Archaeomagnetic Directions Research Articles

A preliminary secular variation reference curve for archaeomagnetic dating in Austria

The construction of a secular variation (SV) reference curve for a region for which little or no archaeomagnetic directions are available is presented here. A SV curve is illustrated for Austria, centred on Radstadt (47.38°N, 13.45°E) and based on data from sites in other countries less than 500 km away. The published data were selected on site characteristics of N = 3 and k = 50, and dated within 400 yr. This yielded 170 directions from which a SV curve was derived using Bayesian techniques. The obtained reference curve represents the past 2300 yr. New data, mainly from Austria, substantiate this curve and confirm the validity of the techniques employed which can, therefore, be applied for similar situations. Another test has been made using the German reference curve for dating the Austrian archaeological sites, here a systematic shift to older times in the order 30­110 yr occurs.

First directional archaeomagnetic results from Syria: evidence from Tell Mishrifeh/Qatna

SUMMARY We report on the first archaeomagnetic directional data obtained from Syria. Reliable palaeomagnetic directions were collected from four pyrotechnological archaeological structures dated by type of pottery and radiocarbon from the 19th to 8th century BC (Middle Bronze to Iron Age) in Tell Mishrifeh/Qatna, Syria. Fairly constant northward declination values (within confidence limits) were obtained, whereas inclination values increased from 42.3 ◦ to 73.0 ◦ from the Middle Bronze to Iron Age. At first approximation, these new data are consistent with coeval archaeomagnetic directions from Western Europe and Bulgaria. The high inclination (73.0 ◦ ± 2.8 ◦ ) documented for the Iron Age (translating to a VGP latitude of 65.9 ◦ N) suggests that very strong secular variation was occurring during the 8th century BC.

Hierarchical versus stratification statistical analysis of archaeomagnetic directions: the secular variation curve for Hungary

SUMMARY Extended by 26 new data, the present archaeomagnetic data set for Hungary consists of 185 couples of inclination and declination distributed between BC 350 and AD 1800. We analysed these data by separately using a stratification and a hierarchical statistical method for unravelling the directional secular variation of the ‘local’ geomagnetic field. For the calculations we applied weighted summation and the moving window technique. In order to have a reasonably quantity of data in every window we selected 100 yr windows with 75 per cent overlaps between consecutive windows. Our calculations ended up with two, almost identical models of secular variation, despite of the widely different weightings that are required in the two approaches. Nevertheless, of the two solutions the one produced by the hierarchical modelling is preferred because the correct statistics in archaeomagnetism are of a hierarchical type.

Comments on “Chronology of Vesuvius' activity from A.D. 79 to 1631 based on archeomagnetism of lavas and historical sources” by C. Principe et al., Bull Volcanol

This paper shortly reviews the various factors which concur to the accuracy of archaeomagnetic dating of volcanic rocks: the error on the laboratory procedure; the assumption that the archaeomagnetic direction matches the paleofield direction present when the rocks formed; the limits of the model used for dating. The very small “analytical error” claimed in the paper by Principe et al. (2004) mainly depends on their focussing on the precision of laboratory procedures, disregarding the other sources of errors, which in fact are known in the literature to be far from negligible.

Archaeomagnetic secular variation in Germany during the past 2500 years

SUMMARY The German archaeomagnetic data set was supplemented with 35 new directions from Ger- man sites mainly dating from the past 3000 yr. Together with archaeomagnetic data from the neighbouring countries of Germany a database of 166 reliable archaeomagnetic directions has been compiled for the reference area, which is defined by a 500-km-radius circle around Gottingen. The retrieved directions come from well-dated archaeological structures and about 40 per cent of the dating relay on natural science methods such as radiocarbon, thermolumines- cence, dendrochronology dating or historical documents. From this data set a secular variation (SV) reference curve has been calculated using a bivariate algorithm, which fits a natural cubic spline based on roughness penalty to declination, inclination and time, simultaneously. The error tube surrounding this curve was obtained from Bayesian modelling of the experimental errors, which can also take stratigraphic information into account. The obtained SV reference curve for the past 2500 yr is similar to that from France, but also significant differences are seen. Comparison of the curves does not show a simple westward drift of the SV pattern. The German reference curve allows archaeomagnetic dating in the reference area and extends this dating technique to sites situated in middle Europe.

Recent achievements in archaeomagnetism in Hungary

Over the last six years the Hungarian archaeomagnetic database has been increased by the addition of 60 new directional results. At present there are a total of 159 archaeomagnetic directions with age estimates falling between BC 350 and AD 1800. Using this more extended data set a new model of the directional secular variation (DSV) was derived for Hungary (Budapest), which is considered an improvement relative to that known six years ago. The main features of the model are an almost sinusoidal variation of the inclination between 58° and 72° and a more complicated variation of the declination between −15° and 25° with several changes of sign. Comparison of this new model DSV with those derived by Daly & Le Goff for other European ‘world sites’ shows general agreement as well as important differences, such as the occurrence of negative declinations during early medieval times, which apparently are not present in the DSV for regions west of Hungary.

