Isotope ratios and trace element concentrations of basalts from DSDP Leg 15 sites (Appendix 1)

Formation of the Cretaceous Caribbean plateau, including the komatiites of Gorgona, has been linked to the currently active Galápagos hotspot. We use Hf-Nd isotopes and trace element data to characterise both the Caribbean plateau and the Galápagos hotspot, and to investigate the relationship between them. Four geochemical components are identified in the Galápagos mantle plume: two 'enriched' components with epsilon-Hf and epsilon-Nd similar to enriched components observed in other mantle plumes, one moderately enriched component with high Nb/Y, and a fourth component which most likely represents depleted MORB source mantle. The Caribbean plateau basalt data form a linear array in Hf-Nd isotope space, consistent with mixing between two mantle components. Combined Hf-Nd-Pb-Sr-He isotope and trace element data from this study and the literature suggest that the more enriched Caribbean end member corresponds to one or both of the enriched components identified on Galápagos. Likewise, the depleted end member of the array is geochemically indistinguishable from MORB and corresponds to the depleted component of the Galápagos system. Enriched basalts from Gorgona partially overlap with the Caribbean plateau array in epsilon-Hf vs. epsilon-Nd, whereas depleted basalts, picrites and komatiites from Gorgona have a high epsilon-Hf for a given epsilon-Nd, defining a high-epsilon-Hf depleted end member that is not observed elsewhere within the Caribbean plateau sequences. This component is similar, however, in terms of Hf-Nd-Pb-He isotopes and trace elements to the depleted plume component recognised in basalts from Iceland and along the Reykjanes Ridge. We suggest that the Caribbean plateau represents the initial outpourings of the ancestral Galápagos plume. Absence of a moderately enriched, high Nb/Y component in the older Caribbean plateau (but found today on the island of Floreana) is either due to changing source compositions of the plume over its 90 Ma history, or is an artifact of limited sampling. The high-epsilon-Hf depleted component sampled by the Gorgona komatiites and depleted basalts is unique to Gorgona and is not found in the Caribbean plateau. This may be an indication of the scale of heterogeneity of the Caribbean plateau system; alternatively Gorgona may represent a separate oceanic plateau derived from a completely different Pacific plume, such as the Sala y Gomez.

(Table S1) Age determination of Emperor Seamounts basement

The Hawaiian-Emperor hotspot track has a prominent bend, which has served as the basis for the theory that the Hawaiian hotspot, fixed in the deep mantle, traced a change in plate motion. However, paleomagnetic and radiometric age data from samples recovered by ocean drilling define an age-progressive paleolatitude history, indicating that the Emperor Seamount trend was principally formed by the rapid motion (over 40 millimeters per year) of the Hawaiian hotspot plume during Late Cretaceous to early-Tertiary times (81 to 47 million years ago). Evidence for motion of the Hawaiian plume affects models of mantle convection and plate tectonics, changing our understanding of terrestrial dynamics.

(Table 1) Isotopic and trace-element ratios of Emperor Seamount basalts

When a mantle plume interacts with a mid-ocean ridge, both are noticeably affected. The mid-ocean ridge can display anomalously shallow bathymetry, excess volcanism, thickened crust, asymmetric sea-floor spreading and a plume component in the composition of the ridge basalts (Schilling, 1973, doi:10.1038/242565a0; Verma et al., 1983, doi:10.1038/306654a0; Ito and Lin, 1995, doi:10.1130/0091-7613(1995)023<0657:OSCHIC>2.3.CO;2; Müller et al., 1998, doi:10.1038/24850). The hotspot-related volcanism can be drawn closer to the ridge, and its geochemical composition can also be affected (Ito and Lin, 1995, doi:10.1130/0091-7613(1995)023<0657:OSCHIC>2.3.CO;2; White et al., 1993, doi:10.1029/93JB02018; Kincaid et al., 1995, doi:10.1038/376758a0; Kingsley and Schilling, 1998, doi:10.1029/98JB01496 ). Here we present Sr-Nd-Pb isotopic analyses of samples from the next-to-oldest seamount in the Hawaiian hotspot track, the Detroit seamount at 51° N, which show that, 81 Myr ago, the Hawaiian hotspot produced volcanism with an isotopic signature indistinguishable from mid-ocean ridge basalt. This composition is unprecedented in the known volcanism from the Hawaiian hotspot, but is consistent with the interpretation from plate reconstructions (Mammerickx and Sharman, 1988, doi:10.1029/JB093iB04p03009) that the hotspot was located close to a mid-ocean ridge about 80 Myr ago. As the rising mantle plume encountered the hot, low-viscosity asthenosphere and hot, thin lithosphere near the spreading centre, it appears to have entrained enough of the isotopically depleted upper mantle to overwhelm the chemical characteristics of the plume itself. The Hawaiian hotspot thus joins the growing list of hotspots that have interacted with a rift early in their history.

Dive depth profile and at surface behaviour data of emperor penguins from Pointe Géologie, Adélie Land, Antarctica, from expedition DDU 2005

Adult male and female emperor penguins (Aptenodytes forsteri) were fitted with satellite transmitters at Pointe-Géologie (Adélie Land), Dumont d'Urville Sea coast, in November 2005. Nine of 30 data sets were selected for analyses to investigate the penguins' diving behaviour at high resolution (doi:10.1594/PANGAEA.633708, doi:10.1594/PANGAEA.633709, doi:10.1594/PANGAEA.633710, doi:10.1594/PANGAEA.633711). The profiles are in synchrony with foraging trips of the birds during austral spring (doi:10.1594/PANGAEA.472171, doi:10.1594/PANGAEA.472173, doi:10.1594/PANGAEA.472164, doi:10.1594/PANGAEA.472160, doi:10.1594/PANGAEA.472161). Corresponding high resolution winter data (n = 5; archived elsewhere) were provided by A. Ancel, Centre d'Ecologie et Physiologie Energétiques, CNRS, Strasbourg, France. Air-breathing divers tend to increase their overall dive duration with increasing dive depth. In most penguin species, this occurs due to increasing transit (descent and ascent) durations but also because the duration of the bottom phase of the dive increases with increasing depth. We interpreted the efficiency with which emperor penguins can exploit different diving depths by analysing dive depth profile data of nine birds studied during the early and late chick-rearing period in Adélie Land, Antarctica. Another eight datasets of dive depth and duration frequency recordings (doi:10.1594/PANGAEA.472150, doi:10.1594/PANGAEA.472152, doi:10.1594/PANGAEA.472154, doi:10.1594/PANGAEA.472155, doi:10.1594/PANGAEA.472142, doi:10.1594/PANGAEA.472144, doi:10.1594/PANGAEA.472146, doi:10.1594/PANGAEA.472147), which backup the analysed high resolution depth profile data, and dive depth and duration frequency recordings of another bird (doi:10.1594/PANGAEA.472156, doi:10.1594/PANGAEA.472148) did not match the requirement of high resolution for analyses. Eleven additional data sets provide information on the overall foraging distribution of emperor penguins during the period analysed (doi:10.1594/PANGAEA.472157, doi:10.1594/PANGAEA.472158, doi:10.1594/PANGAEA.472162, doi:10.1594/PANGAEA.472163, doi:10.1594/PANGAEA.472166, doi:10.1594/PANGAEA.472167, doi:10.1594/PANGAEA.472168, doi:10.1594/PANGAEA.472170, doi:10.1594/PANGAEA.472172, doi:10.1594/PANGAEA.472174, doi:10.1594/PANGAEA.472175).