Appendix A. Whole-rock major, trace and rare earth element analyses of the Ulubey volcanics, Appendix B. Analytical standards and detection limits for analyses

Collisional and post-collisional volcanic rocks in the Ulubey (Ordu) area at the western edge of the Eastern Pontide Tertiary Volcanic Province (EPTVP) in NE Turkey are divided into four suites; Middle Eocene (49.4-44.6 Ma) aged Andesite-Trachyandesite (AT), Trachyandesite-Trachydacite-Rhyolite (TTR), Trachydacite-Dacite (TD) suites, and Middle Miocene (15.1 Ma) aged Trachybasalt (TB) suite. Local stratigraphy in the Ulubey area starts with shallow marine environment sediments of the Paleocene-Eocene time and then continues extensively with sub-aerial andesitic to rhyolitic and rare basaltic volcanism during Eocene and Miocene time, respectively. Petrographically, the volcanic rocks are composed primarily of andesites/trachyandesites, with minor trachydacites/rhyolites, basalts/trachybasalts and pyroclastics, and show porphyric, hyalo-microlitic porphyric and rarely glomeroporphyric, intersertal, intergranular, fluidal and sieve textures. The Ulubey (Ordu) volcanic rocks indicate magma evolution from tholeiitic-alkaline to calc-alkaline with medium-K contents. Primitive mantle normalized trace element and chondrite normalized rare earth element (REE) patterns show that the volcanic rocks have moderate light rare earth element (LREE)/heavy rare earth element (HREE) ratios relative to E-Type MORB and depletion in Nb, Ta and Ti. High Th/Yb ratios indicate parental magma(s) derived from an enriched source formed by mixing of slab and asthenospheric melts previously modified by fluids and sediments from a subduction zone. All of the volcanic rocks share similar incompatible element ratios (e.g., La/Sm, Zr/Nb, La/Nb) and chondrite-normalized REE patterns, indicating that the basic to acidic rocks originated from the same source. The volcanic rocks were produced by the slab dehydration-induced melting of an existing metasomatized mantle source, and the fluids from the slab dehydration introduced significant large ion lithophile element (LILE) and LREE to the source, masking its inherent HFSE-enriched characteristics. The initial 87Sr/86Sr (0.7044-0.7050) and eNd (-0.3 to +3.4) ratios of the volcanics suggest that they originated from an enriched lithospheric mantle source with low Sm/Nd ratios. Integration of the geochemical, petrological and isotopical with regional and local geological data suggest that the Tertiary volcanic rocks from the Ulubey (Ordu) area were derived from an enriched mantle, which had been previously metasomatized by fluids derived from subducted slab during Eocene to Miocene in collisional and post-collisional extension-related geodynamic setting following Late Mesozoic continental collision between the Eurasian plate and the Tauride-Anatolide platform.

(Table 1) Composition of DSDP Hole 37-334 clinopyroxenes

Recent studies of abyssal peridotites (Johnson et al., 1990, doi:10.1029/JB095iB03p02661), mid-ocean-ridge basalts (MORBs) (McKenzie, 1985, doi:10.1016/0012-821X(85)90001-9) and their entrained melt inclusions (Sobolev and Shimizu, 1993, doi:10.1038/363151a0; Humler and Whitechurch, 1988, doi:10.1016/0012-821X(88)90055-6) have shown that fractional melting of the upwelling sub-oceanic mantle produces magmas with a much wider range of compositions than erupted MORBs. In particular, it seems that strongly depleted primary magmas are routinely produced by melting beneath ridges (Johnson et al., 1990, doi:10.1029/JB095iB03p02661). The absence of strongly depleted melts as erupted lavas prompts the question of how long such magmas survive beneath ridges, before their distinctive compositions are concealed by mixing with more enriched magmas. Here we report mineral compositions from a unique suite of oceanic cumulates recovered from DSDP Site 334 (Aumento et al., doi:10.2973/dsdp.proc.37.1977), which indicate that the rocks crystallized from basaltic liquids that were strongly depleted in Na, Ti, Zr, Y, Sr and rare-earth elements relative to any erupted MORB. It thus appears that the magmatic plumbing system beneath the Mid-Atlantic Ridge permitted strongly depleted magmas to accumulate in a magma chamber and remain sufficiently isolated to produce cumulate rocks. Even so, spatial heterogeneity in the compositions of high-calcium pyroxenes suggests that in the later stages of solidification these rocks reacted with infiltrating enriched basaltic liquids.

