Characterization of Gallo-Roman roads in northern France using micromorphological methods

Accepted Manuscript Characterization of Gallo-Roman roads in northern France using micromorphological methods Marie-Caroline Charbonnier, Cécilia Cammas PII:

S1040-6182(17)31040-6

DOI:

10.1016/j.quaint.2018.05.010

Reference:

JQI 7420

To appear in:

Quaternary International

Received Date: 16 July 2017 Revised Date:

2 April 2018

Accepted Date: 10 May 2018

Please cite this article as: Charbonnier, M.-C., Cammas, Cé., Characterization of Gallo-Roman roads in northern France using micromorphological methods, Quaternary International (2018), doi: 10.1016/ j.quaint.2018.05.010. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1

Characterization of Gallo-Roman roads in Northern France using micromorphological

2

methods.

3 4 Marie-Caroline CHARBONNIER, INRAP Grand-Est, 38 rue des Dâts, 51520 Saint Martin-

6

sur-le-Pré ; Unité de Micromorphologie DMOS-AgroParisTech, 1 Avenue Lucien

7

Bretignières, 78850 Thiverval-Grignon.

8

[email protected]

SC

RI PT

5

9

Cécilia CAMMAS, INRAP Centre-Ile-de-France, 36 avenue Paul Vaillant-Couturier, 93120

11

La Courneuve ; Archéologie des Sociétés Méditerranéennes, UMR 5140, 390 route de Pérols,

12

34970 Lattes ; Unité de Micromorphologie DMOS-AgroParisTech, 1 Avenue Lucien

13

Bretignières, 78850 Thiverval-Grignon.

14

[email protected]

17

Abstract

EP

16

TE D

15

M AN U

10

The road network plays an important role in the birth and evolution of town planning and

19

structuring of the landscape. It presents unique and original dynamics that encompass both its

20

status and uses. Whatever their reality in the field, stratigraphic sequences are poorly studied

21

using geoarchaeological methods. To document the formation processes of these road

22

networks, micromorphological analyses were undertaken on a street in a Roman secondary

23

town, Sains-du-Nord, on a road in a Roman civitas capital, Metz Divodurum, and on a lane

24

created in medieval times in Lieusaint. Analyses of the Sains-du-Nord street and the Paille-

25

Maille (Metz) road revealed intense vehicle traffic where traffic-related repairs and traffic

AC C

18

ACCEPTED MANUSCRIPT layers were differentiated. Analyses of the Lieusaint site, enabled specialists to characterize

27

the processes and layers resulting from vehicle traffic. Pedofeatures linked to passage and

28

traffic presented in this paper, can be used to identify such types of space in other case

29

studies. Finally, results show that the use of thin sections of archaeological layers can reveal

30

not only the surface history, but also the original characteristics of traffic circulation space.

31

The microscopic sequences are presented here as descriptive models of formation processes.

32

The monophasic model appears to be the result of ongoing maintenance attributable to the

33

status or continued use of the street (Sains-du-Nord) or lane (Lieusaint). Polyphasic and

34

polycyclic models reflect rhythmic use and allow reconstruction of traffic levels (Paille-

35

Maille (Metz) road).

M AN U

SC

RI PT

26

36 37

Keywords: road, street, lane, vehicle traffic, formation processes, micromorphology.

38 1. Introduction

TE D

39

Roads are one of the most famous features of the Roman Empire. The road network, built and

41

maintained by Rome, was a powerful vehicle for political and administrative control, and also

42

for cultural and commercial development (Grenier, 1985; Chevallier, 1997; Quérel, 2008, p.

43

85). In Gallo-Roman agglomerations, the road network is a decisive element in the structuring

44

of urban space (Ballet et al., 2008). In Roman Gaul, the layout of traffic circulation axes

45

(roads, streets, sidewalks) corresponds to a specific organization. Indeed, on the stereotypical

46

map of a city, town planning is organized around two main axes: the decumanus maximus

47

(oriented east-west) and the cardo maximus (oriented north-south) (e.g. Plinius, Vitruvius).

48

Axes of circulation between the different entities that make up the city, street and sidewalks,

49

are integral elements of the urban landscape (Ballet et al., 2008, p. 51). Thus, the construction

AC C

EP

40

ACCEPTED MANUSCRIPT of roads and their continuity or modifications through time are significant in terms of the

51

evolution of the urban fabric (Ballet et al., 2008).

