Complexity in monogenetic volcanic systems: factors influencing alternating magmatic and phreatomagmatic eruption styles at the 5 ka Mt. Gambier Volcanic Complex, South Australia

Monogenetic volcanoes are the most common type of volcanoes n Earth, however, with the recently growing research interest in them it has been shown that despite the small scale of these volcanic systems, great complexities can exist. Little is known about the complexity these volcanic centres potentially exhibit, especially when the eruptions are of a pre-historic age. However, with the current urban sprawl, cities encroach the borders of monogenetic volcanic fields; a better understanding of these systems is essential for further hazard assessment and emergency planning. The pre-historic Mt. Gambier Volcanic Complex (MGVC), with an age of ~5 ka is Australia’s youngest volcano, and is a complex, monogenetic centre of maars and tuff cones. It is located in the westernmost part of the monogenetic, intraplate, basaltic Newer Volcanics Province of south-eastern Australia (4.6 Ma-Recent), which is still regarded active. The MGVC consists of >14 eruption points and displays a complex stratigraphy with a variety of deposits related to changing magmatic and phreatomagmatic eruption styles. Based on the characteristics of the different deposits, six main facies have been identified. These facies and the interpreted eruption origins are: 1) coherent to vesicular fragmental alkali basalt (effusive/hawaiian spatter and lava flows); 2) massive scoriaceous fine lapilli with coarse ash (strombolian fallout); 3) bedded scoriaceous fine lapilli tuff (violent strombolian fallout); 4) thinmedium bedded, undulating very fine lapilli in coarse ash (dry phreatomagmatic surge modified fallout); 5) palagonite-altered, cross-bedded, medium lapilli to fine ash (wet phreatomagmatic base surges); and 6) massive, palagonite-altered, very poorly sorted tuff breccia and lapilli tuff (phreatovulcanian pyroclastic flows). Based on the established stratigraphy, available well-log data and a 3D computational model, the volumes of the erupted magmas, excavated country rock and the craters have been estimated. The total minimum volume of erupted material is 3.25 x 10^8 m3 (0.325 km3). In total at least 1.08 x 10^8 m3 of trachybasalt (dense rock equivalent, DRE) and 0.89 x 10^8 m3 DRE of basanite were erupted. Overall more magma was involved in phreatomagmatic activity compared to magmatic activity, however, the basanitic magma itself involved more magmatic than phreatomagmatic activity. For the two deposits with the widest dispersal column heights of between 5,000 m and 10,000 m were reconstructed using Tephra2 inversion modelling. Based on the volume and column heights the MGVC eruption was determined to scale 4 on the Volcanic Explosivity Index (VEI), with a magnitude of 4.7 and intensity of 8.2 to 8.9. The energy budgets of most phreatomagmatic phases of Mt. Gambier can be better explained by the incremental growth model than the major explosion dominated model, except for smaller phreatomagmatic and phreato-vulcanian explosions which fit both models. However, this first attempt to quantify the energy involved in the incremental growth of a maar-diatreme is still very simplistic. The MGVC was the product of two simultaneously erupting alkali basaltic magma batches: the more alkaline and light-rare-earth enriched basanites in the west (Mg# 56-62) and the more SiO2 and CaO enriched trachybasalts in the east (Mg# 60-64). Both magmatic suites have similar 87Sr/86Sr and 143Nd/144Nd and all the samples cover a narrow range (0.70399-0.70418, 0.512792-0.51282; εNd 3.0-3.6). Pb isotope compositions also show narrow ranges (206Pb/204Pb 18.518-18.616, 207Pb/204Pb 15.603-15.621, and 208Pb/204Pb 38.795-38.928). The two magma batches originated from the same source where it was formed as a result of the Indian MORB source interacting with a metasomatised source. The two magmatic suites originated from the same asthenospheric depth (2.2 GPa; ~80km) by 4-5% partial melting from a metasomatised lherzolite source. The original basanitic melt fractionated a high-pressure mineral assemblage of amphibole + clinopyroxene ± olivine ± spinel during ascent. At the base of the lithosphere a part of the melt interacted with a near-solidus pyroxenite body. At lithospheric levels, ascent rates increased to >10 cm/s. The two massive, poorly sorted lapilli tuff and tuff breccia deposits (facies 6), originating from the Blue Lake East and Valley Lake maars, have been interpreted as phreatomagmatic pyroclastic flow deposits based on the presence of a fine-grained, stratified base, a massive, extremely poorly sorted middle section, and a fine-grained, laminated deposit at the top. Using palaeomagnetic thermoremanent magnetisation (TRM) analysis of accessory basalt lithic clasts, the emplacement temperatures of these deposits have been estimated. The data show two clusters of unblocking temperatures: 400-500oC and 180-240oC. The palaeomagnetic fields recorded by high temperature data are scattered, indicating that the break between eruptive phases was enough to cool the lava below the Curie temperature, which we evaluate at some months. The lower temperature data are interpreted to correspond to the emplacement temperature of the pyroclastic flow deposits, which is confirmed by the presence of low-grade carbonised wood fragments in the deposits. Analysing the energy budget of these eruptive phases indicates that ~25% of the total thermal energy of the original magma was preserved at the time of deposition.