Variations of 3He/4He ratios in geothermal fluids offer an opportunity to distinguish mantle and crust-derived components in volatiles emanating from island arc volcanoes. Previous works reported settings where the very uniform MORB mantle helium signature is modified by admixture of helium degassed from subducted continental crust (Hilton et al., 1991), and derived from magma ageing and assimilation of lower crustal material (Hilton et al., 1993). A further possibility might be admixture of radiogenic helium degassed from subducting coarse grained terrigenous sediments. The Lesser Antilles islands arc is a candidate to test such a possibility. The southern part of its trench is filled with sediments of proterozoic age from the Brazilian shield (forming the Barbados accretionary prism). The northern part of it culminates in the more than 6000 m deep Puerto Rico trench.
We collected water and gas samples from 19 geothermal sites on the islands St.Lucia, Martinique, Dominica, and Guadeloupe, in the Lesser Antilles arc, and determined isotope and element abundance ratios of He, Ne, and Ar.
Neon and Argon: All samples have 40Ar/36Ar ratios up to 8% higher than in air and 36Ar/38Ar and 20Ne/22Ne ratios not significantly (<1.5 %) differing from atmospheric ratios. Assuming (40Ar/36Ar)MORB = 28000, MORB-like neon and argon components only contribute a few percent (< 8%) to the total neon and argon inventory, the major fraction being dissolved and admixed air. Assuming solubility data of Weiss (1970, 1971) and reasonable temperatures and salinities for water at the recharge and discharge areas, we expect 20Ne/36Ar ratios around 0.11 - 0.19 and 4He/20Ne ratios around 0.25 - 0.29 for water samples, and 0.04 - 0.19 (20Ne/36Ar) and 0.08 - 0.29 (4He/20Ne) for gas samples respectively. Indeed, one third of the 20Ne/36Ar ratios fall within the expected range, nevertheless, all other samples have values close to, or even higher than atmospheric. For the former cases, variations of 20Ne/36Ar ratios might be attributed to admixture of air during sampling or/and by certain geologic processes. At the Soufrière of St. Lucia we observed an example for such a process, where a reaction with atmospheric gases is even visible. The 20Ne/36Ar ratio in samples from a "black" pool (sulfide precipitate) have 20Ne/36Ar = 0.21, in agreement with the solubility data, whereas in samples from a "white" sulfatic pool the 20Ne/36Ar ratio is 0.32 - 0.42. The 4He/20Ne ratios of the same samples are high in the black pool (175) and low in the white pool (6 to 38) again suggesting an admixture of (non-dissolved) air. 20Ne/36Ar ratios higher than expected for air-saturated water do not necessarily reflect a simple admixture of air. In Otsan, St. Lucia, we collect gas with an 20Ne/36Ar ratio 4 times that of atmosphere. Nagao and co-workers (1979) observed a similar enrichment of light noble gases (and light isotopes) in gas samples collected at the Nigorikawa geothermal field, Japan. They attribute the enrichment to atmospheric noble gases, originally dissolved in groundwater, which were transferred into ascending CO2 bubbles by Rayleigh destillation. In some samples we also found Ne, Ar, and He concentrations lower than expected for water saturation. For these samples a complex scenario of mixing of at least two water reservoirs with different temperature and degassing history is required.
Helium: In water samples the 4He/20Ne ratios are near the atmospheric value or slightly above it (up to 6 times). In gas samples the 4He/20Ne ratios average at 230 indicating a non-atmospheric helium component. The latter can be described by its 3He/4He ratio. On each individual island, 3He/4He ratios range from 1.2 Ra ( Ra = 3He/4He in air = 1.4*10-6 ) to a maximum value. The maximum value decreases from typical MORB mantle helium signature in the north (Guadeloupe: 8.5 - 8 Ra; Dominica: 7.5 - 8.2 Ra) to more radiogenic values in the south (Martinique: 6.0 - 7.0 Ra; St. Lucia: < 5.2 Ra). The significance of the helium ratio as a source feature of the magmagenesis beneath the magmatic arc or as a signal of the life time of a magma reservoir or as an indicator for degree of degassing will be discussed.
Hilton, D.R.et al., Geochim. Cosmochim. Acta 56, 851-859 (1991).
Hilton, D.R. et al., Earth Planet. Sci. Lett. 120, 265-282 (1993).
Nagao, K. et al., Earth Planet. Sci. Lett. 44, 82-90 (1979).
Weiss, R.F., Deep Sea Research 17, 721-735 (1970).
Weiss, R.F., J. Chem. Eng. Data 16, 235-241 (1971).