Precision of the paleomagnetic method: An example from the Quaternary Eifel volcanics (Germany)

The total within-site dispersion of paleomagnetic data results from natural misalignment processes and experimentally produced dispersion. Although some of the sources of dispersion may be determined by experiments, this is not possible for others. The total accuracy of the paleomagnetic method so far has been determined only on historic lava flows. Thirty-seven Quaternary basaltic lava flows from the Eifel, Germany, have been re-sampled using the same outcrops as before. The new site-mean directions do not differ significantly from those of the original study. The angular distance between pairs of site-mean directions may be approximated by a Fisher distribution. Assuming that both studies are affected by the same natural dispersion processes, we obtain as the best value for the experimental dispersion s_exp = 4.33°. The total within-site dispersion shows a log-normal distribution with a mean of s_tot = 5.39°. From these values we calculate that the dispersion due to natural processes in the Eifel volcanic field is s_nat = 3.21°.