We present an improved version of the method of Le Goff et al. (2002) for tracing directional variations of the geomagnetic field over millennia. This spherical approach calculates mean directions over time using sliding windows of variable durations, combined with a bivariate extension of Fisher statistics. Window duration is set to include a minimum weighted number (WT) of data points, and each mean direction is computed independently at a constant time step. We apply the method to the French archeomagnetic database derived from the Thellier protocol and assess the influence of key parameters, including WT, data age uncertainties, and relocation errors. Calculations can also be performed using fixed window durations while allowing the weighted number of data points within each window to vary, yielding similar results. The method is further applied to different data selections within circular regions of variable radii across Europe. We identify regions where directional secular variations are best documented, highlighting gaps in southern and eastern Europe. Comparison with the SCHA.DIF.4k regional geomagnetic field model reveals moderate differences in accuracy and variability, likely reflecting the inherently different methodological approaches. We also present three archeomagnetic dating options based on the spherical framework. In particular, we introduce a method for computing probability density functions separately for declinations and inclinations. These options yield comparable dating results and can be selected according to the application context. Combined with our method for constructing reference geomagnetic directional curves, this framework enables consistent archeomagnetic dating across Europe and allows a detailed assessment of the sensitivity of dating results to modeling choices.

The central portion of the 2019 +/- 2 Ma Vredefort (South Africa) impact structure comprises a 40-50 km diameter central uplift of Archean basement rocks surrounded by a 15-20 km wide collar of late Archaean to early Proterozoic Witwatersrand Supergroup sedimentary and volcanic rocks. The collar is characterized by a ring of strongly negative (up to -5 500 nT) aeromagnetic anomalies surrounding much of the structure where the strata dip steeply to overturned. To better understand the origin of this magnetic feature, we undertook a ground survey along 20 transects (340 km) in the Vredefort structure using a three-axis fluxgate magnetometer mounted on a mountain bicycle. Upward continuation of our profiles to 150 m matches the aeromagnetic data in shape and amplitude. From the bicycle measurements, we pinpointed the rocks responsible for the extremely negative anomalies. Field observations and microfabric analyses of the rocks from six outcrops substantiated that the magnetic signal correlates with 10-100 m thick metamorphosed banded iron formations (BIFs) at the base of the supergroup as the main producer of the anomalies. Paleomagnetic samples collected from the rocks at the surface that produce the most intense anomalies (up to -22 000 nT) have extremely high natural remanent magnetization intensities (up to >1000 Am-1) likely arising from lightning strikes. Stepwise demagnetization and rock magnetic experiments establish a new protocol to distinguish samples that escaped remagnetization from lightning and possess the established 2.02 Ga paleodirection at Vredefort. From a suite of thermoremanent magnetization (TRM) experiments, the best estimate for the paleofield intensity at the time of impact was 52 mu T, corresponding to an average remanence of 32.5 Am-1. The results of the TRM experiments together with the paleodirection enabled us to successfully model the prominent negative anomalies in the metasediments only when accounting for the post-impact orientation of the BIFs. We interpret the strongly negative magnetic anomalies in the collar region as being formed directly after crater exhumation and uplift of the rocks. This interpretation implies that Bushveld-related metamorphism at 2.06 Ga created the up to mm-sized magnetite and garnet crystals in the BIFs, which resided at temperatures higher than the Curie temperature of magnetite (580 degrees C) until the impact rapidly brought the BIFs close to the surface, where magnetite cooled to acquire a thermal remanence in the 2.02 Ga field.

We present a two-method spherical approach to archeomagnetic dating based on directional variations of the geomagnetic field after vector treatment of all data, including individual in situ structure-level data used to calculate the reference variation curve and the archeomagnetic direction to be dated. In this paper, the reference curve for France was determined from a compilation of data acquired from kiln structures using a sliding window technique in which the varying durations and time shifts between windows are fixed according to the temporal distribution of individual reference data, as well as the bivariate extension of Fisher's statistics. The first dating method involves identifying the time interval(s) in which the direction to be dated is closest to the mean directions that define the reference curve. The angles between the direction to be dated and reference curve directions, and the Fisher probability density function allow us to derive a coincidence probability density curve, from which it is possible to estimate a 95% probability level-based dating interval. The second dating method involves the determination of an archeomagnetic date in which the direction to be dated is statistically identical to a dated reference curve direction at the 95% confidence level. This approach is much more restrictive than the previous method because it requires an excellent agreement between the test and reference directions to obtain a dating result, while the first method is only based on relative proximity. Using examples of archeomagnetic dating, we show that these two methods are complementarity and should be applied jointly to account for some of the limitations inherent in archeomagnetic dating, particularly due to the dispersion of the reference structure-level data.

