Abstract

We report a comprehensive study of He isotope ( 3He/ 4He) variations on well-characterized lavas from La Palma, the current locus of activity of the Canary hotspot. Cogenetic olivine (OL) and clinopyroxene (CPX) phenocrysts from 11 basaltic lava flows representing all stages of the island’s evolution were analyzed by crushing in vacuo. Additionally, the sample set is supplemented by a CO 2-rich bubbling cold spring in Taburiente caldera. Phenocryst 3He/ 4He ratios vary between 6.3 to 8.9R A; both extremes occur in the 2 to 0.6 Ma shield-building Taburiente lavas. Historic flows vary between 7.0 to 7.8R A, whereas phenocrysts from a single submarine basement sill (4.0–2.9 Ma) have 3He/ 4He ratios of 8.3R A (OL) and 8.4R A (CPX). He isotopic equilibrium characterizes all the phenocryst pairs except one Taburiente lava. The absence of a correlation between 3He/ 4He and [He] contrasts with other ocean island basalts, e.g., Heard Island (Hilton et al., 1995) and, together with equilibrium in phenocrystic 3He/ 4He, suggests that the La Palma sample suite has 3He/ 4He ratios that have suffered minimal crustal contamination. The cold spring (geothermal) sample gives a 3He/ 4He ratio of 9.5R A, which is higher than any of the phenocrysts and in excellent agreement with a previously reported value for Taburiente caldera (Perez et al., 1996). The geothermal 3He/ 4He values indicate that La Palma is the first HIMU-like ocean island (where HIMU implies a distinctive mantle source with high time-integrated 238U/ 204Pb (μ) ratio) with reported 3He/ 4He ratios higher than the canonical value (8 ± 1R A) characteristic of depleted mid-ocean ridge basalt (MORB) mantle. We present two simple mixing scenarios that are compatible with the He–Pb isotope systematics of La Palma: a) a high- 3He mantle plume, extensively but variably depleted in its original He content, mixes with a HIMU source characterized by both radiogenic He and lead isotope compositions; and b) mixing between a composite plume, with high 3He/ 4He and radiogenic lead, and depleted mid-ocean ridge basalt mantle. The latter model is favored as it appears to account more readily for the mid-ocean ridge MORB-like He isotope characteristics of the lavas. Addition of radiogenic He—most likely produced in mantle melts frozen into the sub-Canarian mantle lithosphere—to resultant mixtures may subsequently occur in either model. Such a process can explain the absence of correlations between He and Pb isotope ratios in the lavas, and the lower 3He/ 4He values of the phenocrysts with respect to the geothermal fluids. We find no need to invoke a radiogenic growth/diffusion model (Hanyu and Kaneoka, 1998) to explain the He–Pb isotope relationships for La Palma. These mixing models may also be applicable to other HIMU islands: however, any high- 3He plume signature of St. Helena and Mangaia may be obscured due to the high proportion of the HIMU component at these localities and/or by addition of radiogenic He.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call