52

In this paper, “road” signifies an important axis of circulation, which allows vehicle traffic

53

between two given geographical points, usually two agglomerations (Chevallier, 1997, pp.

54

98-106). “Street” is a specific term of the urban environment, referring to an axis of

55

circulation accessible to all in an urban context, even if it is an extension of a road (Statius,

56

books IV-V; Ballet et al., 2008, p. 63). “Lane” has no urban connotation; it represents the axis

57

of circulation in a rural area, usually poorly built (Brunet, 1992, pp. 102-103), e.g. dirt road.

58

Most current studies focus on “roads” and the network pattern of streets/roads in the

59

urban/rural context (Duval, 1989, pp. 757-764; Querel, 2008, pp. 85-122; Robert and Verdier,

60

2015; Cammara Munoz and Revuelta Pol, 2016; van Lanen et al., 2016; van Lanen and

61

Pierik, 2017). These studies are based on historic documents and maps, and archaeological

62

site maps.

63

Over the past decade in France, a new direction of research has focused on the archaeological

64

and pedosedimentary features related to roads and streets. In the wake of the collective

65

research programme called DYNARIF [(Dynamique et Résilience des Axes de Circulation en

66

Ile-de-France, Sandrine Robert and Nicolas Verdier (dir)], a standard vocabulary was

67

proposed to describe traffic axes (Robert and Verdier, 2015, p. 33) (Fig. 1). Although

68

DYNARIF aimed to investigate the archaeological reality of roads and streets, the limitations

69

of archaeological tools became evident during the project. These limits related mainly to

70

characterization of the layers of circulation axes, their composition and origin, as well as the

71

differentiation of layers related to construction from those related to use (trampling, rolling)

72

(Charbonnier and Cammas, 2015; Robert and Verdier, 2015).

73

Geoarchaeology, and more specifically micromorphology, is applied to gain an understanding

74

of archaeological stratigraphy, site formation processes and the dynamics of soils and

AC C

EP

TE D

M AN U

SC

RI PT

50

ACCEPTED MANUSCRIPT landscape from the city district to territory (Courty et al., 1989; Courty, 2001; Usai, 2001;

76

Goldberg and Macphail, 2006; Cammas and Wattez, 2009; Macphail and Goldberg, 2010).

77

Nevertheless, streets and roads remain underrepresented in studies of soil micromorphology

78

(Cammas and Wattez, 2009, p. 209; Gebhardt and Langhor, 2015, p. 32) as the sedimentary

79

expression of traffic remains poorly investigated in the archaeological context (Charbonnier

80

and Cammas, 2015; Gebhardt and Langhor, 2015, pp. 32-33). A further problem is confusion,

81

due to lack of research, between tracks caused by agricultural implements and rolling tracks

82

or ruts due to traffic. This can be addressed through micromorphology (Gebhardt and

83

Langhor, 2015, pp. 31-38; Deák et al., 2017, pp. 233-259; Nicosia and Stoops, 2017, pp. 281-

84

297). A better knowledge of the impact of traffic on soils and sediments could help clarify

85

such problems. Broadly speaking, investigations should address the characterization of traffic

86

areas, as well as their method of construction and organization, in order to restore the

87

formation processes of traffic space, and their relation to the urban dynamic and landscape

88

structure.

89

The main objective of this paper is to provide approaches to the identification and

90

characterization of traffic features typical of roads, streets and lanes with the help of

91

micromorphology. In order to reconstruct past uses of axes of circulation, the present paper

92

focuses on microscopic features related to the degree of compaction and redoxic processes

93

associated with the presence and intensity of vehicle traffic, and the sedimentary formation

94

processes of traffic.

SC

M AN U

TE D

EP

AC C

95

RI PT

75

96 97

2. Material and method

98 99

2.1. Context and sampling

ACCEPTED MANUSCRIPT 100 Three sites are presented in this study: Sains-du-Nord (Nord), Metz (Moselle) and Lieusaint

102

(Seine-et-Marne) (Fig. 2). They are situated in different geological pedo-sedimentary

103

contexts, and present different archaeological features. Their study allows evaluation of the

104

role of micromorphology within the main archaeological questions related to the nature and

105

continuity of the Gallo-Roman road network up to the Medieval period in North-East France.

106

The first two case studies are located in an urban context, in a secondary town (Sains-du-

107

Nord) and a civitas capital (Metz), the third is a rural lane (Lieusaint).

