High‐resolution inversion for South Atlantic Plate kinematics using joint altimeter and magnetic anomaly data

Abstract

We present an inversion for plate kinematics that solves for finite rotation parameters using fracture zone (FZ) and magnetic anomaly location data jointly. We define misfit functions that incorporate properties unique to each data type; in particular, the FZ misfit function does not depend upon aligm•ent of conjugate FZ traces in the same way as magnetic linearions under the firrite rotations. This property is useful for FZ locations, in which the signals on conjugate sides of the ridge may include systematic differences, or where data from one side of the ridge are sparse or missing. Formal error bounds estimated for the pole parameters show that the magnetic and FZ data are complementary in their information content. Error bounds computed for the joint inversion are substantially smaller than for either the FZ or magnetics data used separately, indicating that simultaneous use of the data in an inversion is crucial. We apply this method to Seasat altimeter data and magnetic anomaly picks in the South Atlantic. We solve for finite poles corresponding to magnetic anomalies 5, 6, 8, 13, 21, 22, 25, 30, 33, 33r, and 34; these define a smooth path over the past 84 m.y. indicating that the plate motions have not been as erratic as found previously.

The tectonic evolution of the South Atlantic from Late Jurassic to present

“…The series of finite rotation poles proposed in this paper results in a platekinematic model without severe, short-term changes in directions of relative plate motions as proposed previously. Hence our results for the time interval from Chron 34 to present day concur with Shaw and Cande's (1990) conclusion that available data do not support a model of erratic changes in sea floor spreading directions in the South Atlantic.…”

“…Ma~~~ti~ anomaly pick locations for Chron 34 to 3 correspond to the young end of tbe normal polarity interval Shaw, 1987;Cande et al. 1988;Shaw and Cande, 1990;Craig and Sandwell, 1988;MiilIer et al,, in press MacArthur et al, 19871, 2.9 times better than the 10 cm accuracy of Seasat data for wave heights less than 20 m (Tapley et al, 1982). Geosat has an ~q~~~torial track spacing of 164 km (McConathy and Kilgus, 1987).…”

Paleogene opening of Drake Passage

“…We have therefore extended the technique of Shaw and Cande [15], involving joint inversion of sea floor isochrons, derived from picks of magnetic anomalies, together with continuous flow lines derived from satellite gravity anomalies, to permit the inclusion of data from only one flank of a spreading segment. This dnon-conjugateT fitting involves matching each isochron with every other isochron on the same flank (Fig.…”

Factors influencing the locations of hot spots determined by the hot‐spotting technique

“…One can then use a trial set of stage rotations to estimate the (pseudo) age for all seamounts in a seamount chain and use these estimated ages to "back-track" each seamount to its source. The distances between CVA-based hot spot locations and "back-tracked" seamounts may then be used to design an iterative, nonlinear inversion scheme to optimize stage pole parameters, in some respects similar to standard inversions based on isochron data [Shaw and Cande, 1990]. If hot spots are stationary then time is not required to determine the stage poles and opening angles, but if they migrate we need estimates of seamount ages in order to back-track the hot spot to where it was at the time the seamount was emplaced.…”