Petrogenesis of Davidson Seamount lavas and its implications for fossil spreading center and intraplate magmatism in the eastern Pacific

Abstract

[1] Seafloor spreading causes abundant magmatism along active ocean spreading centers, but the cause of magmatism along fossil spreading centers is enigmatic. Samples collected from Davidson Seamount, a typical volcanic ridge along an abandoned spreading center in the eastern Pacific, consist of an alkalic basalt to trachyte lava series; transitional basalts were sampled from another part of the abandoned axis, 20 km from the seamount. All samples experienced complex fractional crystallization prior to eruption, but they all share a common, compositionally heterogeneous mantle source. The parental magmas of the transitional basalts were produced from this source at higher degree of melting than those of the alkalic lava series. The composition of Davidson lavas overlaps with those of ridges along other fossil spreading centers and isolated near-and off-ridge seamounts in the eastern Pacific. Together they define a compositional continuum ranging from tholeiitic, normal mid-ocean ridge basalt (MORB)-like to alkalic, ocean island basalt (OIB)-like, similar to lavas that form linear island chains and ridges. We propose that this entire compositional spectrum of intraplate lavas that do not form linear volcanic chains in the eastern Pacific results from variations in the degree of partial melting of a common, compositionally heterogeneous mantle source. This source consists of more easily melted, geochemically enriched components of varying sizes and amounts embedded in a depleted lherzolitic matrix. Large degree of partial melting produces normal MORB-like melts represented by some near-ridge seamount lavas, whereas small degree of melting produces OIB-like fossil spreading center lavas. The small degree of partial melting beneath recently abandoned spreading centers results from either buoyancy-driven decompression melting of the hot lithospheric and asthenospheric mantle material beneath active spreading centers or rapid motion, with respect to the underlying asthenosphere, of abandoned spreading axes that are thickening over a fertile mantle. Mid-Tertiary volcanic rocks in coastal California, which are compositionally akin to intraplate lavas, are interpreted to be small degree partial melts of the same compositionally heterogeneous sub-Pacific mantle that has upwelled through windows in a subducted slab.

Insights into mantle composition and mantle melting beneath mid‐ocean ridges from postspreading volcanism on the fossil Galapagos Rise

“…Comparison of trace element and isotope compositions of postspreading lavas from the Galapagos Rise (data from this study) with lavas from other fossil spreading centers worldwide (data from Choe et al [2007], Choi et al [2008], Castillo et al [2010], and Haase et al [2011]). Postspreading lavas include some of the most extreme EMORB compositions (high Nb/Zr, La/Sm, and 87 Sr/ 86 Sr) found far from “hot spots.” Data for lavas from the EPR and Pacific near‐ridge seamounts are shown for comparison (data sources are the same as for Figure 4).…”

“…Lavas erupted at intratransform spreading segments in the Garrett Fracture Zone, interpreted to result from melting of mantle that recently underwent melt extraction beneath the adjacent spreading ridge, have trace element and isotope compositions that are more depleted than lavas from neighboring ridge segments [ Wendt et al , 1999]. Conversely, lavas believed to result from small degrees of mantle melting tend to have more enriched incompatible trace element and isotope compositions [e.g., Haase , 1996; Janney et al , 2000; Regelous et al , 2003; Hirano et al , 2006; Konter et al , 2009; Castillo et al , 2010]. Clinopyroxenes in residual abyssal peridotites tend to have more radiogenic Nd isotope compositions than those of lavas from the same section of ridge, consistent with preferential melting‐out of eclogite or pyroxenite with lower Sm/Nd and 143 Nd/ 144 Nd ratios during decompression melting [ Salters and Dick , 2002].…”

“…Fossil spreading ridges, formed when active spreading centers are abandoned, may continue to erupt magma for several million years after plate separation has ceased [ Batiza , 1977; Batiza et al , 1982; Batiza and Vanko , 1985; Davis et al , 2002; Clague et al , 2009; Haase et al , 2011]. Previous geochemical studies of the youngest lavas erupted at fossil spreading ridges have shown that many are EMORB, with highly enriched trace element and isotope compositions [ Batiza and Vanko , 1985; Bohrson and Reid , 1995; Choe et al , 2007; Choi et al , 2008; Clague et al , 2009; Castillo et al , 2010; Haase et al , 2011; Tian et al , 2011]. Postspreading lavas erupted at fossil ridges may therefore preserve the purest expression of the EMORB source among all oceanic basalts away from hot spots, and they apparently result from smaller degrees of mantle melting of the same “normal” mantle that melts beneath actively spreading ridges to produce MORB.…”

Timing of rifting in the southern Gulf of California and its conjugate margins: Insights from the plutonic record

“…2 and 3) sampled a mainly granitic and andesitic section here (Figs. There is clear geomorphic evidence of an inactive strike-slip fault with at least 30 km of right-lateral offset along the Cerralvo Island (southwest) side of the trough (Lonsdale and Kluesner, 2010;Kluesner, 2011), and we infer that the trough was a transtensional fault zone that, before the Alarcón Basin opened, linked Baja-North America spreading in the mouth of the gulf (i.e., the axes of the Los Frailes and Maria Magdalena Rises, predecessors of the East Pacifi c Rise; Castillo et al, 2010) to sites of oblique extension within the gulf. The intersecting NW-striking scarp is at the southern end of a major structural lineament, which Lyle and Ness (1991) named the Santa Cruz fault zone and speculated that it was an extinct intracontinental transform fault zone that after ca.…”

Untitled

“…Samples DANA 46a, ROCA 3J 4a, and ROCA 24J 33 were analyzed by X-ray fl uorescence spectrometry and by inductively coupled plasma-mass spectrometry at the GeoAnalytical Laboratory of Washington State University with the methods described in Castillo et al (2010).…”