SC

RI PT

101

108

110

M AN U

109

2.1.1. The Roman street of Sains-du-Nord

111

- Study area and archaeological context

113

The Sains-du-Nord commune (Fig. 2) is situated on an interfluve between two tributaries of

114

the Sambre river, the Helpe Majeure in the north and the Helpe Mineure in the south. The

115

town developed on the southeastern slope of a small-enclosed valley (the Rieux Wiart valley)

116

in the Helpe Majeure basin (Fig.2). Geologically this region is composed of paleozoic shale

117

covered by sedimentary cretaceous marls and eocene sand and sandstone (Fig. 2) (Neaud,

118

2015).

119

An archaeological excavation undertaken by the Institut national de recherches

120

archéologiques préventives (Inrap) in 2013 revealed a stretch of streets and the remains of

121

living quarters of an old district of the secondary town of Sains-du-Nord. These discoveries

122

enabled researchers to determine the chronology of the agglomeration: it was founded in the

123

early Augustan period, developed during the 1st century, reached its apogee in the 2nd and 3rd

AC C

EP

TE D

112

ACCEPTED MANUSCRIPT centuries AD, and seems to have been abandoned in the 4th century AD (Neaud, 2014; Neaud,

125

under press)

126

The space studied is in the southeastern area of the excavation (Fig. 3a), is part of a domestic

127

dwelling bordered by a 2.60 m-wide street (Fig. 3b). It was thoroughly excavated.

128

Micromorphological analysis of the site aimed to identify the means of installation,

129

construction and maintenance of the street, to facilitate an understanding of its role and status

130

within the site.

RI PT

124

SC

131 - Sampling

133

Field observations showed the street to be formed of a layer of Tournai limestone,

134

interspersed locally with red tile and pottery fragments lying on shale bedrock. Samples were

135

taken in the west-east section (Fig.3a), and the entire section was sampled from the top to the

136

geological shale layer at the bottom. To test the spatial heterogeneity of pedosedimentary

137

facies, two samples were taken in a homogeneous traffic area (1040-1; 1040-2) and one in a

138

rut (1058) (Fig. 3b).

141 142

TE D

EP

140

2.1.2. Paille-Maille (Metz) road

AC C

139

M AN U

132

143

- Study area and archaeological context

144

Paille-Maille (Metz) road is at the foot of the Cuesta, "the coasts of Moselle". The soil

145

consists of Taorcian marls covered by Bajocian limestone. The city is in the Moselle valley

146

(Fig. 2) which is filled by recent fluvial deposits composed of thin layers of more or less

147

sandy silt, covering about ten metres of sand, gravel and siliceous pebbles coming from the

148

Vosges mountains (Fig. 4).

ACCEPTED MANUSCRIPT In 2012, an archaeological excavation carried out by the Metz-Métropole Archaeological

150

Service in an existing road, Paille-Maille (Pontiffroy district) (Fig. 4) revealed a section of the

151

Metz-Trier road that follows the left bank of the Moselle. The Roman town was built on the

152

celtic oppidum of Mediomatrics, known under the Latin name of Divodurum. Since 1983, half

153

a dozen excavations have helped determine the chronology and outline the main

154

organizational features of the Pontiffroy district (Fig. 4). As a result of these excavations, this

155

area of the town is well documented as regards Antiquity and the beginning of the Middle

156

Ages (Brkojewitsch, 2017). The Pontiffroy district developed about the last decade of the 1st

157

century BC, around the Roman Metz-Trier road. This portion of the Consular Road between

158

Metz and Trier is indicated on the Peutinger Table and in the Itinerary of Antoninus, and is

159

probably part of Agrippa's primary network (Chevalier, 1997; Brkojewitsch, 2017). Wooden

160

palisades to the west of the road delimited a funerary area with tombs dating from the first

161

two stages of the road (Fig. 4). Subsequently, a residential area was built on both sides, dated

162

to the first half of the 1st century AD. In the field, the axis of traffic consists of 6 construction

163

stages ranging from the second half of the 1st century AD to the Medieval Period (Fig. 5)

164

(Brkojewitch, 2017). Construction stages IV-VI comprise a superimposition of sand and

165

pebble layers covered by a dense stack of calcareous blocks. The earlier stages (I to III) are

166

composed of finer pebbles. The site provided an opportunity to investigate the significance of

167

these layers with reference to the unanswered questions posed in the DYNARIF project: Do

168

the interfaces or layers between coarse backfill or finer material correspond to traffic,

169

construction, maintenance, or even abandonment?