Since 2004, numerous archeomagnetic intensity data have been obtained using the vibrating sample magnetometer Triaxe, which measures full-vector magnetization directly at high temperatures, in either an applied or zero field. Satisfactory comparisons have been made between Triaxe intensity data and results derived from more conventional Thellier-Thellier type techniques, indicating the reliability of Triaxe data. For each specimen analyzed, a Triaxe archeointensity value is obtained from the average of R'(Ti) data. The R'(Ti) parameter is determined every 5 degrees C and corresponds to the ratio, multiplied by the laboratory field intensity, between the natural remanent magnetization (NRM) and laboratory-thermoremanent magnetization (TRMlab) fractions that are lost between reference temperature T1 and a given temperature Ti between T1 and reference temperature T2. Here, we introduce an additional parameter, based on so-called AutoR'(Ti) data, to facilitate and improve the interpretation of Triaxe measurements. Each individual AutoR'(Ti) datum corresponds to an averaged R'(Ti) value; the AutoR'(Ti) dataset is then obtained by gradually decreasing the temperature range from T1 to T2 to a minimum temperature interval near T2. Several examples of Triaxe measurements show the value of using AutoR'(Ti) data to isolate the most appropriate temperature range for an intensity determination, as well as to characterize the cooling rate effect on TRM acquisition. In particular, these experiments confirm that the Triaxe procedure minimizes this effect because, when it is present, it appears to be largely due to magnetic grains with high unblocking temperatures (> similar to 350 degrees C). Moreover, the AutoR'(Ti) dataset provides alternatives for estimating mean archeointensity values at both the fragment and fragment-group levels. We show that the simple approach used so far, based on the average of the R'(Ti) data determined over a single temperature interval, provides results as reliable as those derived from other options.

Abstract New archaeointensity results were obtained from 14 groups of baked-brick fragments collected in and around Pisa (Tuscany, Italy). The fragments were assembled from civil and religious buildings whose dating of construction or renovation, over the past millennium, was constrained by documentary sources. This collection, analysed using the Triaxe protocol, was found particularly suitable for intensity experiments, with a success rate of c. 84% corresponding to 276 fruitful specimens associated with 125 independent brick fragments. The Tuscan data clearly show a peak in intensity at the transition between the sixteenth and seventeenth centuries. They are also in very good agreement with, and complementary to, a dense dataset previously obtained in France. Considering the results available within a 700 km radius of Beaune (between Paris and Pisa), all satisfying a set of quality criteria, a mean geomagnetic field intensity variation curve was constructed for the past millennium using a newly developed transdimensional Bayesian technique. This curve, which thus incorporates the new Tuscan results, allows a better recognition of three intensity peaks (during the twelfth century, the fourteenth century and around AD 1600) in western Europe. The detail of this curve is a clear illustration of the centennial-scale resolution that can be achieved using accurate archaeointensity data.

Nearly 40 years ago, Emile Thellier published an article summarizing the archeomagnetic data he had obtained during his career, which had allowed him to recover the main features of the directional variations of the geomagnetic field in France for the last two millennia. This database went on to be significantly completed 25 years ago by Ileana Bucur, who had taken over Thellier's work on archeomagnetism; this forms the current basis of our knowledge of the directional evolution of the geomagnetic field in France. Since then, archeomagnetic studies have been continued at Thellier's historical laboratory in Saint Maur. This article presents the directional archeomagnetic data obtained in France over the past 25 years in this same laboratory. A total of 528 new data are presented, which, together with the 170 results obtained on the French territory previously listed in Bucur (1994), constitute the French directional database (698 data in all). All but two of these data were obtained using the experimental protocol developed by E. Thellier based on the analysis of large samples and on a magnetic viscosity test that has proved its reliability on numerous occasions. The directions from the entire French database have been precisely defined, with 95% of the alpha(95) values being less than 1.9 degrees, and similar to 50% being less than 0.8 degrees. The selection of 286 data with dating uncertainties of <= 50 years allowed us to compute a new reference directional variation curve for France since the first century BC using sliding windows, the variable durations and shifts of which were adapted to the time distribution of the available data, and the bivariate extension of the Fisher statistics. This shows ample and smooth variations, fairly similar to those previously determined from a much smaller database. The resulting secular variation curve is particularly well suited for use in archeomagnetic dating.

