A multi-scale analysis of factors controlling the dynamics of basaltic volcanic fields: Newer Volcanics Province, Australia

2017-04-23T23:22:09Z (GMT) by Jackson Cornelius van den Hove
Basaltic Volcanic Fields (BVFs) occur on all continents, in all tectonic environments and host a diverse range of small scale basaltic volcanoes, which include the most numerous types of volcanic edifices on Earth. They are associated with small, often monogenetic, eruptions with long periods of quiescence. BVF are poorly studied compared with more obviously threatening, and high eruptive flux volcanic systems (volcanic arcs, mid-ocean ridges, continental rifts, and mantle plumes). The range of models used to explain what drives volcanism at the dozens of BVFs studied globally, stems from the variety of tectonic settings in which they occur. The complexity and inconsistency between the various models accounts for the need for further research on how these fields develop and the hazards they pose to society. BVFs that have been studied in detail are predominantly geometrically smaller examples, with characteristics that suggest relationships with local tectonic processes.

The Newer Volcanics Province (NVP) is an expansive Pliocene to Recent intraplate basaltic plains province, located in south-eastern Australia. The NVP has several aspects that make it an interesting and unique BVF to study. It is not readily relatable to any tectonic processes expressed at the surface, it occurs in a compressional lithospheric stress field setting, and it is host to some of the world’s largest maar volcanoes. The NVP therefore provides an ideal case study for larger end-member examples of both volcanoes and BVFs. The NVP as a whole, and many of its volcanoes, have been the subject of geochemical, deep geophysical imaging, physical volcanology, and limited age dating studies over the past half century. Despite this, there are significant gaps in understanding what controls on volcanism in a compressive stress regime, and the formation of very large maar volcanoes. Potential field modelling and spatial analysis methods are proven effective methods in studying basaltic volcanoes and volcanic fields. Hence, they are used in this thesis to investigate the aforementioned unique aspects of the NVP, with the implications being relevant to cases of basaltic volcanoes and BVFs worldwide.

Lake Purrumbete Maar (LPM) is a ~50 ka yrs old, large maar volcano with a crater that is up to 2,800 m in diameter. Despite its age, Lake Purrumbete’s near circular crater is well preserved, having undergone only minor erosion and is. It is one of a number of maar volcanoes of the NVP that rank amongst the largest examples in the world. Forward and inverse potential field modelling is used to constrain the subsurface structures related to the maar to assist in determining the factors that control the formation of such a large maar. Results show that LPM is the result of at least four coalesced vents that have produced a large shallow bowl shaped diatreme system, and not a deep conical feature. This is consistent with features typical of maars hosted in unconsolidated sediments, which is suggested for LPM by the occurrence of irregular marl lithic clasts with peperitic textures in the tephra ring deposits. Geometry inversions of the magnetic data indicate that the vents extend to a greater depth than inferred by accidental lithics present in the volcanic deposits (<250 m). This supports recent work that shows accidental lithics present in the volcanic successions likely provide an underestimate of the maximum depth of explosive eruptions. The vents of LPM show no discernible alignment, although previous authors suggest its emplacement was likely influenced by pre-existing crustal structures which it is known to overly.

At a larger scale, this study has undertaken spatial analysis of the location of volcanoes of the NVP to assess if crustal structures or tectonic processes have influenced their locations. Before undertaking alignment analysis on a point-set of NVP volcanoes, the capability of several reproducible alignment identification methods (Hough transform, random sample consensus (RANSAC), three-point regression, and two-point azimuth methods) were tested and compared against one another to determine their suitability. This was done using synthetic point-sets that have enforced alignments and clustered point distributions aimed at replicating the distribution of BVFs. The Hough transform method was the most robust method for identifying enforced alignment trends within the point-sets. The three-point regression method also was effective in identifying significant alignment trends, but produces a high percentage of coincidental alignments in clustered point-sets. The RANSAC method adopted from the field of image analysis was the least effective method tested. It was unable to consistently reproduce enforced alignment trends in clustered point-sets. The two-point azimuth method is ideal for validating alignment trends identified by the other methods that identify and define the locations of individual significant alignments.

The Hough transform and two-point azimuth methods were used along with cluster analysis methods to interrogate the distribution of NVP eruption centres (the term “eruption centre” is used in this thesis to refer to small monogenetic volcanoes that primarily form from a single eruption event) (Boyce, 2013) and a point-set of coeval vents (An eruption centre may have a one or more vents/eruption points that form during the single eruption event. Two or more vents related to the same eruption centre are termed “coeval vents” throughout this thesis) (Tibaldi, 1995).. Results show that alignments between vents are primarily oriented parallel with σHmax, whilst the alignment trends between eruption centres are preferentially aligned with pre-existing faults and predominantly along fault trends oriented near parallel with σHmax throughout most of the NVP. It is suggested that pre-existing faults could play an important role in preventing dikes from stalling and forming sills where BVFs such as the NVP are hosted in a compressive tectonic setting. Like most BVFs the NVP has a clustered distribution of eruption centres, whilst individual clusters are observed with more random to uniform distributions. The distribution of vents the NVP shows a good correlation with areas of thin lithosphere. The thickness of the lithosphere is likely a major factor controlling the location of volcanism in conjunction with edge-driven convection asthenospheric upwelling. The shear-driven upwelling of the asthenosphere into zones of thin/extended lithosphere is used to explain volcanism in other cases worldwide including; the Pannonian Basin (Central Europe), and volcanism along the Rio Grande rift (North America).

Controls on volcanism at sixteen BVFs including the NVP are discussed, based on a comparison of their size, volume and eruptive flux. First, measurement of the dimensions of the NVP was undertaken by mapping and amending current geological maps of extrusive volcanic deposits using high-resolution aeromagnetic data, and modelling volume using deposit thickness data from >1472 boreholes. Results show the NVP is large (23,100±530 km2) voluminous (680–900 km3 DRE) and high-flux (0.15–0.2 km3/ka) example relative to comparable low-flux IBVFs (0.0001 – 0.1 km3/ka). All the BVFs used for comparison have eruptive fluxes an order of magnitude or more less than examples of plume related volcanoes (Kilauea) and BVFs (Eastern Snake River Plains). Most lower flux BVFs also show no systematic age migration pattern in volcanism suggestive of a fixed mantle plume, and those with detailed geochronology and volume data often show a correlation between their eruptive flux and the rate of local tectonic processes. It is suggested that the NVP and most low- and high-flux BVFs are the result of upwelling occurring in the asthenosphere, related to tectonic processes; without requiring additional thermal input from a deep mantle source, as is inferred for several cases. Considering a control on volcanism by tectonic processes, the range of eruptive flux of tectonically controlled BVFs is related to variations in the rate of the effecting tectonic process, mantle composition, and the size of the mantle source zone where melt generation and accumulation is taking place.