HE-3 EVIDENCE FOR A WIDE ZONE OF ACTIVE MANTLE MELTING BENEATH THE CENTRAL ANDES

We report results of a regional survey of helium isotopes measured in water and gas samples in volcanic sulfataras and geothermal springs fr om the Central Andes of northern Chile and Bolivia between the latitud es 15 degrees S and 23 degrees S. The highest He-3/He-4 ratios (report ed as R/R(A) ratios: R = sample 3He/ 4He, R(A) = air He-3/He-4) are as sociated with the active volcanic are of the Western Cordillera (0.92 < R/R(A) < 5.52) and approach ratios found at other convergent margins in the circum-Pacific region. A significant He-3 component is also pr esent in fluid and gas samples from the high Altiplano plateau (0.48 < R/R(A) < 3.56) and the Eastern Cordillera (0.03 < R/R(A) < 1.2), up t o 300 km east of the active are and more than 300 km above the subduct ing slab. This wide zone of He-3 anomalies is delineated both to the e ast and the west by regions with low He-3/He-4 ratios (less than or eq ual to 0.2R(A)), typical of radiogenic helium production in the crust. Studies of the regional groundwater regime suggest that the wide zone of elevated He-3/He-4 values away from the active volcanic are is unl ikely to be caused by lateral and shallow transport of magmatic helium and there is no evidence for significant crustal sources of He-3. The high He-3/He-4 ratios are interpreted as reflecting degassing of vola tiles from mantle-derived magmas emplaced over an area 400 km wide ben eath and into crust up to 75 km thick. The subducting slab is at depth s of 100-350 km in this region. In the west, underneath the active vol canic arc, mantle melting is probably largely controlled by mantle hyd ration and dehydration and the helium isotope data can be used to deli neate the extent of the asthenospheric mantle wedge at depth. In contr ast, mantle melting behind the are, beneath the Altiplano and Eastern Cordillera, may be a result of convective removal of the base of the l ithosphere. The sharp cut-off in the mantle helium signal in the east is interpreted as marking the western edge of thick and relatively col d lithosphere, devoid of mantle melts, which could transport mantle vo latiles towards the surface. This may coincide with the limit of under thrusting of the Brazilian shield beneath the eastern margin of the Ce ntral Andes.