Skip to navigation – Site map

HomeNuméros45-12. Methods and innovationsMineral-Magnetic Characterization...

2. Methods and innovations

Mineral-Magnetic Characterization as a Key to Explain Differences in Magnetic Contrast and Improve Archaeological Interpretation

An Example of the Roman Site at Auritz/Aurizberri, Navarre
Ekhine Garcia-Garcia, Hana Grison, Neli Jordanova, Philippe De Smedt and Eneko Iriarte
p. 161-164

Abstract

– New approach to mineral-magnetic characterization.

– Evaluation of factors influencing geophysical results.

– Origin of a fringe without magnetic contrast explained by waterlogging.

Top of page

Full text

This research was supported by a STSM Grant from COST Action SAGA (CA17131), supported by COST (European Cooperation in Science and Technology) and the INTER-EXCELLENCE program of the Ministry of Education, Youth and Sports of the Czech Republic (MEYS), grant No. LTC19029. The authors wish to thank the many volunteers from the Aranzadi Science Society and the involved municipalities. Special thanks to Eder Martinez, Xabier Errea and Anika Villanueva, who contributed to data collection.

Introduction

1Geophysical and geoarchaeological methods can be used to delimit the extent and characterise the nature of settlement in the past. To this end, a magnetic survey was conducted at the site of the ancient Roman town of Zaldua (Auritz, Navarre). Although the results of this survey provided a firm basis for a preliminary archaeological assessment of the site (Garcia-Garcia et al., 2016), in some parts of the study area the magnetic response was difficult to interpret. This is most striking in the north of the main area, where a narrow band without clear magnetic variation separates two zones with clear evidence of intensive occupation. The same result was reproduced in the Ground Penetrating Radar (GPR) and Electromagnetic Induction (EMI) surveys performed in this area. The absence of clear responses is believed not to reflect archaeological reality but the reason for this contrast-free fringe is unknown, and remains one of the archaeological questions of the site.

2A Short Time Scientific Mission (STSM), conducted as part of the ongoing COST Action 17131 – SAGA program, brought together specialists from the fields of geophysics, archaeology, environmental magnetism and soil science from four countries with the main aim to improve in situ measurements using soil magnetic characteristics to understand the factors affecting geophysical results, in order to better interpret archaeological features. The presentation here showcases the investigations focused on the fringe area described above. The objective was to compare the magnetic properties of the fringe with those of settled areas, the expectation being that it could help to establish the origin of the fringe and lead to an understanding of why there are no archaeological remains present in this area.

Site description and methodology

3The extensive remains of the ancient Roman city of Zaldua in the north of the Iberian Peninsula, functioning from the 1st century AD to its abandonment in the 4th century AD, was discovered in 2012 and geophysical methods have been used extensively to delimit and characterise the settlement (Garcia-Garcia et al., 2017). Archaeological excavations have been ongoing annually since 2015 and the partial results have been compared to the results of previous geophysical investigations. The main area of about 4.5 ha has been shown to be densely occupied and separated from a small secondary area (0.45 ha) by a narrow fringe devoid of a clear magnetic contrast (Fig. 1).

4

Figure 1. The gradiometer response map (-7 nT black, +9 nT white) and the magnetic susceptibility map from the EMI survey (in-phase HCP response, highest positive sensitivity 0–0.5 m), with core positions indicated. The location of the fringe zone is marked on a general map of the settlement.

Figure 1. The gradiometer response map (-7 nT black, +9 nT white) and the magnetic susceptibility map from the EMI survey (in-phase HCP response, highest positive sensitivity 0–0.5 m), with core positions indicated. The location of the fringe zone is marked on a general map of the settlement.

5The study was performed on a set of samples from 13 cores, 1-2 m long, drilled with a handheld coring machine. The samples were made from 1 to 10 cm thick sections, predetermined by the archaeologically identified stratigraphic layers. Laboratory measurements were done at the Institute of Geophysics of the Czech Academy of Science, Prague, Czech Republic.

6For all samples, magnetic susceptibility (MS) was measured at low (976 Hz) and high frequency (15616 Hz), using Kappabridge MFK1-FA (AGICO, Brno, Czech Republic). Frequency-dependent magnetic susceptibility (χFD) was calculated as a relative and absolute change of MS obtained for low and high frequencies. This approach allows the relative content of pedogenic superparamagnetic (SP) grains and discriminating potential archaeological layers to be estimated (e.g., Dearing, 1996; Fassbinder, 2015; Hrouda, 2011). Bi-plots of the MS and χFD permitted a comparison of the relations between different magnetic contributions in the investigated cores.

