LATE CENOZOIC RATES OF MAGMATIC ACTIVITY IN THE CENTRAL ANDES AND THEIR RELATIONSHIPS TO CONTINENTAL-CRUST FORMATION AND THICKENING

Data from four sources are used to assess the rate of magmatic addition to the central Andean crust : (i) a closely studied segment of the are between latitudes 21 degrees and 22 degrees S; (ii) all the potentially active volcanoes within the Central Volcanic Zone; (iii) all the volcanic additions in the last 1 Ma; and (iv) those within the last 10 Ma. These indicate a consistent rate of effusive volcanism of 3 x 10(-6) km(3) a(-1) km(-1). Rates of silicic pyroclastic volcanism, the ratios of extrusive to intrusive rocks, and the fraction of material newly derived from the mantle are more difficult to constrain, but conservative estimates indicate that the overall rate of magmatic addition to the crust is of the order of 1.3 x 10(-5) km(3) a(-1) km(-1). This rate is inadequate to account for the crustal thickening since 15 Ma ago, but is consistent with estimates from, for example, the Aleutian are (3.4 x 10(-5) km(3) a(-1) km(-1)). These data support Isacks' contention that the thick crustal root beneath the Central Andes results from tectonic shortening and thickening, rather than magmatism.

There is little evidence of sediment subduction at the Chile-Peru trench at the present day, but tectonic erosion at a rate of about 1 x 10(-6) km(3) a(-1) km(-1) may have continued since the Jurassic. Their net effects are difficult to determine, primarily because of uncertainties about petrological processes in the lower crust. The extensive ignimbrite province of the Sierra Madre Occidental of Mexico has some similarities with the Central Andes, but eruption rates are 20 times larger (3.7 x 10(-5) km(3) a(-1) km(-1)). epsilon(Nd) values for the Sierra Madre dacites are high (+0.2-+0.48) suggesting that they evolved by crystal fractionation from mantle-derived basaltic parent magmas, and are consistent with rapid formation of continental crust. In the Central Volcanic Zone, however, higher Rb/Sr ratios in similar dacites suggest large scale melting of middle to lower continental crust, perhaps in association with lithospheric thinning. If lithospheric thinning is an important cause of melting, it will reduce estimates of net magmatic additions to the crust.