AC C

EP

TE D

M AN U

SC

RI PT

149

170 171

- Sampling

172

The 2.60 m-thick section (Fig. 4; Fig. 5) was composed of very coarse and very hard deposits.

173

At this point in the study, archaeological scrutiny and micromorphological sampling focused

ACCEPTED MANUSCRIPT 174

on the different limits between coarse deposits interpreted as traffic layers, such as an abrupt

175

layer (log 2.1, road IV), a 5 cm-thick sandy layer (log 2.2, road IV) (end of the 1st century

176

AD), and a micro layered sequence (log 2.3 haut et bas, road II) dating to the 1st quarter of the

177

1st century AD).

179 180

2.1.3. The Lieusaint lane

181

SC

RI PT

178

- Study area and archaeological context

183

The third archaeological site comprises a rural lane in Lieusaint commune, on the Sénart

184

Plateau on the right bank of the Seine river (Fig. 2). The Sénart Plateau is part of the western

185

end of the French Brie plateau, and is cut by the Seine and Yerres river valleys (Fig. 2). The

186

pedological and geological context is composed of Luvisols formed on silt that covers clayey

187

“argiles à meulières” and sandy “sables de Beauchamps (Beauchamps sand)” deposits

188

overlying the altered Brie limestone. The archaeological remains appear at the interface of the

189

existing ploughed horizon and the Holocene Bt horizon. When loessic deposits or holocene

190

soil are absent due to erosion or unfavorable depositional conditions, the archaeological

191

structures were excavated directly in the “argiles à meulières” or in the Beauchamp sands

192

(Broutin, 2013).

193

Since 1999, development of the new town of Sénart has led to a proliferation of

194

archaeological interventions (Broutin, 2013). Remains of human occupation were found

195

spanning the Paleolithic to the Modern Period. Excavations show evidence of homogeneous

196

development during the early Middle Ages. From the Carolingian period the settlement is

197

structured around one or several roads, and by the 10th and 11th century AD is permanently

198

organized around the Brie-Comte-Robert to Saint-Germain-lès-Corbeil secondary road. The

AC C

EP

TE D

M AN U

182

ACCEPTED MANUSCRIPT 199

questions addressed in this paper concern methods of construction and characterization of

200

Lieusaint lane: Is it possible to distinguish layers related to construction from those related to

201

use of the traffic axis (Fig. 6)

202 - Sampling

204

The excavation revealed a crossroads at the western part of the site (Fig. 6). In the north-

205

eastern area of the excavation another section of the lane on a NNW-SSE axis was excavated.

206

The total thickness of the latter circulation axis was sampled throughout a continuous

207

stratigraphic column (P3 and P4) (Fig. 6).

M AN U

208 209 210

SC

RI PT

203

2.2. Method

211

Micromorphology is the study of the nature and organization of soil components (Bullock et

213

al., 1985; Fitzpatrick, 1993; Fedoroff and Courty, 1994). Roads sections were described and

214

sampled for micromorphological analyses as described by Courty et al. (1989), Courty (2001)

215

and Cammas and Wattez (2009). Following procedures described by Guilloré (1985), the

216

samples were air- and oven dried, at the AgroParisTech microphology laboratory, then

217

impregnated with polyester resin, and thinned to 25 microns on glass.

218

The resulting thin sections (13,5 x 6,5 cm) were analyzed using a polarizing microscope

219

(Nikon E200 Pol petrographic microscope, magnification 10x to 400x), described according

220

to Bullock et al. (1985) and Fitzpatrick (1993), abundance of features or components was

221

estimated with their abundance charts. The hierarchy of signatures of the different

222

mechanisms involved in the formation of features observed through micromorphology allow a

223

chronological series of natural processes and anthropogenic actions to be determined (Courty

AC C

EP

TE D

212

ACCEPTED MANUSCRIPT et al., 1989; Golberg and Macphail, 2006; Cammas and Wattez, 2009; Stoops et al., eds.

225

2010; Cammas, 2015).

226

Descriptions of thin sections are presented in Tables 1, 2, 3, where each thin section (named

227

1058, log 2.1 or P3, left column of the table) is divided into microstratigraphic units or

228

microfacies (named a, b, c, second column of the table) (Courty, 2001; Stoops, 2003). The

229

third column from the left indicates the thickness of each microstratigraphic unit. Others

230

columns refer to microstructure-porosity-c/f pattern, groundmass, inclusions, pedofeatures

231

and interpretation.

