Testing corrections for paleomagnetic inclination error in sedimentary rocks: A comparative approach

Abstract

Paleomagnetic inclinations in sedimentary formations are frequently suspected of being too shallow. Recognition and correction of shallow bias is therefore critical for paleogeographical reconstructions. This paper tests the reliability of the elongation/inclination ( E/ I) correction method in several ways. First we consider the E/ I trends predicted by various PSV models. We explored the role of sample size on the reliability of the E/ I estimates and found that for data sets smaller than ∼ 100–150, the results were less reliable. The Giant Gaussian Process-type paleosecular variation models were all constrained by paleomagnetic data from lava flows of the last five million years. Therefore, to test whether the method can be used in more ancient times, we compare model predictions of E/ I trends with observations from five Large Igneous Provinces since the early Cretaceous (Yemen, Kerguelen, Faroe Islands, Deccan and Paraná basalts). All data are consistent at the 95% level of confidence with the E/ I trends predicted by the paleosecular variation models. The Paraná data set also illustrated the effect of unrecognized tilting and combining data over a large latitudinal spread on the E/ I estimates underscoring the necessity of adhering to the two principle assumptions of the method. Then we discuss the geological implications of various applications of the E/ I method. In general the E/ I corrected data are more consistent with data from contemporaneous lavas, with predictions from the well constrained synthetic apparent polar wander paths, and other geological constraints. Finally, we compare the E/ I corrections with corrections from an entirely different method of inclination correction: the anisotropy of remanence method of Jackson et al. [Jackson, M.J., Banerjee, S.K., Marvin, J.A., Lu, R., Gruber, W., 1991. Detrital remanence, inclination errors and anhysteretic remanence anisotropy: quantitative model and experimental results. Geophys. J. Int. 104, 95–103] which relies on measurement of remanence and particle anisotropies of the sediments. In the two cases where a direct comparison can be made, the two methods give corrections that are consistent within error. In summary, it appears that the E/ I method for recognizing and corrected the effects of sedimentary flattening is reasonably robust for at least the Mesozoic and Cenozoic when the source of scatter is geomagnetic and sedimentary flattening in origin.