Enriched components in the Hawaiian plume: Evidence from Kahoolawe Volcano, Hawaii

Abstract

The geochemical differences between individual Hawaiian shields provide clues to the magma source components in the Hawaiian plume. Lavas from Koolau (Makapuu‐stage) and Kahoolawe volcanoes define the enriched, i.e., relatively high 87Sr/86Sr and low 143Nd/144Nd, extreme for Hawaiian shield lavas. There are, however, important geochemical differences between these shields; Kahoolawe lavas lack the relatively high SiO2, low CaO, and high Sr/Nb and La/Nb that are characteristic of Makapuu‐stage Koolau lavas, and they are offset from other Hawaiian shield lavas to high 87Sr/86Sr at a given 143Nd/144Nd. Consequently, a varying role for recycled plagioclase‐rich gabbro is inferred, in particular, lower amounts of the low 87Sr/86Sr component in Kahoolawe lavas. Also, lavas from Loa‐trend volcanoes, such as Kahoolawe, define trends ranging toward high 208Pb*/206Pb* and 87Sr/86Sr and low 143Nd/144Nd and 176Hf/177Hf. Such trends are consistent with variable amounts of recycled sediment sampled by Loa‐trend volcanoes, with the largest proportion in Makapuu‐stage Koolau lavas. Therefore the enriched component in the Hawaiian plume, the Koolau component, is recycled oceanic crust, which is heterogeneous because of varying proportions of sediment, basalt, and gabbro. Hawaiian shield‐stage lavas range widely in 87Sr/86Sr, 143Nd/144Nd, 176Hf/177Hf, and 206Pb/204Pb, but they have similar ratios of Sr/Nd, Nd/Hf, and Hf/Pb, each varying by a factor of <3 among the Hawaiian shields. This observation has important consequences. Namely, the similar Hf/Pb ratios are inconsistent with a two‐component (i.e., Kea and Koolau) mixing model for explaining the hyperbolic trend of 176Hf/177Hf versus 206Pb/204Pb defined by shield lavas. Such a model requires end‐members with very different Hf/Pb (a factor of 15 to 40), but this is not observed; therefore a third component must be involved. On the basis of trends of 208Pb*/206Pb* versus 87Sr/86Sr, 143Nd/144Nd, and 176Hf/177Hf, we infer that Loa and Kea trend shield lavas contain variable amounts of the Loihi source component.

Ancient carbonate sedimentary signature in the Hawaiian plume: Evidence from Mahukona volcano, Hawaii

“…The inset shows a blow up of Mahukona data, along with the 2 σ error bars. Data sources are as follows: Kahoolawe lavas, West et al [1987] and Huang et al [2005b]; Koolau (Makapuu), Roden et al [1984, 1994]; Mauna Kea, Lassiter et al [1996] and Bryce et al [2005]; Mauna Loa, Cohen et al [1996] and Kurz et al [1995]; Kauai, Mukhopadhyay et al [2003]; Loihi, Garcia et al [1993, 1995b, 1998]; Kohala, Hofmann et al [1987]; rejuvenated stage lavas, Stille et al [1983], Roden et al [1984], Chen and Frey [1985], Tatsumoto et al [1987], West et al [1987], Frey et al [2000], and Lassiter et al [2000]; EPR MORB, Niu et al [1999], Regelous et al [1999], and Castillo et al [2000].…”

“…For example, Abouchami et al [2005] showed that at a given 206 Pb/ 204 Pb, Loa trend lavas have higher 208 Pb/ 204 Pb, i.e., higher 208 Pb*/ 206 Pb*, than Kea trend lavas. Loa and Kea trend lavas form different trends in plots of 208 Pb*/ 206 Pb* versus Sr, Nd and Hf isotopic ratios [ Abouchami et al , 2005; Huang et al , 2005b]. Specifically, Kea trend lavas form horizontal trends in plots of 208 Pb*/ 206 Pb* versus Sr, Nd and Hf isotopic ratios.…”

“…In contrast, Loa trend lavas form steep trends in these plots, with Makapuu‐stage Koolau lavas (Oahu) defining the high 208 Pb*/ 206 Pb*‐ 87 Sr/ 86 Sr and low 143 Nd/ 144 Nd‐ 176 Hf/ 177 Hf end. Loa trend lavas, especially Makapuu‐stage Koolau lavas, also have high La/Nb and Sr/Nb, which Huang and Frey [2005] and Huang et al [2005b] proposed were derived from a recycled, ancient, carbonate‐rich, phosphate‐bearing sedimentary source component. In this paper, we present major and trace element abundances, and Sr‐Nd‐Hf‐Pb isotopic data for lavas from Mahukona volcano, a submarine Loa trend Hawaiian volcano between Hualalai and Kahoolawe (Figure 1).…”

Magma dynamics of ancient Mt. Etna inferred from clinopyroxene isotopic and trace element systematics

“…Etna through time. Lassiter et al, 2003;Stracke et al, 2003;Gaffney et al, 2004;Huang et al, 2005;Xu et al, 2007;Sims et al, 2008;Blichert-Toft and Albarède, 2009;Yamasaki et al, 2009;Garcia et al, 2010;Peate et al, 2010;Chekol et al, 2011;Salters et al, 2011;Viccaro et al, 2011); mantle components from Zindler and Hart, 1986;Salters and White, 1998;Workman et al, 2004;Stracke et al, 2005;Workman and Hart, 2005. Hafnium isotopic values for Italian sediments (Conticelli et al, 2002;Brems et al, 2013) are calculated from Nd isotopic data and both cases following the seawater array (SA) and the terrestrial array (TA) of Vervoort et al (2011) Trua et al (1998).…”

The fate of subducted oceanic crust: a mineral segregation model

“…Compared to MORB, OIB is usually more enriched in incompatible elements, such as U, Nb, Ta, Th, and K. Its isotopic composition is typically more radiogenic (with the exception Nd) and far more variable than MORB(e.g. Jacobsen and Wasserburg 1979;Sun 1980;Zindler and Hart 1986;Hofmann 1997;Allegre 2002;Frey et al 2002;Abouchami et al 2005;Frey et al 2005;Huang et al 2005;Sobolev et al 2007). The distinctively different geochemical characteristics of MORB and OIB led to the development of layered-mantle-convection models, which were once widely accepted by geochemists.…”