7In order to identify magnetic grain size and mineral composition, the hysteresis and remanence characteristics were determined using a Vibrating-sample magnetometer (ADE Corporation, VSM EV9 VSM 2900). The measurement cycle was composed of the hysteresis loop, acquisition of the Isothermal Remanent Magnetization Curve (IRM) and the DC demagnetisation remanence curve (DCD). Data were subsequently evaluated in a Day plot (Day et al., 1977).

Results

8Data from the MS measurement of cores and geophysical data, gradiometric and EMI, are very consistent (Fig. 2). Cores located in the area with no magnetic contrast show the lowest MS values (cores 16, 20 and 21; average value of 16 × 10-8 m3/kg). The exceptionally high values of MS in some samples (178 × 10-8 m3/kg) can be related to small inclusions of slag or pottery. Cores located in areas with high magnetic contrast show the highest MS values (P1004, P1006, P27). In most cases, increased MS values (range of 324–2026 × 10-8 m3/kg) correspond to the results obtained for samples from the archaeological deposits. The rest of the investigated cores show “intermediate” MS values (in the range of 7–270 × 10-8 m3/kg, average 57 × 10-8 m3/kg). The bi-plot of the absolute change of χFD and MS shows that the bulk of cores from the area follows the same magneto-mineralogical trend (Fig. 2b). This is interpreted as representing similar pedogenic processes going on throughout this area (Evans & Heller, 2003; Jordanova, 2016). The exception is core P1004, which has a notably lower SP fraction indicating a different origin of the magnetic minerals.

9

Figure 2. (a) Vertical distribution of magnetic susceptibility along cores from Area 2 (horizontal axis broken); (b) bi-plot of the absolute change of frequency dependence and mass magnetic susceptibility.

Figure 2. (a) Vertical distribution of magnetic susceptibility along cores from Area 2 (horizontal axis broken); (b) bi-plot of the absolute change of frequency dependence and mass magnetic susceptibility.

10The Day plot in Figure 3 summarizes magnetic grain size of selected samples in cores P15, P16, P27, P1004 and P1006. The hysteresis curves were corrected for the paramagnetic part by subtracting the linear part of the loop from the whole signal. Core P16, positioned in a poorly magnetic area, shows a mixture of different minerals. The hysteresis curve of P16-38, for instance, is not closed, indicating that it is influenced by a high coercivity mineral, such as goethite. Sample P16-04 is located in the topsoil horizon and shows large, multidomain magnetic grain size, most probably of anthropogenic origin (modern pollution). Samples of the “strongly magnetic” core P1004 are located in the pseudo-single domain (PSD) area; therefore, only one dominant mineral type is expected (Dunlop, 2002).

11

Figure 3. Day-plot and hysteresis loops of selected samples from Area 2 (vertical scale adapted to each loop).

Figure 3. Day-plot and hysteresis loops of selected samples from Area 2 (vertical scale adapted to each loop).

Discussion and conclusions

12Analysis of the mineral magnetic characterization of the samples leads to the conclusion that the provenance of magnetic minerals is similar for all of the cores. Indeed, they show a similar frequency dependence of magnetic susceptibility, suggesting similar pedogenic influence. Therefore, an external contribution of magnetic minerals should be discarded, and areas without magnetic contrast correspond to areas where the pedogenesis did not result in significant magnetic enhancement. Core P1004 is different. Here, an external contribution of magnetic minerals should be considered. Indeed, the gradiometer response map indicates thermally enhanced areas, which could indicate industrial activity.

13However, the hysteresis parameters show a notable difference in core P16, located inside the fringe without magnetic contrast. The high-coercivity component combined with low magnetic susceptibility suggest the existence of a mixture of magnetic minerals with different coercivities (Maxbauer, 2017). By contrast, the presence of a single dominant magnetic mineral is to be expected in the other cores.

14Waterlogging in the fringe could be considered as an explanation for these observations. It would create an environment, in which iron (hydro)oxides could be transformed into hard-coercivity goethite instead of magnetite (Cornell & Schwertmann, 2003). From an archaeological point of view, a higher water table level would explain why there are no construction remains in the fringe. Therefore, the preliminary conclusion in this case is that the fringe reflects inconvenient terrain rather than an intentional marking of a separation between two areas of occupation.

Top of page

Bibliography

Cornell, R., Schwertmann, U., 2003. The Iron Oxides: Structures, Properties, Reactions, Occurrences, and Uses, 2nd ed., Wiley-VCH, Weinheim.

Day, R., Fuller, M., Schmidt, V.A., 1977. Hysteresis properties of titanomagnetites: Grain-size and compositional dependence. Physics on the Earth and Planetary Interiors, 13(4): 260-267. DOI: 10.1016/0031-9201(77)90108-X (1977).

Dearing, J.A., Dann, R.J.L., Hay, K., Lees, J.A., Loveland, P.J., Maher, B.A., O’grady, K., 1996. Frequency-dependent susceptibility measurements of environmental materials. Geophysical Journal International, 124: 228-240. DOI: 10.1111/j.1365-246X.1996.tb06366.x.

