Horizontal Contraction of Oceanic Lithosphere Tested Using Azimuths of Transform Faults

A central hypothesis or approximation of plate tectonics is that the plates are rigid, which implies that oceanic lithosphere does not contract horizontally as it cools (hereinafter "no contraction"). An alternative hypothesis is that vertically averaged tensional thermal stress in the competent lithosphere is fully relieved by horizontal thermal contraction (hereinafter "full contraction"). These two hypotheses predict different azimuths for transform faults. We build on prior predictions of horizontal thermal contraction of oceanic lithosphere as a function of age to predict the bias induced in transform-fault azimuths by full contraction for 140 azimuths of transform faults that are globally distributed between 15 plate pairs. Predicted bias increases with the length of adjacent segments of mid-ocean ridges and depends on whether the adjacent ridges are stepped, crenellated, or a combination of the two. All else being equal, the bias decreases with the length of a transform fault and modestly decreases with increasing spreading rate. The value of the bias varies along a transform fault. To correct the observed transform-fault azimuths for the biases, we average the predicted values over the insonified portions of each transform fault. We find the bias to be as large as 2.5°, but more typically is ≤ 1.0°. We test whether correcting for the predicted biases improves the fit to plate motion data. To do so, we determine the sum-squared normalized misfit for various values of γ, which we define to be the fractional multiple of bias predicted for full contraction. γ = 1 corresponds to the full contraction, while γ = 0 corresponds to no contraction. We find that the minimum in sum-squared normalized misfit is obtained for γ = 0.9 ±0.4 (95% confidence limits), which excludes the hypothesis of no contraction, but is consistent with the hypothesis of full contraction. Application of the correction reduces but does not eliminate the longstanding misfit between the azimuth of the Kane transform fault with respect to those of the other North America-Nubia transform faults. We conclude that significant ridge-parallel horizontal thermal contraction occurs in young oceanic lithosphere and that it is accommodated by widening of transform-fault valleys, which causes biases in transform-fault azimuths up to 2.5°.