A new geomagnetic polarity time scale for the Late Cretaceous and Cenozoic

We have constructed a magnetic polarity time scale for the Late Cretaceous and Cenozoic based on an analysis of marine magnetic profiles from the world's ocean basins. This is the first time, since Heirtzler et al. (1968) published their time scale, that the relative widths of the magnetic polarity intervals for the entire Late Cretaceous and Cenozoic have been systematically determined from magnetic profiles. A composite geomagnetic polarity sequence was derived based primarily on data from the South Atlantic. Anomaly spacings in the South Atlantic were constrained by a combination of finite rotation poles and averages of stacked profiles. Fine‐scale information was derived from magnetic profiles on faster spreading ridges in the Pacific and Indian Oceans and inserted into the South Atlantic sequence. Based on the assumption that spreading rates in the South Atlantic were smoothly varying but not necessarily constant, a time scale was generated by using a spline function to fit a set of nine age calibration points plus the zero‐age ridge axis to the composite polarity sequence. The derived spreading history of the South Atlantic shows a regular variation in spreading rate, decreasing in the Late Cretaceous from a high of almost 70 mm/yr (full rate) at around anomaly 33–34 time to a low of about 30 mm/yr by anomaly 27 time in the early Paleocene, increasing to about 55 mm/yr by about anomaly 15 time in the late Eocene, and then gradually decreasing over the Oligocene and the Neogene to the recent rate of about 32 mm/yr. The new time scale has several significant differences from previous time scales. For example, chron C5n is ∼0.5 m.y. older and chrons C9 through C24 are 2–3 m.y. younger than in the chronologies of Berggren et al. (1985b) and Harland et al. (1990). Additional small‐scale anomalies (tiny wiggles) that represent either very short polarity intervals or intensity fluctuations of the dipole field have been identified from several intervals in the Cenozoic including a large number of tiny wiggles between anomalies 24 and 27. Spreading rates on several other ridges, including the Southeast Indian Ridge, the East Pacific Rise, the Pacific‐Antarctic Ridge, the Chile Ridge, the North Pacific, and the Central Atlantic, were analyzed in order to evaluate the accuracy of the new time scale. Globally synchronous variations in spreading rate that were previously observed around anomalies 20, 6C, and in the late Neogene have been eliminated. The new time scale helps to resolve events at the times of major plate reorganizations. For example, anomaly 3A (5.6 Ma) is now seen to be a time of sudden spreading rate changes in the Southeast Indian, Pacific‐Antarctic, and Chile ridges and may correspond to the time of the change in Pacific absolute plate motion proposed by others. Spreading rates in the North Pacific became increasingly irregular in the Oligocene, culminating in a precipitous drop at anomaly 6C time.