Dunlop, D.J., 2002. Theory and application of the Day plot (Mrs/Ms versus Hcr/Hc) 1. Theoretical curves and tests using titanomagnetite data. Journal of Geophysical Research, 107, B3, 2056. DOI: 10.1029/2001JB000486.

Fassbinder, J.W.E., 2015. Seeing beneath the farmland, steppe and desert soil: Magnetic prospecting and soil magnetism. Journal of Archaeological Science, 56: 85-95.

Evans, M., Heller, F., 2003. Environmental Magnetism: Principles and Applications of Enviromagnetics. Academic Press, San Diego, CA.

Garcia-Garcia, E., Mtz. Txoperena, J.M., Sala, R., Aranburu, A., Agirre-Mauleon, J., 2016. Magnetometer Survey at the Newly-discovered Roman City of Auritz/Burguete (Navarre). Results and Preliminary Archaeological Interpretation. Archaeological Prospection, 23(4): 243-256. DOI: 10.1002/arp.1537.

Garcia-Garcia, E., Andrews, J., Iriarte, E., Sala, R., Aranburu, A., Hill, J., Agirre-Mauleon, J. 2017. Geoarchaeological Core Prospection as a Tool to Validate Archaeological Interpretation Based on Geophysical Data at the Roman Settlement of Auritz/Burguete and Aurizberri/Espinal (Navarre). Geosciences, 7(4): 104. DOI: 10.3390/geosciences7040104.

Hrouda, F., 2011. Models of frequency-dependent susceptibility of rocks and soils revisited and broadened. Geophysical Journal International, 187: 1259-1269. DOI: 10.1111/j.1365-246X.2011.05227.x.

Jordanova, N., 2016. Soil Magnetism – Applications in Pedology, Environmental Science and Agriculture. Elsevier, London.

Maxbauer D. P., Freiberg, J. M., Fox, D. l., Nater, E. A., 2017. Response of pedogenic magnetite to changing vegetation in soils developed under uniform climate, topography, and parent material. Scientific Reports, 7. DOI: 10.1038/s41598-017-17722-21.

Top of page

List of illustrations

Title Figure 1. The gradiometer response map (-7 nT black, +9 nT white) and the magnetic susceptibility map from the EMI survey (in-phase HCP response, highest positive sensitivity 0–0.5 m), with core positions indicated. The location of the fringe zone is marked on a general map of the settlement.
URL http://journals.openedition.org/archeosciences/docannexe/image/9280/img-1.jpg
File image/jpeg, 764k
Title Figure 2. (a) Vertical distribution of magnetic susceptibility along cores from Area 2 (horizontal axis broken); (b) bi-plot of the absolute change of frequency dependence and mass magnetic susceptibility.
URL http://journals.openedition.org/archeosciences/docannexe/image/9280/img-2.jpg
File image/jpeg, 425k
Title Figure 3. Day-plot and hysteresis loops of selected samples from Area 2 (vertical scale adapted to each loop).
URL http://journals.openedition.org/archeosciences/docannexe/image/9280/img-3.jpg
File image/jpeg, 399k
Top of page

References

Bibliographical reference

Ekhine Garcia-Garcia, Hana Grison, Neli Jordanova, Philippe De Smedt and Eneko Iriarte, “Mineral-Magnetic Characterization as a Key to Explain Differences in Magnetic Contrast and Improve Archaeological Interpretation”ArcheoSciences, 45-1 | 2021, 161-164.

Electronic reference

Ekhine Garcia-Garcia, Hana Grison, Neli Jordanova, Philippe De Smedt and Eneko Iriarte, “Mineral-Magnetic Characterization as a Key to Explain Differences in Magnetic Contrast and Improve Archaeological Interpretation”ArcheoSciences [Online], 45-1 | 2021, Online since 16 August 2021, connection on 29 March 2024. URL: http://journals.openedition.org/archeosciences/9280; DOI: https://doi.org/10.4000/archeosciences.9280

Top of page

About the authors

Ekhine Garcia-Garcia

Corresponding author, SOT Archaeological Prospection, Barcelona, Catalunya

By this author

Hana Grison

Institute of Geophysics of the Czech Academy of Sciences, Prague, Czech Republic

Neli Jordanova

National Institute of Geophysics, Geodesy and Geography, Bulgarian Academy of Sciences, Sofia, Bulgaria

Philippe De Smedt

Department of Environment, Department of Archaeology, Ghent University, Ghent, Belgium

By this author

Eneko Iriarte

Universidad de Burgos, Spain

Top of page

Copyright

CC-BY-NC-ND-4.0

The text only may be used under licence CC BY-NC-ND 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search