Spatio-temporal evolution of a dispersed magmatic system and its implications for volcano growth, Jeju Island Volcanic Field, Korea

Abstract

Jeju Island is the emergent portion of a basaltic volcanic field developed over the last c. 1.8 Ma on continental crust. Initial volcanism comprised dispersed, small-volume (< 0.01 km³) alkali basaltic eruptions that incrementally constructed a tuff pile. Lavas and scoria from continuing small-scaled monogenetic volcanism capped this foundation. From c. 0.4 Ma large-volume (> 1 km³) eruptions began, with lavas building a composite shield. Three magma suites can be recognized: Early Pleistocene high-Al alkali (HAA), and Late Pleistocene to Holocene low-Al alkali (LAA) and subalkali (SA). The chemical similarity between small-volume and primitive large-volume eruptions suggests analogous parent magmas and fractionation histories that are independent of erupted volumes. The large-volume magmas evolved to trachyte, which erupted in two distinct episodes: the HAA Sanbangsan suite at c. 750 ka and the LAA Hallasan suite at c. 25 ka. Sr and Nd isotopes indicate that the early trachytes were contaminated by upper crustal material, whereas the later magmas were not. Both suites bear a Nd isotope signature indicative of lower crustal interaction. Sub-suites transitional between HAA and LAA, and between LAA and SA, indicate that melting occurred in discrete, but adjacent, mantle domains. Throughout the evolution of this volcano, each magma batch erupted separately, and a centralized plumbing system was never created. The Island's central peak (Mt. Halla 1950 m a.s.l.) is therefore not a sensu stricto stratovolcano, but marks the point of peak magma output in a distributed magmatic system. Jeju's shape and topography thus represent the spatial variation of fertility of the mantle below it. An increase in melt production in the Late Pleistocene was related to a deepening of the melting zone due to regional tectonic rearrangements. Temporal coincidences between magmatic pulses on Jeju and large-scale caldera eruptive events along the nearest subduction system in Kyushu, Japan, suggest that tectonic extension and changing strain rates may drive volcanism on a regional basis, influencing the intraplate volcanism of Jeju Island.

Introduction

Intraplate volcanism is a widespread phenomenon on Earth, occurring in both oceanic and continental settings and ranging in scale between the extremes of small-volume dispersed volcanism, forming monogenetic volcano fields, up to large-volume plume or rift-related volcanoes that can build polygenetic volcanic edifices of considerable size. The mode of magma generation, storage, ascent and eruption in these systems reflects tectonic stress and strain distribution (Brenna et al., 2011, Nakamura, 1977, Takada, 1994, Valentine and Hirano, 2010, Valentine and Perry, 2007), the magmatic footprints (Valentine and Perry, 2006) and the degree of partial melting (Frey et al., 1978, Sato et al., 1990), or the sustainability of magma supply (Fedotov, 1981).

Intraplate volcanism is most commonly monogenetic, i.e., magmas are sourced and erupted independently. This contrasts with polygenetic volcanism, in which magmas are interconnected at depth and a repeatedly re-used conduit system develops, resulting in an edifice made up of the products of many thousands of eruptions (Cañón-Tapia and Walker, 2004). Consequences of these systematic differences include low cumulative volumes of commonly primitive compositions in monogenetic eruptions but large volumes and evolved magmas in polygenetic volcanoes. If the two types of system are superposed, distinguishing between them requires a detailed knowledge of chemical variability and eruption chronology.

Small monogenetic volcanic centers can often be sampled throughout their entire eruption sequence, which allows detailed insights into their magmatic origins (Blondes et al., 2008, Brenna et al., 2010, Brenna et al., 2011, McGee et al., 2011, Smith et al., 2008, Sohn et al., 2012, Strong and Wolff, 2003). These single-volcano studies provide snapshots of the evolution of distributed volcanic fields that may contain many hundreds of monogenetic centers sporadically erupted over hundreds of thousands to millions of years. A similar investigation for a large composite volcano is however hindered by the burial of the majority of early eruption products, which conceal a large part of the volcano development history. Jeju Island, the subaerial part of a volcanic field immediately south of the Korean Peninsula, has been extensively and deeply drilled to explore and exploit groundwater resources (Koh, 2005, Won et al., 2006). This has provided the basis for a thorough understanding of the geological history, age and structural development of the island (Koh and Park, 2010a, Koh and Park, 2010b, Koh et al., 2008, Sohn and Park, 2004, Sohn et al., 2008). Drill cores on Jeju also provide a unique opportunity to investigate the evolution of an entire volcanic system, comprising both dispersed, small-volume, monogenetic volcanism and large-volume volcanism that has built a central composite volcanic edifice.

Here we address the relationship between small- and large-volume monogenetic volcanism in focused versus dispersed settings, as exemplified on Jeju Island. Further, we discuss the implications of our model for other intraplate volcanic areas and volcanic edifice building and briefly compare it to the stable magmatic conduit systems established beneath typical andesitic stratovolcanoes (Annen et al., 2006, Turner et al., 2011, Zellmer and Annen, 2008). We present new major and trace element and Sr–Nd–Pb chemical data from both surface samples and three deep cores (400–500 m) penetrating the entire lava succession in the central part of Jeju Island. From this we develop a model for the spatio-temporal evolution of the island's magmatic system from its infancy that explains its form and time-varying eruptive behavior. This example provides new insights into how long-lived mantle-fed volcanism operates in continental intraplate areas.