprecedented detail the bathymetry offshore Terceira Island and, particularly, in the Serreta Ridge area. The latter is characterized by several volcanic features such as linear eruptive centres, volcanic cones and lava flows. Furthermore, bathymetric backscatter data and seafloor samples (collected through rock- and gravity- coring, grabbing and dredging) were preliminary analyzed on board and gave the following results: 1) The main vent of 1998-2001 Serreta eruption was identified; it is a half-cone, some 150 m high with a diameter of 700 m, resting on the top of the Serreta Ridge at 250-300 m water depth. 2) Most samples collected in the area surrounding the vent have been interpreted as related to the 1998-2001 Serreta eruption. They include, in fact, a large amount of fresh, highly vesciculated basaltic scoria, with the typical layered vescicularity observed in the lava balloon crust recovered intact during the 1998-2001 eruption. Material sampled from the top of Serreta Ridge encompasses different types, from large scoriaceous blocks (up to 60 cm length) of black fresh basalt characterized by numerous, large, highly convoluted vesicles with smooth internal walls, to fragments of scoriaceous material with dimensions <20 cm and irregular shapes. The origin of the larger clasts may be related to the deposition of collapsed ballons or autoclastic brecciation of lava flows. Moreover, the cores from the Serreta Ridge summit sampled normal-graded, poorly sorted layers of ash-sized glassy fragments, scoriaceous and poorly vesiculated lapilli and bombs with a maximum clast size <10 cm. Deposit dispersal, morphological features of larger clasts, as well as stratigraphy of lapilli layers lead to hypothesize the occurrence of different pulses of activity and/or some changes in the eruptive style. 3) The steep volcaniclastic apron south of the Serreta Ridge was also sampled through piston cores, from ~520 to over 1500 m bsl. Some main lithological units were identified, possibly related to the occurrence of mass and gravity-flows. As far as the comparison with the Pantelleria 1891 eruption the following points can be highlighted: 1) Despite the very different geodynamic setting, the geomorphology of the hypothesized vents and the macroscopic characteristics of volcanic products are surprisingly similar. Highly vesciculated, scoriaceous lavas were erupted in both cases close to the vent, while sand-sized glass shards are present in the surrounding areas; 2) In both cases the eruptions produced conical edifices (even if in Serreta half of the cone is missing); whereas, summit craters and significant lava flows on the flanks were not observed; 3) The volume of such cones, although very different (~700,000 cubic meters in Pantelleria; ~36,000,000 in Serreta) seem to be proportional to the duration of the lava balloons emission observed at the sea surface.