Archaeomagnetic results from mural paintings and pyroclastic rocks in Pompeii and Herculaneum

This work investigates the magnetic remanence associated with red pigments from murals at Pompeii and compares their directions to those of the pyroclastic rocks from the Vesuvius AD 79 eruption. The remanence of the murals is shown, using X-ray analyses, to be carried by haematite. Murals in Thermae Stabianae, known to have been painted a few years before AD 79, yield an archaeomagnetic direction ( D=1.2°, I=58.0°; α 95=5.5°) indistinguishable from that of a nearby kiln ( D=358.0°, I=59.1°; α 95=1.7°) ( Evans and Mareschal, 1989) probably last used immediately prior to the eruption. The directions are also consistent with those of fine-grained pyroclastic rocks from the eruption ( D=351.2°, I=57.9°; α 95=3.4°) and lithic and tile fragments embedded within them ( D=358.5°, I=60.4°; α 95=8.5°). Other paintings of the 1st century AD yield similar directions, with a lower statistical definition. This study shows that murals can retain their remanent magnetization for centuries and demonstrates the viability in principle of pictorial remanence as an archaeomagnetic tool.

Anomalous archaeomagnetic directions and site formation processes at archaeological sites in Israel

As part of a program to investigate archaeomagnetic secular variation in Israel and implications for archaeomagnetic dating, we have identified “anomalous” results that have yielded information about site formation processes. Stereonets, vector demagnetization diagrams, and sample location maps are most useful in examining stability of magnetization and consistency of archaeomagnetic directions with each other and with what would be expected from normal secular variation. A collection of examples is presented, including: strong and hard remanence due to vitrification, unstable magnetization due to lightning strikes, superimposed thermal magnetizations, chemical remagnetization, magnetization of a hearth fragment before falling, magnetization of a hot brick after falling, and mechanical deformation of an archaeomagnetic feature. © 1999 John Wiley & Sons, Inc.

Anomalous archaeomagnetic directions and site formation processes at archaeological sites in Israel

As part of a program to investigate archaeomagnetic secular variation in Israel and implications for archaeomagnetic dating, we have identified “anomalous” results that have yielded information about site formation processes. Stereonets, vector demagnetization diagrams, and sample location maps are most useful in examining stability of magnetization and consistency of archaeomagnetic directions with each other and with what would be expected from normal secular variation. A collection of examples is presented, including: strong and hard remanence due to vitrification, unstable magnetization due to lightning strikes, superimposed thermal magnetizations, chemical remagnetization, magnetization of a hearth fragment before falling, magnetization of a hot brick after falling, and mechanical deformation of an archaeomagnetic feature. © 1999 John Wiley & Sons, Inc.

A North American Archaeomagnetic Database.

North American archaeomagnetic data have been compiled using a relational database management system. There are currently 1875 archaeomagnetic directions and 151 paleointensity results in the database.

Archaeomagnetic directions: the Hungarian calibration curve

Abstract The first archaeomagnetic dating in Hungary was made using a reference inclination curve derived by interpolation between three inclination curves then available from continental Europe, namely from France, Bulgaria and the Ukraine. As the corresponding declination curves could not be interpolated with confidence, all three declination curves were used only for an estimation of the time interval to which the measured declinations might be assigned. As data accumulated it became feasible for dating to be made exclusively on the basis of archaeomagnetic and direct observational results for Hungary. The results now cover the last 2000 years with relatively short gaps (sixth and thirteenth centuries ad ), for which data interpolation is plausible. The variation of inclination is fairly sinusoidal and well resolved. It exhibits two maxima ( c. 70°) at approximately ad 800 and ad 1500 and three minima at about ad 300 ( c. 58°), ad 1300 ( c. 55°) and in the twentieth century ( c. 62°). Although the variation of the declination is twice as large as that of the inclination, the declination record has inherently less resolution (d D = d I /cos I > d I ). Approximately zero declination values during the first half of the first millennium ad were followed by westerly values during the sixth to eighth centuries ( c. 10°). It appears that by ad 900 the declination was already easterly and rapidly increasing until ad 1000 when it peaked with a value over 20°E. A newly discovered feature is a short oscillation beginning with 16° at ad 1300 and ending with 16.5° at ad 1600 during which the declination became westerly with a minimum of −10° in the first half of the fifteenth century. From ad 1600 on the declination rapidly decreased and described a negative half circle between about ad 1620–30 and 1950 with a minimum of −18° at ad 1800. The general pattern of the directional secular variation for Hungary is in agreement with that for France, Sicily, Britain, the Ukraine and the Balkans.

