Palaeomagnetic study of the Xitle-Pedregal de San Angel lava flow, southern Basin of Mexico

Abstract

A detailed study of the ∼ 2000 years old Xitle-Pedregal de San Angel volcanic field in the southern Basin of Mexico was undertaken to assess the reliability of the palaeomagnetic record as derived from fresh well-preserved and exposed lava flows. The Xitle vent is on the slope of the Ajusco volcano, which results in a topographic difference of over 800 m in less than 12 km of horizontal distance. Most sites present a mean direction, with a small within-site dispersion, around the dipolar direction for the locality, but some sites, particularly in the basin sector of flat relief away from the vent, show shallow inclinations. Anisotropy of magnetic susceptibility (AMS) shows normal ‘flow’ fabrics with horizontal foliation planes and small anisotropy degree. AMS lineations correlate with observed flow directions. The magnetic properties vary systematically across flow units, but directions do not show a consistent pattern. Secular variation effects do not apparently contribute to the shallow inclinations or directional scatter. Some lava structures like pressure crests and blocky fronts give shallow inclinations and scattered directions, respectively. The resulting overall mean direction is well defined and close to the dipolar direction ( B = 26, Dec = 359.8°, Inc = 32.8°, k = 167, α 95 = 2.2°), but this excludes two apparent directional groups. The mean direction for one group, with B = 19, Dec = 359.0°, Inc = 35.1°, k = 247, α 95 = 2.1°, may be the representative estimate for the field. Shallow inclinations are considered anomalous and associated to a characteristic specific to given sectors of lava flows. Palaeointensity determinations have been obtained for six samples from five sites by using the Thellier and Shaw methods. Results agree well with previous studies, however, the standard deviation calculated for the mean value remains high after incorporation of the new data. Mean palaeointensity based on five new determinations and eight early data is N = 13, 56.50 ± 6.16 mT (and the mean for the overall data set is N = 20, 60.50 ± 9.21 mT). Results illustrate some practical considerations and limitations in methodology and the reliability of the palaeomagnetic record in volcanic rocks.