Abstract A complexity is emphasized in the distribution of French archaeomagnetic directions during the thirteenth and the fourteenth centuries AD. Data uncertainties, and the smoothing introduced when estimating an average secular variation curve, prevent scrutiny of the very nature of this complexity. It might correspond to a directional yaw, the nature of which would be compatible with the recent geomagnetic field evolution as traced by the gufm1 model. In order to emphasize this indeterminacy, a reference secular variation curve was constructed for dates between AD 1000 and 1500, including the yaw in question, and synthetic databases that mimic the accuracy and density characteristics of the true French archaeomagnetic database were considered for some of these. The synthetic curves hence obtained show that the dating accuracy of archaeomagnetic data is the crucial parameter for constructing a detailed secular variation path. The significant impact of the experimental data accuracy is also illustrated. Even more crucial is the fact that the precision of the data dating required to describe the directional variability over the century timescale largely exceeds the precision of the archaeological dates available for the structures generally studied. This highlights the intrinsic limitation of archaeomagnetism for regional reconstruction of century-scale geomagnetic field variations.

We present new archeointensity results obtained at two multi-layer archeological sites, Tell Atij and Tell Gudeda (northeastern Syria), dated from the Early Bronze Period in the third millennium BC. The archeointensity data were obtained using the experimental protocol developed for the Triaxe magnetometer. In total, 68 fragments (204 specimens) of 151 fragments analyzed passed our selection criteria, allowing average intensity values to be estimated for 14 archeological layers, nine at Tell Atij and five at Tell Gudeda. Based on the available archeological constraints, the different archeological layers of Tell Atij and Tell Gudeda were dated between similar to 2900 BC and similar to 2600 BC and between similar to 2550 BC and similar to 2325 BC, respectively. The Tell Atij data show a significant increase in intensity over the dated period, while the results from Tell Gudeda exhibit a V-shape evolution. Using high-quality data available from Syria, the Levant and Turkey, a regional geomagnetic field intensity variation curve spanning the entire third millennium BC was constructed using a trans-dimensional Bayesian method. It clearly shows two intensity peaks, around 2600 BC and at similar to 2300 BC, associated with variation rates of similar to 0.1-0.2 mu T/yr. This indicates that the occurrence of century-scale intensity peaks with rates of variation comparable to or even slightly higher than the maximum rates observed in the modern geomagnetic field is an ubiquitous feature of the geomagnetic secular variation. From an archeological point of view, the new archeointensity data strengthen the hypothesis that the successive occupation of Tell Atij and Tell Gudeda was synchronous with the two first urban phases of Mari, making possible a sustained trade network between these settlements during the third millennium BC. We further suggest that the end of Mari's first urban phase, contemporaneous with the abandonment of Tell Atij, might have been caused by a regional drought episode around 2600 BC. More generally, the Bayesian approach used to estimate the new reference intensity variation curve offers promising chronological constraints for archeological purposes. (C) 2020 Elsevier B.V. All rights reserved.

This study presents new archeointensity results from the multilayered settlement Yarim Tepe I located today in Northern Iraq. Archeological evidences and new radiocarbon dates indicate that this site was occupied for about four centuries, between the end of the 7th and the beginning of the 6th millennium BC, leading to a 6.5 m-thick sequence of archeological deposits. A series of 16 groups of potsherds, with a total of 76 fragments, were collected from superimposed stratigraphic layers, with thicknesses of similar to 30 cm on average. Archeointensity measurements were carried out using the Triaxe procedure, which takes into account both anisotropy and cooling rate effects on thermoremanent magnetization acquisition. In this study, 114 specimens from 40 fragments have fulfilled our selection criteria and mean archeointensity results were derived from nine different groups of fragments, with a minimum of three fragments per group. According to an age model constructed using a bootstrap approach, the new archeomagnetic record spans a time interval of similar to 220 years between similar to 6070 BCE and similar to 5850 BCE. No significant intensity variations were observed during this time interval, with an overall mean intensity value of 42.0 +/- 1.6 mu T. A comparison of the Yarim Tepe I data with other archeointensity results spanning the 7th and 6th millennia BC previously obtained from the Near East, as well as from Eastern and Western Europe was conducted. The Yarim Tepe I results enabled to constrain the short duration, one century at most, of an intensity peak evidenced around 5750 BCE from the Bulgarian database. Intensity variation rates associated with the ascending branch of the peak reached values as high as similar to 0.12-0.15 mu T/year. Summarizing the geomagnetic field intensity variations in the Near East and Eastern Europe during the entire 6th millennium BC, it appears that they were likely characterized by the occurrence of two short-lasting intensity peaks at similar to 5750 BCE and similar to 5500 BCE, with intensity variation rates similar or slightly higher than the maximum rates prevailing in the modern field. Finally, it is worth mentioning that the four radiocarbon dates reported in our study provide new constraints for deciphering the temporal correlation between the cultural/historical phases (Halaf and Hassuna) that were independently defined from excavations carried out in Iraq and Syria.