(Table 1) Chemical and isotopic analyses of Rajmahal Traps and Ninetyeast Ridge volcanics

The hypothesis that hotspots are the sources of many continental flood basalts is evaluated geochemically for the proposed Rajmahal Traps-Ninetyeast Ridge-Kerguelen hotspot system. It appears that the Kerguelen hotspot did not directly feed Rajmahal magmas, although it may have provided a source of heat for Rajmahal activity.

(Table 1) Sr, Nd, and Pb isotopic composition of ODP Leg 129 basalts

Basaltic rocks recovered from three drill sites in the western Pacific during Ocean Drilling Program Leg 129 have fairly distinct Sr, Nd, and Pb isotopic compositions. The Cretaceous alkali olivine dolerites from Site 800 in the northern part of Pigafetta Basin have fairly low 87Sr/86Sri (0.70292-0.70320) and 143Nd/144Ndi (0.51277-0.51281) and high present-day Pb isotopic ratios (206Pb/204Pb = 20.53-21.45; 207Pb/204Pb = 15.70-15.77; 208Pb/204Pb = 40.02-40.68). The Middle Jurassic tholeiites from Site 801 in the southern part of the basin have low 87Sr/86Sri (0.70237-0.70248), high 143Nd/144Ndi (0.51298-0.51322), and moderate present-day Pb isotopic ratios (206Pb/204Pb = 18.20-19.12; 207Pb/204Pb = 15.47-15.60; 208Pb/204Pb = 37.56-38.18); isotopic compositions of the alkali olivine basalts overlying the tholeiites fall between those of the tholeiites and Site 800 dolerites. The Cretaceous tholeiites from Site 802 in the East Mariana Basin have high 87Sr/86Sri (0.70360-0.70372), fairly low 143Nd/144Ndi (0.51277-0.51280), and fairly low and homogeneous present-day Pb isotopic ratios (206Pb/204Pb = 18.37-18.39; 207Pb/204Pb = 15.49-15.51; 208Pb/204Pb = 38.34-38.39). Isotopic compositions of Site 801 tholeiites are indistinguishable from those of modern mid-ocean ridge basalts, consistent with the proposal that these tholeiites are a part of the oldest Pacific crust. The diverse isotopic compositions of the younger basalts appear to be the result of Jurassic Pacific plate migration over the geologically anomalous south-central Pacific region, wherein they acquired their distinct isotopic compositions. The anomalous region was volcanically more active during the Cretaceous than at present.

The Cenomanian/Turronian Boundary Event at ODP Hole 103-641A

Drilling at ODP Site 641 (on the western margin of Galicia Bank, off northwestern Spain) revealed a thin, but pronounced, interval of black shale and gray-green claystone. Our high-resolution study combines the sedimentology, micropaleontology (palynomorphs and others), organic and inorganic geochemistry, and isotopic values of this layer to demonstrate the distinct nature of the sediment and prove that the sequence represents the local sedimentary expression of the global Cenomanian/Turonian Oceanic Anoxic Event (OAE) of Schlanger and Jenkyns (1976), Arthur and Schlanger (1979), and Jenkyns (1980), also called the Cenomanian/Turonian Boundary Event (CTBE). The most striking evidence is that the strong positive d13C excursion characterizing the CTBE sequences in shallow areas can be traced into a pronounced deep-sea expression, thus providing a good stratigraphic marker for the CTBE in various paleosettings. The isotopic excursion at Site 641 coincides with an extremely enriched trace metal content, with values that were previously unknown for the Cretaceous Atlantic. Similar to other CTBE occurrences, the organic carbon content is high (up to 11%) and the organic matter is of dominantly marine origin (kerogen type II). The bulk mineralogy of the CTBE sediments does not differ significantly from the general trend of Cretaceous North Atlantic sediments (dominance of smectite and zeolite with minor amounts of illite and scattered palygorskite, kaolinite, and chlorite); thus, no evidence for either increased volcanic activity nor a drastic climatic change in the borderlands was found. Results from Site 641 are compared with the CTBE section found at Site 398, DSDP Leg 47B (Vigo Seamount at the southern end of the Galicia Bank).

