DTM Attends the 2014 Geological Society of America Meeting in Vancouver

GSA 2014
Thursday, October 30, 2014 

Steven Shirey
Sunday, 19 October 2014 
"Looking at Ocean Islands From the Bottom: The Osmium Isotopic Compositions of Savai'i and Tubai'i (Samoa) Mantle Xenoliths"

Abstract: Ocean islands are one of the best probes of Earth’s deep mantle. Hawaii, Earth's major hotspot, has been a key scientific laboratory for the productive career of Fred Frey of MIT. In early important work, Frey used its basalts to understand mantle source enrichment and melting, inverting trace elements through melting models [1]. Even earlier, Frey analyzed Hawaiian mantle xenoliths for the rare earth element partitioning between clinopyroxene and garnet and its relationship to Hawaiian lavas [2] showing how valuable xenoliths are in ocean island geochemical studies.

A suite of 25 spinel peridotites from the islands of Savai’i and Tubuai’i hosted in alkali basalts were analyzed for their osmium isotopic compositions, and major and trace elements. The goal is to examine their mantle source compositions for domains that record ancient depletion events that were recycled along with a component of continental and oceanic crust present in the host basalts [3]. In Samoan xenoliths, Re abundances are typically low (0.0003 to 0.035 ppb), but Re abundances are higher in the Tubaui xenoliths (0.030 to 0.354 ppb). Os abundances are also typically low (<1ppb) but range up to values exceeding 5 ppb in both suites. Osmium isotopic compositions are typical of the convecting mantle (187Os/188Os = 0.1240 to 0.1304), but several xenoliths from Samoa have isotopic compositions <0.120 while just one xenolith from Tubuai'i has such low isotopic compositions. The xenoliths are thought to originate in the oceanic lithospheric mantle. The 187Os/188Os therefore records preexisting heterogeneities in the upper mantle “locked into” the lithosphere at the mid-ocean ridge, and that may have been modified by the metasomatic processes associated with the upwelling plume [4]. Carbonatite signatures are clearly recorded in the trace element signatures of the xenoliths, and the carbonatite metasomatism may have modified the Os-isotopic composition of the xenoliths so that they record slightly more radiogenic values. The unradiogenic Os component in the xenoliths may relate to subcontinental lithospheric mantle or residues of ancient depletion events that are preserved in the convecting mantle.

[1] Clague and Frey, 1982, J Pet 23, 447. [2] Reid and Frey, 1971, JGR 76, 1184. [3] Jackson et al., 2007, Nature 448, 684. [4] Hauri et al., 1993, Nature 365, 221.

Hanika Rizo
Monday, 20 October 2014
"Reconciling 182W/184W Variability in the Archean Mantle with W Partition Coefficients for Metal-Silicate Differentiation"

Abstract: Short-lived isotope systems, such as 182Hf-182W (T½ = 8.9 Ma), provide important information on the earliest phases of planetary differentiation and evolution. Because of the short half-life of 182Hf, variations in 182W can be only produced during the first ~ 50 Ma of Earth’s history. Tungsten is siderophile while Hf is lithophile, so metal-silicate segregation is commonly invoked to explain the Hf/W changes needed to create 182W/184W variability. W is also substantially more incompatible during silicate crystal-liquid fractionation, so magma ocean crystallization also could lead to variations in Hf/W. Finally, the182W/184W of the Earth, as well as the highly siderophile element (HSE) abundances in the mantle, were also modified to an unknown extent as a result of addition of late accreted materials.
In this study, we present high-precision 182W data for mafic and ultramafic samples from the Isua supracrustal belt (southwest Greenland), with ages that range between 3.3 Ga and 3.8 Ga. The Isua samples show enrichments in 182W of up to 15 ppm, relative to terrestrial standards and modern rocks, in general agreement with the previous results [1]. These excesses were previously interpreted to reflect the incomplete mixing of late-accreted materials into the mantle sources of these rocks during the period between 4.5 and 3.8 Ga. However, the HSE abundances in the studied samples suggest that their sources had HSE abundances similar to those in the estimates for the modern mantle. The “normal” HSE abundances of the Isua rocks suggest that their mantle source had already received the full complement of the late accretion component of HSE, and hence W. In this case, the 182W/184W ratio of the source of Isua’s samples before late accretion, assuming that the late accretion component represent ~0.5% of Earth’s mass, is calculated to have been between +23 to +32 ppm.
If the Hf/W fractionation was caused by silicate-metal differentiation, excesses of +23 to +32 ppm in 182W require a partition coefficient for W between 36 and 37, assuming that the core formed in one stage at ~ 30 Ma after the Solar system formation. These partition coefficients are in the range of those experimentally determined (between 30 and 50) which include 2% of S at the end of the accretion [2].
1. Willbold et al., 2011, Nature. 2. Wade et al., 2012, Geochim. Cosmochim. Acta.

Lara Wagner
Tuesday, 21 October 2014
"The Role of Subducted Ridges in the Evolution of Flat Slabs"

Abstract: Flat slab subduction has long been considered as a potential contributing factor in the formation of inland basement cored uplifts, volcanic sweeps and gaps, and widespread ignimbrite volcanism. However, the processes responsible for the formation of flat slabs are poorly constrained. This is of particular importance when evaluating the putative Farallon flat slab, which would have been significantly larger in scale than any flat slab observed today. We present new constraints on the role of ridge subduction in the formation of flat slabs from recent seismic data collected in southern Peru and northernmost Bolivia as part of the PULSE, CAUGHT, and PERUSE deployments. We find several lines of evidence that strongly suggest that the ridge is a necessary, if not sufficient, condition for the formation and stability of the flat slab in southern Peru. Specifically, we find that the flat slab is shallowest directly adjacent to the ridge, and that north of the southward migrating ridge, the flat slab that had formed earlier when the ridge was present is now resuming a normal subduction geometry due to sagging and tearing of the plate. Our results suggest that in Peru, the contributing factors in the formation of the flat slab include 1) trench retreat and rapid overriding plate motion, 2) suction forces between established flat slab and the base of the continent and 3) ridge subduction. However, when the ridge is removed, the first two contributing factors are not sufficient to maintain the flat slab geometry. This has important implications for the tectonic evolution of the Western United States where the buoyant bathymetric feature was not a ridge, but rather a discrete plateau.