L’atelier de la Goulande (La Haute-Chapelle, Orne) se trouve au sein d’un vaste ensemble de fosses d’extraction d’argiles utilisées pour fabriquer de la poterie commune, des grès et des tuiles. Les prospections géophysiques menées à cet endroit ont permis de distinguer deux anomalies magnétiques, correspondant à trois fours référencés A, B et C. Le premier (A) est un four à pots longitudinal, daté par une analyse archéomagnétique de la seconde moitié du xiiie siècle ou du début du xive siècle. Sa morphologie et les productions qu’on peut lui associer sont proches de celles de l’atelier voisin de Saint-Georges-de-Rouelley, dont le réexamen des données archéomagnétiques recueillies au milieu des années 1980 conduit à une datation plus haute d’un demi-siècle (1260-1280 au lieu de 1325-1350). Les deux ateliers de potiers (Saint-Georges-de-Rouelley et four A de la Goulande) seraient donc sensiblement contemporains. À environ 1,50 m plus à l’est, se trouvent les restes des deux autres fours. Le premier (B) est un four à tuiles, auquel s’est substitué un four à pots longitudinal (C). En accord avec les données archéologiques, les résultats archéomagnétiques obtenus donnent des datations différentes pour chacun des fours étudiés. La production de tuiles qui s’y est développée a probablement pris fin au début du xve siècle. Finalement, une production de grès s’est substituée à la production de tuiles, vraisemblablement jusqu’au dernier tiers du xve siècle.

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 .

Large directional changes of remanent magnetization within lava flows that cooled during geomagnetic reversals have been reported in several studies. A geomagnetic scenario implies extremely rapid geomagnetic changes of several degrees per day, thus difficult to reconcile with the rate of the earth's core liquid motions. So far, no complete rock magnetic model provides a clear explanation. We revisited lava flows sandwiched between an underlying reverse and an overlying normal polarity flow marking the last reversal in three distinct volcanic sequences of the La Palma Island (Canary archipelago, Spain) that are characterized by a gradual evolution of the direction of their remanent magnetization from bottom to top. Cleaning efficiency of thermal demagnetization was not improved by very rapid heating and cooling rates as well as by continuous demagnetization using a Triaxe magnetometer. We did not observe partial self-reversals and minor changes in magnetic grain sizes are not related to the within-flow directional changes. Microscopic observations indicate poor exsolution, which suggests post-cooling thermochemical remagnetization processes. This scenario is strongly reinforced by laboratory experiments that show large resistance to thermal demagnetization when thermoremanence was acquired over a long time period. We speculate that in the present situation exsolution was reactivated during in field reheating and yielded formation of new magnetite, yet magnetic domain state rearrangements could also play a role. Initial reheating when the overlying flow took place, albeit moderate (less than 200-300 degrees C), was enough to produce overlying components with significantly higher unblocking temperatures.

The workshop of la Goulande (La Haute-Chapelle, Orne) is located close to a large complex of pits used to extract clay to make earthenware pottery, stonewares and tiles. Geophysical surveys have identified two magnetic anomalies corresponding to three kilns referenced A, B, and C. he first (A) is a longitudinal pottery kiln, dated by archaeomagnetic analysis to the second half of the 13th century or to the beginning of the 14th century. Its morphology and productions are similar to those of the neighbouring workshop of Saint-Georges-de-Rouelley, for which the re-examination of the archaeomagnetic data collected during the 1980s provides us with a date earlier by half a century (1260-1280 instead of 1325-1350). he two potters workshops (Saint-Georges-de-Rouelley and kiln A at la Goulande) are therefore contemporary. Around 1,50 m to the east, the remains of two other kilns were found. he first (B), a tile kiln, was replaced by a longitudinal pottery kiln (C). In correspondence with the archaeological data, archaeomagnetic results obtained provide different dates for each of the kilns studied. Finally, the production of stoneware most probably substituted the production of tiles up until the last third of the 15th century.