A method for correcting geographically separated remanence directions for the purpose of archaeomagnetic dating

SUMMARY Spatial variations in the geomagnetic field must be taken into account if secular variation master curves and directional magnetic dates are to be optimized. Two methods for relocating remanence vectors have been proposed and in this paper their relative accuracies are compared using a numerical model based on the present-day field. A method which converts archaeomagnetic directions via a virtual geomagnetic pole is shown to be the more efficient transformation. For an‘archaeomagnetic region’the size of the British Isles, (900 km radius), the maximum error in relocating vectors to a central location is predicted to be of the order of 1.2°.

Archaeomagnetic evidence for the age and duration of firing of mediaeval hearths from Coffee Yard, York

SUMMARY Archaeomagnetic directions in two hearths at Coffee Yard, York, have provided evidence for their age and mode of use. The earliest structure was last fired in approximately 1440 AD and the second about 90 years later. The depth of remagnetization (>16 cm) has been simulated using a simple analytical model which suggests that one hearth was fired continuously for at least 56 h.

Secular variation of archaeomagnetic direction in the American Southwest, A.D. 750–1425

Oriented archaeomagnetic samples were collected from 158 in situ features at 33 archaeological sites in the southwestern United States. Seventy‐three independently dated features were used for analysis of secular variation. A moving‐window technique with outlier rejection was developed to compute a smoothed secular variation curve. This technique incorporates weighted Fisher statistics to account for imprecision in both the age of remanence acquisition and the direction of magnetization for each feature. A mean direction and measure of dispersion is generated for each window of time. The secular variation record covers the period A.D. 750–1425, during which time the virtual geomagnetic pole (VGP) moves in a generally counterclockwise fashion from 86°N, 83°E at A.D. 750 to the lowest latitude position of 74°N, 192°E at A.D. 1075, and then to 85°N, 236°E at A.D. 1425. The median rate of VGP movement is 0.036° yr−1, similar to present‐day rates. The Southwestern record shows close correspondence to other North American archaeomagnetic, limnomagnetic, and speleomagnetic records. The archaeomagnetic record can serve as the master curve for the derivation of archaeomagnetic dates.

Archaeomagnetic investigations of geomagnetic secular variations

Bulgarian archaeomagnetic directions and intensities show a satisfactory agreement with those for Ukraine and Paris. Bauer plots of declination against inclination show both clockwise and anticlockwise looping through the last eight millennia. The necessity for accurate dating of samples studied is stressed.

Ancient Geomagnetic Field Intensities?I Historic and Archaeological Data: Sets H1-H9

Regular lists of palaeomagnetic directional data, which are of great benefit to other workers in the field, have been compiled by Irving (1960-1963, 1965) and Irving & Stott (1963). Greater attention is now being paid to determinations of ancient field intensities; hence it is thought that regular tabulation of the results of these would likewise be useful. This first list contains ancient field intensities covering the historic and archaeological time scale that are available up to 1966 December. For the present purpose historic and archaeological time has been defined as the past 10 OOO years; and all results up to this limit have been included. A similar compilation of geological time scale field intensity data is at present being prepared. The results are presented in the accompanying Table. The data from each original author or group of authors have been given a reference number prefixed with the letter H (historic and archaeological). Each individual field intensity determination has also been allocated a number which follows the set number after a dash. Thus, for example, H3-1 (column 2) represents the first field intensity value in the third set of historic and archaeological data (in this case Russian). Ages of the materials used are given in column 3 where dates AD have been expressed as positive numbers and dates BC as negative numbers. The ' age ' of a sample as here recorded refers to the age of the natural remanence and not necessarily to the physical age of the sample. For example, the ' age ' of a piece of pottery will refer to the time at which it was fired, and the ' age ' of a furnace will generally refer to the last time it was used. Data within each set are arranged in order of increasing age. Column 4 shows the material localities as precisely as the original publications will allow; and column 5 gives the type of material used in each determination. Experimental methods are recorded in column 6, where Th refers to the technique developed by Thellier & Thellier (1959) for materials possessing TRM and De refers to artificial sedimentation under water in a laboratory field for materials possessing DRM. The main tests of reliability and precautions taken to obtain reliable results are listed in column 7 where the code letters used refer to the list beneath the Table. The convention used in palaeomagnetism whereby specimens are cut from separately collected samples has also been used here even though samples are frequently unoriented. Numbers of samples and specimens used are given in columns 8 and 9 respectively, but where no distinction has been made between the two in the originals, numbers are entered in column 8. The basic field intensity data are listed in columns 10-13, where applicable, and from these the ancient dipole moments and equatorial field intensities have been calculated assuming an essentially dipolar field. Generally archaeomagnetic directions have not been presented in the originals, and hence in each case the equatorial * Publication authorized by the Director, U.S. Geological Survey.