Maxime Le Goff

Three millennia of directional variation of the Earth’s magnetic field in western Europe as revealed by archeological artefacts

Directional secular variation of the geomagnetic field over the last 2000 years has been defined in western Europe from numerous archeomagnetic studies. However, the number of archeomagnetic results for older periods is much more limited. For this reason, we present new data obtained from fired archeological structures found in two French sites (Loupiac and Aspiran) dated within the first millennium b.c. (latest Bronze-earliest Iron Age transition, ∼850–700 b.c. and Iron Age, ∼525–475 b.c., respectively). From a compilation of archeomagnetic results from western Europe (Great Britain, Italy and France) and northern Africa (Tunisia), we propose a directional secular variation curve for western Europe that covers the entire first millennium b.c. This curve exhibits a large clockwise motion with rapid changes during the first half of the millennium, while the last four centuries b.c. are characterized by weak variations.

Archeomagnetic dating of Mediterranean volcanics of the last 2100 years: validity and limits

Abstract Archeomagnetic dating developed at St. Maur laboratory has been applied to the Mediterranean volcanoes Etna, Vesuvius and Ischia. The method involves samples from lava flows or high temperature emplaced pyroclasts (welded scoriae, pumice, etc.) weighing 0.5–1 kg each, that allows reaching a precision of a few tenths of a degree on the direction of their thermoremanent magnetization, and hence a semi-angle of the Fisher 95% confidence cone between 0.6 and 1.8° for every volcanic unit. Among the factors reducing precision on the mean magnetic direction, the most important appears to be a distortion of the ambient field induced by magnetization of the cooling lava, which means that a number of samples should be collected over a large area. Age determination is based upon similarity between variation curves of the Direction of Earth’s Magnetic Field (DEMF) reconstructed in France from 120 well-dated archeological sites, and on Italian volcanoes from historically dated eruptions. A total of 63 lava flows and pyroclastic units, such as cinder cones or nuee ardente deposits, are shown to be dated with an overall precision of ±40 years for the last 1500 years, and ±50 to ±100 years from AD 500 to 150 BC, this lesser precision resulting from both an increasing uncertainty on the shape of the DEMF curve and a smaller variation of the DEMF itself. This irregularity of the DEMF path plus an increasing number of ambiguities, related to similarity of the DEMF at different times further into the past, are the most serious limitations of the method. Though well-dated eruptions are known for the last two millennia, retrieval of their products is often misleading and about 50% of volcanics presumed of known date prior to the 17th century are in fact of older age, discrepancies usually reaching several hundreds of years. Owing to good agreement between the DEMF curves of France and southern Italy, the method may confidently be extended to volcanic materials from the whole of Mediterranean Europe, provided there are firm constraints that they were erupted within the last 2100 years.

On archeomagnetic secular variation curves and archeomagnetic dating

Abstract Secular variation (SV) of the Earth’s magnetic field can be used for dating purposes by comparing archeomagnetic directions of unknown ages with a well-dated reference curve. In this study, we propose a dating technique based on the statistics of McFadden and McElhinny [Geophys. J. Int. 103 (1990) 725] for testing the hypothesis that two Fisherian distributions of individual directions share a common mean direction. The statistics are adapted to test the degree of compatibility between one individual Fisherian mean direction and a reference curve constructed using the bivariate extension of the Fisher distribution. Furthermore, as the density of the data which define the archeomagnetic reference curve varies in time, we suggest that one computes the mean directions with moving windows of varying duration, where both the window widths and the time shifts between successive mean directions are fixed when a minimum threshold density of data is reached within each time interval. In our paper, we apply this new procedure to the French archeomagnetic data set.

Static stress demagnetization of single and multidomain magnetite with implications for meteorite impacts

Stress demagnetization effects on ferromagnetic minerals are poorly known, especially above 1 GPa, and when initially magnetized under pressure and then subjected to further stress. Our experiments on pure magnetite under quasi-hydrostatic loads in the presence of a small (Earth’s) field show that stress demagnetization depends on domain state and stress history. Viewed globally, the results follow a simple law where the percentage loss in magnetic moment is the inverse of pressure (e.g., 50% loss in moment at 1 GPa, 67% loss at 2 GPa, etc.). Our experiments also quantify the effect of demagnetization upon stress release, where the moment upon full decompression is two-thirds less than the moment when decompression first began. Given the magnitude of the stress demagnetization effect, we conclude that the presence or absence of a planetary magnetic field cannot be deduced from the magnetic fields measured over meteorite craters, such as those on Mars.

Rock magnetic property and paleointensity determination on historical Santorini lava flows

Eight historical dacitic lava flows from Santorini with ages between 46 A.D. and 1950 A.D. (four of them within the past century) have been subjected to detailed rock magnetic analyses and various experiments of absolute paleointensity. Thermomagnetic measurements and acquisition of isothermal magnetization have revealed the presence of two physically distinct magnetic phases with Curie temperatures of 280°C and 500°C. In most of the samples, the second phase does not play a prominent role for the characteristic remanent magnetization, which is dominated by titanomagnetite. Magnetostatic interaction is very limited and does not considerably change upon heating. Back-field curve spectra indicate a good thermochemical stability of these dacitic lava samples, which is also supported by the absence of noticeable changes in the remanent coercive force prior heating to 450°C. Hysteresis measurements show typical pseudo-single-domain behavior without noticeable superparamagnetism. Such characteristics were favorable to conduct and to test the most widely used experimental approaches for absolute paleointensity determination. Despite a success rate of 38%, the microwave technique has provided rather scattered within-flow determinations. The results obtained from approaches involving alternating field demagnetization were biased by considerable differences between the NRM and the TRM coercive force spectra. We have also noticed that most determinations obtained by microwave heating differ from the historical field value at the site for the most recent flows. Last, techniques involving double-heating protocols were successful due to a dominant low Curie temperature phase with a narrow grain size distribution. The results were characterized by low dispersion and were found in good agreement with the historical field.

A reappraisal of instrumental magnetic measurements made in Western Europe before AD 1750: confronting historical geomagnetism and archeomagnetism

We present a new compilation and analysis of historical geomagnetic measurements made in Western Europe before AD 1750. The dataset in its ensemble provides a coherent evolution of magnetic field directions. Several data points excluded from previous analyses actually appear very consistent with most of the present compilation. A new average historical curve is computed for Paris, which is in very good agreement with the archeomagnetic data obtained in France, while significantly differing from the directional curve expected for Paris before AD 1675 based on the gufm1 model (Jackson et al. in Philos Trans R Soc Lond A 358:957–990, 2000). This finding suggests that the older segment of the gufm1 model lacks reliability and should be improved. Similarly, the historical part of the regional geomagnetic field model built for Europe by Pavón-Carrasco et al. (Geochem Geophys Geosyst 10:Q03013, 2009) should be revised because it erroneously incorporates directions derived from the gufm1 model.Graphical abstract.