Sediment geochemistry of ODP Leg 127 sites

The relative effects of paleoceanographic and paleogeographic variations, sediment lithology, and diagenetic processes on the final preserved chemistry of Japan Sea sediments are evaluated by investigating the rare earth element (REE), major element, and trace element concentrations in 59 squeeze-cake whole-round and 27 physical-property sample residues from Sites 794, 795, and 797, cored during ODP Leg 127. The most important variation in sedimentary chemical composition is the increase in SiO2 concentration through the Pliocene diatomaceous sequences, which dilutes most other major and trace element components by various degrees. This biogenic input is largest at Site 794 (Yamato Basin), moderately developed at Site 797 (Yamato Basin), and of only minor importance at Site 795 (Japan Basin), potentially reflecting basinal contrasts in productivity with the Yamato Basin recording greater biogenic input than the Japan Basin and with the easternmost sequence of Site 794 lying beneath the most productive waters. There are few systematic changes in solid-phase chemistry resulting from the opal-A/opal-CT or opal-CT/quartz silica phase transformations. Most major and trace element concentrations are controlled by the aluminosilicate fraction of the sediment, although the effects of diagenetic silica phases and manganese carbonates are of localized importance. REE total abundances (Sum REE) in the Japan Sea are strongly dependent upon the paleoceanographic position of a given site with respect to terrigenous and biogenic sources. REE concentrations at Site 794 overall correspond well to aluminosilicate chemical indices and are strongly diluted by SiO2 within the upper Miocene-Pliocene diatomaceous sequence. Eu/Eu* values at Site 794 reach a maximum through the diatomaceous interval as well, most likely suggesting an association of Eu/Eu* with the siliceous component, or reflecting slight incorporation of a detrital feldspar phase. SumREE at Site 795 also is affiliated strongly with aluminosilicate phases and yet is diluted only slightly by siliceous input. At Site 797, SumREE is not as clearly associated with the aluminosilicate fraction, is correlated moderately to siliceous input, and may be sporadically influenced by detrital heavy minerals originating from the nearby rifted continental fragment composing the Yamato Rise. Ce/Ce* profiles at all three sites increase essentially monotonically with depth and record progressive diagenetic LREE fractionation. The observed Ce/Ce* increases are not responding to changes in the paleoceanographic oxygenation state of the overlying water, as there is no independent evidence to suggest the proper oceanographic conditions. Ce/Ce* correlates slightly better with depth than with age at the two Yamato Basin sites. The downhole increase in Ce/Ce* at Sites 794 and 797 is a passive response to the diagenetic transfer of LREE (except Ce) from sediment to interstitial water. At Site 795, the overall lack of correlation between Ce/Ce* and Lan/Ybn suggests that other processes mask the diagenetic behavior of all LREEs. First-order calculations of the Ce budget in Japan Sea waters and sediment indicate that ~20% of the excess Ce adsorbed by settling particles is recycled within the water column and that an additional ~38% is recycled at or near the seafloor. Thus, because the remaining excess Ce is only ~10% of the total Ce, there is not a large source of Ce to the deeply buried sediment, further suggesting that the downhole increase in Ce/Ce* is a passive response to diagenetic behavior of the other LREEs. The REE chemistry of Japan Sea sediment therefore predicts successive downhole addition of LREEs to deeply buried interstitial waters.

Age model and Nd isotope ratios of DSDO Hole 43-386 and ODP Hole 210-1276A

Modern thermohaline circulation plays a role in latitudinal heat transport and in deep-ocean ventilation, yet ocean circulation may have functioned differently during past periods of extreme warmth, such as the Cretaceous. The Late Cretaceous (100-65 Ma) was an important period in the evolution of the North Atlantic Ocean, characterized by opening ocean gateways, long-term climatic cooling and the cessation of intermittent periods of anoxia (oceanic anoxic events, OAEs). However, how these phenomena relate to deep-water circulation is unclear. We use a proxy for deep-water mass composition (neodymium isotopes; e-Nd) to show that, at North Atlantic ODP Site 1276, deep waters shifted in the early Campanian (~78-83 Ma) from e-Nd values of ~-7 to values of ~-9, consistent with a change in the style of deep-ocean circulation but >10 Myr after a change in bottom water oxygenation conditions. A similar, but more poorly dated, trend exists in e-Nd data from DSDP Site 386. The Campanian e-Nd transition observed in the North Atlantic records is also seen in the South Atlantic and proto-Indian Ocean, implying a widespread and synchronous change in deep-ocean circulation. Although a unique explanation does not exist for the change at present, we favor an interpretation that invokes Late Cretaceous climatic cooling as a driver for the formation of Southern Component Water, which flowed northward from the Southern Ocean and into the North Atlantic and proto-Indian Oceans.

