Temporal coincidence between continental flood basalts and breakup has been
noted for almost three decades. Eight major continental flood basalts have
been produced over the last 300 Ma. The most recent, the Ethiopian traps,
erupted in about 1 Myr at 30 Ma. Rifting in the Red Sea and Gulf of Aden, a
nd possibly East African rift started at about the same time. A second trap
-like episode occurred around 2 Ma and formation of true oceanic crust is d
ue in the next few Myr. We find similar relationships for the 60 Ma Greenla
nd traps and opening of the North Atlantic, 65 Ma Deccan traps and opening
of the NW Indian Ocean, 132 Ma Parana traps and South Atlantic, 184 Ma Karo
o traps and SW Indian Ocean, and 200 Ma Central Atlantic Margin flood basal
ts and opening of the Central Atlantic Ocean. The 250 Ma Siberian and 258 M
a Emeishan traps seem to correlate with major, if aborted, phases of riftin
g. Rifting asymmetry, apparent triple junctions and rift propagation (towar
ds the flood basalt area) are common features that may, together with the r
elative timings of flood basalt, seaward dipping reflector and oceanic crus
t production, depend on a number of plume- and lithosphere- related factors
. We propose a mixed scenario of 'active/passive' rifting to account for th
ese observations. In all cases, an active component (a plume and resulting
flood basalt) is a pre-requisite for the breakup of a major oceanic basin.
But lifting must be allowed by plate-boundary forces and is influenced by p
re-existing heterogeneities in lithospheric structure. The best example is
the Atlantic Ocean, whose large-scale geometry with three large basins was
imposed by the impact points of three mantle plumes. (C) 1999 Elsevier Scie
nce B.V. All rights reserved.