DISCUSSION

The lack of significant sediment intercalations in the Site 1001 basalt stratigraphy and the similarity of the radiometric ages indicate that the penetrated lavas erupted over a relatively short time period (at least less than the uncertainty of the dating). Furthermore, the chemical variations of the basalts are very slight and indicate that all of the recovered basalts are from a single volcanic source (Sigurdsson, Leckie, Acton, et al., 1997). The shipboard XRF major and trace element analyses show that the basalts are similar to mid-ocean ridge basalts (MORB) and unpublished isotopic data from the Site 1001 basalts are also consistent with derivation from a MORB source (D.G. Pearson, pers. comm., 1998). The compositional similarity of the Site 1001 basalts and the Site 152 basalts, as well as the similarity in the age of the overlying sediments, suggests that they are related to the same magmatic event. Taken together, the uppermost igneous basement in this region of the lower Nicaraguan Rise/Hess Escarpment consists of MORB-like basalts that erupted at roughly 81 Ma.

Comparison between Site 1001 basalts and MORB erupted at the East Pacific Rise (EPR) shows that there are some subtle but significant compositional differences. Using the Joint Oceanographic Institutions, Inc., database for EPR glass analyses from 14°S to 16°N within the range of 7.2%-7.4% MgO, the Site 1001 glass (7.32% MgO) is at the lower end of the spectrum in terms of incompatible elements such as TiO2 (Fig. 2), Na2O, K2O, and P2O5. In terms of mantle source and/or melting conditions, this may signify that the Site 1001 lavas were derived from either a more depleted mantle source or higher degrees of partial melting relative to MORB.

In addition to the 90-Ma Gorgona Island basalts and komatiites (Kerr et al., 1996) and possibly some Cretaceous basalts in Haiti (Sen et al., 1988), the Site 1001/152 basalts can be included in the few MORB-like lavas from the Caribbean plateau. All of the other lavas have some "enriched" component (Sinton et al., 1998). Taken to- gether, the mantle sources for the Caribbean plateau are diverse and laterally heterogeneous.

Based on radiometric ages from the obducted margins of the plateau, a large proportion of the plateau erupted by at least ~90 Ma, and the synchroneity of ages across the region is consistent with a flood basalt origin for the bulk of the Caribbean plateau. A volumetrically secondary, but widespread magmatic event occurred at roughly 76 Ma (Sinton et al., 1998). The ~76-Ma magmatism on the Caribbean plateau could have resulted from lithospheric extension that caused mantle upwelling and partial melting. At this time the plateau was being emplaced between North and South America (Pindell and Barrett, 1990). It is possible that associated extension produced lithospheric thinning and allowed upwelling and decompression melting of residual hot mantle. Given the uncertainty of the radiometric and fossil ages, the Sites 1001 and 152 area may fall into the ~76-Ma "event," but it is apparent that submarine magmatism after 90 Ma occurred in various Caribbean locations (Sinton et al., 1998) and that the coincidence of ~76-Ma magmatism at several locations may not necessarily indicate a discrete event.

NEXT