Major and trace element analyses of ODP Leg 120 volcanic rocks (Table 1)

Widespread Lower Cretaceous magmatism occurred along the Indian-Australian/Antarctic margins, and in the juvenile Indian Ocean, during the rifting of eastern Gondwana. The formation of this magmatic province probably began around 120-130 Ma with the eruption of basalts on the Naturaliste Plateau and at Bunbury, western Australia. On the northeast margin of India, activity began around 117 Ma with the Rajmahal continental basalts and associated lamprophyre intrusions. The formation of the Kerguelen Plateau in the Indian Ocean began no later than 114 Ma. Ultramafic lamprophyres (alnoites) were emplaced in the Prince Charles Mountains near the Antarctic continental margin at ~ 110 Ma. These events are considered to be related to a major mantle plume, the remnant of which is situated beneath the region of Kerguelen and Heard islands at the present day. Geochemical data are presented for each of these volcanic suites and are indicative of complex interactions between asthenosphere-derived magmas and the continental lithosphere. Kerguelen Plateau basalts have Sr and Nd isotopic compositions lying outside the field for Indian Ocean mid-ocean ridge basalts (MORB) but, with the exception of Site 738 at the southern end of the plateau, within the range of more recent hotspot basalts from Kerguelen and Heard Islands. However, a number of the plateau tholeiites are characterized by lower 206Pb/204Pb ratios than are basalts from Kerguelen Island, and many also have anomalously high La/Nb ratios. These features suggest that the source of the Kerguelen Plateau basalts suffered contamination by components derived from the Gondwana continental lithosphere. An extreme expression of this lithospheric signature is shown by a tholeiite from Site 738, suggesting that the southernmost part of the Kerguelen Plateau may be underlain by continental crust. The Rajmahal tholeiites mostly fall into two distinct geochemical groups. Some Group I tholeiites have Sr and Nd isotopic compositions and incompatible element abundances, similar to Kerguelen Plateau tholeiites from Sites 749 and 750, indicating that the Kerguelen-Heard mantle plume may have directly furnished Rajmahal volcanism. However, their elevated 207Pb/204Pb ratios indicate that these magmas did not totally escape contamination by continental lithosphere. In contrast to the Group I tholeiites, significant contamination is suggested for Group II Rajmahal tholeiites, on the basis of incompatible element abundances and isotopic compositions. The Naturaliste Plateau and the Bunbury Basalt samples show varying degrees of enrichment in incompatible elements over normal MORB. The Naturaliste Plateau samples (and Bunbury Basalt) have high La/Nb ratios, a feature not inconsistent with the notion that the plateau may consist of stretched continental lithosphere, near the ocean-continent divide.

Geochemistry of the oldest Atlantic oceanic crust

Controversy has surrounded the issue of whether mantle plume activity was responsible for Pangaean continental rifting and massive flood volcanism (resulting in the Central Atlantic Magmatic Province or CAMP, emplaced around 200 Ma) preceding the opening of the central Atlantic Ocean in the Early Mesozoic. Our new Sr-Nd-Pb isotopic and trace element data for the oldest basalts sampled from central Atlantic oceanic crust by deep-sea drilling show that oceanic crust generated from about 160 to 120 Ma displays clear isotopic and chemical signals of plume contamination (e.g., 87Sr/86Sr(i) = 0.7032-0.7036, epsilonNd(t) =+6.2 to +8.2, incompatible element patterns with positive Nb anomalies), but these signals are muted or absent in crust generated between 120 and 80 Ma, which resembles young Atlantic normal mid-ocean ridge basalt. The plume-affected pre-120 Ma Atlantic crustal basalts are isotopically similar to lavas from the Ontong Java Plateau, and may represent one isotopic end-member for CAMP basalts. The strongest plume signature is displayed near the center of CAMP magmatism but the hotspots presently located nearest this location in the mantle reference frame do not appear to be older than latest Cretaceous and are isotopically distinct from the oldest Atlantic crust. The evidence for widespread plume contamination of the nascent Atlantic upper mantle, combined with a lack of evidence for a long-lived volcanic chain associated with this plume, leads us to propose that the enriched signature of early Atlantic crust and possibly the eruption of the CAMP were caused by a relatively short-lived, but large volume plume feature that was not rooted at a mantle boundary layer. Such a phenomenon has been predicted by recent numerical models of mantle circulation.