Nineteen new archeointensity results were obtained from the analysis of groups of French pottery fragments dated to the Early Middle Ages (6th to 10th centuries AD). They are from several medieval ceramic production sites, excavated mainly in Saran (Central France), and their precise dating was established based on typo-chronological characteristics. Intensity measurements were performed using the Triaxe protocol, which takes into account the effects on the intensity determinations of both thermoremanent magnetization anisotropy and cooling rate. Intensity analyses were also carried out on modern pottery produced at Saran during an experimental firing. The results show very good agreement with the geomagnetic field intensity directly measured inside and around the kiln, thus reasserting the reliability of the Triaxe protocol and the relevance of the quality criteria used. They further demonstrate the potential of the Saran pottery production for archeomagnetism. The new archeointensity results allow a precise and coherent description of the geomagnetic field intensity variations in Western Europe during the Early Medieval period, which was until now poorly documented. They show a significant increase in intensity during the 6th century AD, high intensity values from the 7th to the 9th century, with a minimum of small amplitude at the transition between the 7th and the 8th centuries and finally an important decrease until the beginning of the 11th century. Together with published intensity results available within a radius of 700km around Paris, the new data were used to compute a master curve of the Western European geomagnetic intensity variations over the past 1500years. This curve clearly exhibits five intensity maxima: at the transition between the 6th and 7th century AD, at the middle of the 9th century, during the 12th century, in the second part of the 14th century and at the very beginning of the 17th century AD. Some of these peaks are smoothed, or nearly absent when the selection of the data is extended to a 1250km radius around Paris. The apparent regularity in the occurrence of intensity maxima, with a recurrence of ∼250years, is particularly intriguing and might reflect a new characteristic of the secular variation, at least in Western Europe. It clearly requires further investigation and in particular the acquisition of new data from older periods.

Measuring the intensity of the Moon's ancient magnetic field is an important goal of lunar sample analysis. Paleointensity estimates using thermal methods (Thellier–Thellier) raise the problem of sample alteration, especially for lunar samples carrying sulfides. To address this problem, we made real-time measurements of sample magnetization during heating with a three-axis vibrating sample magnetometer (the Triaxe, LeGoff and Gallet, 2004), in the hope that rapid heating would minimize alteration (Coe et al., 2014). We studied the Millbillillie meteorite, which has a lunar-like mineralogy. We find that after repeated heating phases to ∼600°C, we form pyrrhotite, magnetite, and magnetic phases with low (200–270°C) Curie temperatures. These low-temperature phases appear after pyrrhotite has apparently been destroyed by subsequent heating phases, and their mineralogy is unknown. These results have implications for the paleomagnetic study of any extraterrestrial samples with even small amounts of sulfides, and further experiments are required to understand them.

Une étude archéomagnétique à des fins de datation a été menée sur deux foyers en carreaux de terre cuite mis au jour sur le site de l’ancienne abbaye cistercienne de Morimond (Haute-Marne). La désaimantation thermique complète des échantillons prélevés a permis de définir pour chacun des deux foyers une direction archéomagnétique moyenne précise, acquise lors de leur dernière utilisation. Pour dater cet instant, les directions archéomagnétiques ont été comparées à une courbe des variations directionnelles du champ géomagnétique construite à partir de données obtenues en France et dans des pays voisins. Nous obtenons, à 95% de confiance, une datation comprise entre 1585 et 1615 après J.-C. pour le premier foyer et entre 1525 et 1605 après J.-C. pour le second foyer. Bien que leurs deux intervalles d'âge se recouvrent partiellement, les deux directions archéomagnétiques moyennes ne sont pas compatibles à 95% ce qui indique que les arrêts de fonctionnement des deux foyers ne sont pas contemporains. Ces résultats archéomagnétiques confirment deux phases métallurgiques observées indépendamment lors des fouilles. Ils précisent les datations issues des radiocarbones qui ne permettaient pas de discriminer chronologiquement les deux phases. Les datations envisagées se calent avant l’abandon du bâtiment suite à plusieurs saccages évoqués dans les sources écrites.