The French archeomagnetic directional database for the past two Millennia and its application in geomagnetism and archeology M. Le Goff (1), N. Warmé (2),

The French archeomagnetic directional database for the past two Millennia

and its application in geomagnetism and archeology

M. Le Goff (1), N. Warmé (2), Y. Gallet (1), and A. Genevey (3)

(1) Institut de Physique du Globe de Paris, CNRS UMR 7154, Laboratoire de Paléomagnétisme, Paris, France, (2) Institut National de Recherches Archéologiques Préventives, France,

(3) Centre de Recherche et de Restauration des Musées de France, CNRS UMR 171, Paris, France

The last article of Pr Thellier (1904-1987), in PEPI-1981, shows the directional variation curve of the Earth Magnetic Field, in France, between

0 and 1800AD, obtained with only 63 dated structures selected in a set of more than 200

ovens and 50 sets of bricks .

13 years later, Ileana Bucur, in PEPI 1994, upgraded the French data base with 110 new data ; 67 being archeologically dated.

Including the Thellier results, she computed this French reference curve with more than 120 dated directions.

Since her retirement, in 1997, a young team has continued the

directional data acquisition in the Saint-Maur Laboratory, with until now, nearly 400 new structures

measured

Over the past two millennia, the directional variations stand in a small

aperture cone of less than 20°. Recovering these variations thus requires

a high precision in measurement

N

EW

I = 0

I = 3

0°I = 6

0°

N

The expected parasitic magnetizations are :- partial or total TRM due to refiring- IRM due to lightning effect- and most often, significant VRM acquisition

TRM directions can be biased by magnetic anisotropy

Mechanical effects, such as structure dislocation or tilting, contribute to disperse the directional results

ABOUT ACCURACY AND RELIABITY

The methodology used in our laboratory,

not « standard », follows the Thellier’s one

Extensive sampling of the fired structures, avoiding as much as possible walls, or clay mixed with pieces of bricks or tiles.

Big samples (12cm in size) precisely oriented, without further cutting or induration.

Systematic measurement of magnetic viscosity.

AF demagnetization in selected cases.

« Big » sample inductometer

Sampling and Magnetic measurements

« Big » sample AF demagnetizer

Sampling Domestic Ovens in JOSSIGNY

Horizontal surface and sun orientation

12 cm

Inverse Measurem

ent

Z

X

Y

TRM

± VRM

Samples with a Viscosity ratio v (VRM/TRM) > 10 % are discarded

No

rth L

ab d

irection

Viscosity ExperimentsViscosity Experiments (Thellier, 1981)

By Vectorial operation

VRM and TRM

2 - Samples in “Inverse” position during at least 2 weeks

Direct Measurem

ent

1 - Samples in “Direct” position during at least 2

weeks

Two Examples of directional results with v=7%

Structure : Rungis 1277Mean TRM direction: D=-18.5°, I=64.7°

a95=0.34°, N=16, k=10860

Structure : Villeparisis 4069Mean TRM direction: D=5.4°, I=67.7°

a95=1.82°, N=8, k=743

VRM VRM

TRMTRM

Declination (°) Declination (°)

Inc

lin

ati

on

(°)

Inc

lin

ati

on

(°)

0

5

10

15

20

25

<0.2< 0.

4< 0.

6< 0.

8< 1

< 1.2

< 1.4

< 1.6

< 1.8

< 2< 2.

2< 2.

4< 2.

6< 2.

8< 3

< 3.2

< 3.4

ou plu

s...

Alpha 95 (°)

Fre

qu

ency

(%

)

0

20

40

60

80

100

363 results ; 95% < 1.6°

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

0 5 10 15 20 25 30 35Number of Samples

Alp

ha

95

(°)

363 new results

0

2

4

6

8

10

12

14

16

18

20

<1 <2 <3 <4 <5 <6 <7 <8 <9 <10 <11 <12 <13 <14 <15 <16 >16Mean Viscosity (%)

freq

uen

cy (

%)

0

20

40

60

80

100

in which30 results

with v>10%were

discarded

393 new collected sites

Alpha 95histogram

Viscosityhistogram

-200 0 200 400 600 800 1000 1200 1400 1600 1800 2000

384 with mean date < 1000 AD 149 with mean date > 1000 AD

All Resultswith « true » or indicative

datation(533)

Some examples of other European databasesfrom archeolabs using standard methodology

All directionsreduced in Paris

Path 1300-1750 from Lübeck

363 new results from France

NEW RESULTS

THELLIER BUCUR

Know n Date

Unknow n Date

Know n Date

Unknow n Date

Léry, 28 results

Marines 21 results

Villiers le Sec 42 results

Saint-Pathus 59 results

The Geographical Distribution

is very uneven through-out the country.

The Ile de France region contains more than half of the sampling

locations

These 393 new sampled structures represent more than 4400 kg-sized specimens.

-200 0 200 400 600 800 1000 1200 1400 1600 1800 2000

-200 0 200 400 600 800 1000 1200 1400 1600 1800 2000

-200 0 200 400 600 800 1000 1200 1400 1600 1800 2000

130 dated(in Bucur 94)

163 new datedstructures

Resulting ina total of 293

dated structures

CONSIDERING NOW ONLY THE WELL DATED RESULTS...

50

55

60

65

70

75

80

0 200 400 600 800 1000 1200 1400 1600 1800 2000

-30

-20

-10

0

10

20

30

40

0 200 400 600 800 1000 1200 1400 1600 1800 2000

The updated French reference curve is calculated using the self adaptable, variable width sliding

windows (LeGoff et al, 2002)

Inclination

Declination

293 dated structures

20 < windows < 100 years

WHICH MODEL FITS ?

The most precise directions (small alpha95) are alsothe less scattered around the mean curve

82 data with Alpha95 < 0.75°

82 data with Alpha95 > 0.75°

(selection of age incertitude <= ±100 years)

CONCLUSION

• The Thellier’s methodology combined with suitable instrumentation provides highly precise directional data.

• The French reference database, which now contains 293 dated directions, is currently used as an accurate dating

tool for archeological purpose.

Our future article (in preparation) will bring more information !

Thank you

The Saint-Maur Laboratory

Examples of archeological sites with a long and continuous

occupation

Excavating a ditch Digging the kiln in silt

Heating the kiln Cooking meals

Abandonment of the kiln And digging a new one

and, now, before sampling

Haut Empire millieu Ier BC ; fin IIIème -50 -> 300

antiquités Tardive (ex Bas-Empire) déb IV, fin Vème (476, fin Emp Rom) 300 -> 500

Haut Moyen Age (HMA) Vème, Xème 450 -> 1000

dont Mérovingien Vème, VIIIème 450 -> 750

Carolingien VIIIème, Xème 750 -> 1000

Moyen Age Classique (ex Médiéval) XIème, XIIIème 1000 -> 1300

Bas Moyen Age (BMA) XIVème, XVème 1300 -> 1500 (1492)

Moderne fin XVème, fin XVIIIème 1500 -> 1800 (1789)

Contemporain fin XVIIIème -->> 1800 (1789) ->>>>>