Continental rifts and passive continental margins show fundamental along‐axis segmentation patterns that have been attributed to one or a number of different processes: extensional fault geometry, variable stretching along strike, preexisting lithospheric compositional and structural heterogeneities, oblique rifting, and the presence or absence of eruptive volcanic centers. The length and width scales of the rift stage fault‐bounded basin systems change during the late evolution of the new plate boundary, and the role of magmatism may increase as rifting progresses to continental rupture. Along obliquely spreading ridges, first‐order mid‐ocean ridge geometries originate during the synrift stage, indicating an intimate relationship between magma production and transform fault spacing and location. The Gulf of Aden rift is a young ocean basin in which the earliest synrift to breakup structures are well exposed onshore and covered by thin sediment layers offshore. This obliquely spreading rift is considered magma‐poor and has several large‐offset transforms that originated during late stage rifting and control the first‐order axial segmentation of the spreading ridge. Widely spaced geophysical transects of passive margins that produce only isolated 2‐D images of crust and uppermost mantle structure are inadequate for evaluation of competing rift evolution models. Using closely spaced new geophysical and geological observations from the Gulf of Aden we show that rift sectors between transforms have a large internal variability over short distances (∼10 km): the ocean‐continent transition (OCT) evolves from a narrow magmatic transition to wider zones where continental mantle is probably exhumed. We suggest that this small‐scale variability may be explained (1) by the distribution of volcanism and (2) by the along‐strike differences in time‐averaged extension rate of the oblique rift system. The volcanism may be associated with (1) the long‐offset Alula‐Fartak Fracture Zone, which may enhance magma production on its younger side, or (2) channeled flow from the Afar plume material along the newly formed OCT and the spreading ridge. Oblique extension and/or hot spot interactions may thereby have a significant control on the styles of rifting and continental breakup and on the evolution of many magma‐poor margins.
In order to evaluate the activity of the Hawaiian hot spot, we calculate both the magma production rate, associated with volcanism, and the rate of swell formation, characteristic of the plume behavior. Both computations are made along the Emperor‐Hawaii track. Depth anomaly is calculated by correcting the 2′ bathymetry grid of Smith and Sandwell [1997] from thermal subsidence. A new filtering method is then used to separate the topography associated with volcanism and the swell surrounding the hot spot chain. The volume of magma includes the compensating root underlying the volcanoes, computed assuming either an Airy compensation (local) or a flexural root (regional) associated with the volcanic load. Next, the volume corresponding to the swell is calculated between the swell amplitude map and the zero value of the depth anomaly. Temporal variations of both volumes are then computed by the means of 1° × 10° windows translated along the hot spot track. Both volume fluxes are correlated through time and present (1) a general increase in amplitude for the last 30 Ma, indicating an increase in hot spot activity, and (2) short‐wavelength oscillations with a 5 m.y. period, which may reflect the presence of solitary waves in the plume conduit. Contrary to the swell volume flux, the magma production rate estimation is not dependent on the subsidence model and is still valid for the older part of the chain. It is thus the most relevant parameter to describe the temporal variation of the Hawaiian hot spot behavior.
Hotspot theory was first proposed on the basis of the observation of linear volcanic chains on the Pacific plate and assumed age progression within these chains. Knowledge of the ages of islands and seamounts is therefore of primary importance to analyzing intraplate volcanism and deciphering the history of hotspot tracks. In this paper we review published radiometric ages of islands and seamounts on the Pacific plate to help further reconstruction. We present a compilation of 1645 radiometric ages sorted by chain and further by island or seamount, along with a brief overview of each chain. Paleomagnetic ages obtained from seamount magnetism have not been considered, except for some oceanic plateaus (e.g., Shatsky rise). We do not consider foraminifer ages, which only give minimum ages of seamounts. Reliability problems intrinsic to the samples and to the radiometric dating methods must be considered. Dating of whole rocks must generally be disregarded unless they have been subject to special treatment, Ar/Ar incremental heating dating should be preferred over other methods, and data that do not pass the reliability criteria discussed by Baksi (this volume) should be disregarded. Thus use of the ages compiled in our database must be done in the light of filtering, and we encourage the user to check critically the initial papers in which the dates were published.