The Late Roman Antiquity walls of Le Mans in northwest France are one of the most representative and preserved examples of the urban fortifications that developed in the Roman provinces of Gaul and Germany at this period. Because of a lack of reliable chronological data, the construction of the walls was poorly dated, which made unclear its historical context. The main objective of this study was to reassess the date of construction using several methods on a well-preserved sector (sector 11) of the Late Roman Antiquity walls. The dated masonries, thoroughly studied in building archaeology beforehand, are characterized by an alternating of stone and brick courses. Sampling focused on one hand on mortars with four radiocarbon dates (14C) and six single-grain optically stimulated luminescence (SG-OSL) dates, and on the other hand on bricks with two OSL dates (quartz fine grain technique) and one archaeomagnetic date on a set of 104 bricks. The consistency between the dates on the two types of materials discards a possible reuse of the bricks from former Roman buildings. They were produced for the construction of the walls with the presence of several types of bricks likely reflecting a supply from at least two workshops. The dating program in the sector 11 also included 21 14C dates and six SG-OSL on protohistoric structures, Early Empire masonries and large medieval buildings, in order to investigate the evolution of the area over the long-term and to better constrain the chronology of the Late Roman Antiquity walls in Bayesian modelling. The chronological model (Chronomodel software) dates this construction between 301 and 423 CE at 95% of confidence. This date clearly excludes that the Late Roman Antiquity walls were built during the crises of the 3rd century, as previously thought, but rather in the stable political and economic context of the 4th century.

Absolute Thellier–Thellier archaeointensity determinations were conducted on 137 tiles and sedimentary blocks from five different structures from the ‘Site de la Molère’, southern France, to constrain the archaeological interpretation of a relative age difference between structures and add new data to the French secular variation database. We present 89 new archaeointensity and 20 new directional results, with averages of 57 (±8) μT in the second century ad and 68 (±7) μT in the third century ad . The overall data trend is consistent with the limited available data, indicating rapidly increasing field intensity from the second to the third century ad .

Hawaii is an ideal place for reconstructing the past variations of the Earth's magnetic field in the Pacific Ocean thanks to the almost continuous volcanic activity during the last 10000 yrs. We present here an updated compilation of palaeomagnetic data from historic and radiocarbon dated Hawaiian lava flows available for the last ten millennia. A total of 278 directional and 66 intensity reference data have been used for the calculation of the first full geomagnetic field reference secular variation (SV) curves for central Pacific covering the last ten millennia. The obtained SV curves are calculated following recent advances on curve building based on the Bayesian statistics and are well constrained for the last five millennia while for older periods their error envelopes are wide due to the scarce number of reference data. The new Bayesian SV curves show three clear intensity maxima during the last 3000 yrs that are accompanied by sharp directional changes. Such short-term variations of the geomagnetic field could be interpreted as archaeomagnetic jerks and could be an interesting feature of the geomagnetic field variation in the Pacific Ocean that should be further explored by new data. (C) 2017 Elsevier B.V. All rights reserved.

The efforts of geophysicists to describe geomagnetic field behaviour in the past lead to creation of different geomagnetic field models. On the other hand, the established regional palaeosecular variations of geomagnetic elements are increasingly used for dating purposes in archaeology. Both of these goals can be achieved if sufficient amounts of long archaeomagnetic data sets exist for different geographical regions. The accumulation of archaeomagnetic determinations began at the middle of the last century, parallel with the progressive development of experimental methodology and acceptance criteria. The presence of great number of old determinations requires their critical assessment. The important question about the reliability of the associated dating intervals should be also re-assessed. All this requires the continuous refinement and extension of the accumulated databases. This paper presents the last synthesis of Bulgarian archaeomagnetic database and the local palaeosecular variation curves obtained using a statistical treatment based on Bayesian approach (RenCurve software). The rock-magnetic characteristics of the newly included, non-published results are summarized.

The available Bayesian European Palaeosecular Variation Curves (PSVC) based on archaeomagnetic data have been used to derive an initial regional model for the geomagnetic field in Europe for the last 2000 years by using the Spherical Cap Harmonic Analysis (SCHA) technique. The resulting SCHA.DI.00 model provides the directional behaviour of the Earth’s magnetic field, but no information about intensity is supplied because input data are only directional. The first spherical cap harmonic coefficient, g00, has been used to normalise the rest of the coefficients. The spherical expansion extends up to Kint=2 which, given the size of the spherical cap used and the maximum data resolution, is equivalent, in terms of spatial wavelength, to a maximum degree of approximately 5 in the ordinary Spherical Harmonic Analysis (SHA). The SCHA.DI.00 model fits the present archaeomagnetic database for Europe more accurately than global models, although it is necessary to consider that regional SCHA model and global models did not use the same database. Palaeosecular variation curves given by SCHA.DI.00 are compared with available archaeomagnetic data from Bulgaria (Sofia), Italy (Etna) and Austria (Radstadt), not used in the development of the model. The SCHA.DI.00 model improves the fit to these palaeomagnetic data with respect to the CALS7K.2 [Korte, M., Constable, C.G., 2005. Continuous geomagnetic field models for the past 7 millennia: 2. CALS7K. Geochem. Geophys. Geosyst. 6, Q02H16. doi:10.1029/2004GC000801] global model, especially for Italy.

Reference archaeomagnetic secular variation (SV) curves recently have been proposed for the Iberian Peninsula and may now be used for archaeomagnetic dating. Archaeomagnetic dating is a relative dating technique that is strongly dependent on the age control of the data used to construct the reference curves. In order to test the method, an archaeological structure from central Spain has been studied. Samples have been taken for both archaeomagnetic and radiocarbon dating, and the results are compared. Close agreement is observed between both techniques, with the archaeomagnetic age of AD 603–999 overlapping the calibrated age of AD 770–890. These results demonstrate the reliability of the proposed reference curves as a dating tool within the Iberian Peninsula during this archaeological period.

This paper examines the limitations and deficiencies of the current British archaeomagnetic calibration curve and applies several mathematical approaches in an attempt to produce an improved secular variation curve for the UK for use in archaeomagnetic dating. The dataset compiled is the most complete available in the UK, incorporating published results, PhD theses and unpublished laboratory reports. It comprises 620 archaeomagnetic (directional) data and 238 direct observations of the geomagnetic field, and includes all relevant information available about the site, the archaeomagnetic direction and the archaeological age. A thorough examination of the data was performed to assess their quality and reliability. Various techniques were employed in order to use the data to construct a secular variation (SV) record: moving window with averaging and median, as well as Bayesian statistical modelling. The SV reference curve obtained for the past 4000 years is very similar to that from France, most differences occurring during the early medieval period (or Dark Ages). Two examples of dating of archaeological structures, medieval and pre-Roman, are presented based on the new SV curve for the UK and the implications for archaeomagnetic dating are discussed.

New archeointensity values have been determined from seven Spanish kilns sampled at the same archeological site and dated by archeological constraints as between 1100 and 1200 a.d. The directions of the characteristic remanent magnetization and paleointensities have been obtained from classical Thellier experiments conducted on 69 samples collected from the walls of the kilns. Thermoremanent magnetization (TRM) anisotropy and the cooling rate effect upon TRM intensity acquisition have been investigated in all the samples. Differences between mean archeointensities per site, before and after corrections for cooling rate are between 3 and 17%. The study of these seven “contemporaneous” kilns coming from the same archeological site and treated in the same way in the laboratory allows us to discuss the different factors that can cause the dispersion between sites generally observed in archeointensity studies. The results confirm, on one hand, the need for cooling rate corrections in this kind of study and, on the other hand, the need to obtain several archeointensity determinations per century in order to construct a reliable archeointensity curve. Finally, the new data provide a robust mean intensity (60.3±3.3μT at the latitude of Paris) for the middle of the XII century in Western Europe and constrain better the evolution of the geomagnetic field intensity during medieval times in this area.

A total of 58 new archaeomagnetic directions has been determined from archaeological structures in Spain. Together with five previous results they allow the compilation of the first archaeomagnetic catalogue for Spain, which includes 63 directions with ages ranging between the 2nd century BC and the 20th century AD. Characteristic remanence directions have been obtained from stepwise thermal and alternating field demagnetization. The hierarchical structure has been respected in the calculation of the mean site directions. Rock magnetic experiments reveal that the main magnetic carrier is magnetite or titanomagnetite with different titanium contents. The age estimate of the studied structures is generally well justified by archaeological constraints. For six structures the proposed date is also supported by physical methods. The data are in close agreement with the French secular variation (SV) curve. This catalogue represents the first step in the construction of a SV curve for the Iberian Peninsula, which will be of much use in archaeomagnetic dating and in modelling of the Earth's magnetic field in Western Europe.