SC

RI PT

224

233 234

3. Results and interpretation

235

M AN U

232

The following results and interpretations are presented sequentially for each site. The main

237

microscopic features of interest and results are summarized in Tables 1, 2 and 3.

239 240

3.1. The Sains-du-Nord street

EP

238

TE D

236

In the field, the street was characterized by a coarse backfill that seemed to lie directly on

242

shale. Under the microscope, the shale at the base of the three sampled sections is composed

243

of sub-horizontal plates (Table 1). Its upper limit is sharp and there are no noticeable

244

pedological or biological features, this indicates that the superficial horizons were scraped

245

before construction of the street. Furthermore, no feature is related to rain or climatic agents at

246

this boundary, suggesting that the studied deposits accumulated shortly after the superficial

247

horizons had been scrapped.

AC C

241

ACCEPTED MANUSCRIPT Under the microscope, superimposed microlayers are identifiable on the geological deposit

249

(Table 1) in the three sampled sections (Fig. 7). Two samples were taken in the street (1040-1,

250

1040-2). 1040-2 (Fig. 7), shows three superimposed microstratigraphic units. The lower unit

251

is composed of a mix of sub-angular shale plate fragments and anthropogenic components,

252

specifically tile and burnt brick fragments (1040-2, C, Table 1). As revealed by archaeologists

253

in the field (Neaud, 2015), this layer is contemporaneous with construction of the porticus,

254

thus this microscopic unit would correspond to a craft area. The fissural microstructure of the

255

unit indicates strong compaction. Microlayer 1040-2, C is covered by microlayer 1040-2, B

256

(Table 1). In microlayer 1040-2, shale plates are sub-rounded and there are more

257

anthropogenic components indicating another deposit where the morphology of voids

258

(fissures, planar voids) also indicates strong compaction. The upper microlayer (1042, A,

259

Table 1) is different; particle size is coarser (sandy silt), and some in situ broken components

260

imply direct pressure (Banerjea et al., 2015; Deák et al., 2017, pp. 233-259; Huisman and

261

Milek, 2017, pp. 113-118.)

262

The shale bedrock in sample 1040-1 shows an oblique truncation lined by alignments of shale

263

platelets (1040-1, D, Table 1), suggesting oblique shear stress, probably related to

264

deformation and incision of the shale by wheels, and is very likely the base of a rut (Cammas,

265

2003; Charbonnier and Cammas, 2015; Nicosia and Stoops, 2017). As the upper part of this

266

incision is not evident in thin section, it is impossible to determine whether or not the traffic

267

was directly on the shale or on upper layers. Under the microscope, the geological layer next

268

to the rut in samples 1040-1 is covered by a thin layer of a dense, randomly distributed

269

(millimeters to centimeters) stack of shale plates, relating to the first stage of construction of

270

the axis of traffic in this part of the street. The thinness of the layer does not suggest a large

271

backfill, but rather a levelling (1040-1, C, Table 1). The massive, sandy microstructure of the

AC C

EP

TE D

M AN U

SC

RI PT

248

ACCEPTED MANUSCRIPT upper microlayer (1040-1, A, Table 1) includes in situ broken components which together

273

indicate strong compaction (Usai, 2001).

274

Sample 1058 was taken in a wheel rut identified in the field. Under the microscope, the first

275

layer (1058, B, Table 1) of the geological deposit shows a dense packing of obliquely aligned

276

shale plates caused by shear stress (Cammas, 2003; Hamard et al., 2017, pp. 8-9), probably

277

the result of the wheel rut (Charbonnier and Cammas, 2015; Rentzel et al., 2017, p. 287). As

278

with the two other samples, the upper layer (1058, A, Table 1) is sandy, indicating sediments

279

of different origin, and showing strong compaction.

280

All the archaeological micro layers of this case study are a mixture of geological and

281

anthropogenic components, implying that sediments were prepared before or at the time of the

282

deposit. The scanty fine fraction is brown, limestone-clayey, and massive to fissural, this

283

latter characteristic indicating strong compaction. Ferruginous pedofeatures indicate poor

284

drainage (Lindbo et al., 2010, pp. 129-147) due to moisture of the surrounding soils, enhanced

285

by compaction (Gebhardt and Langohr, 2015, pp. 31-38). The clear boundaries of the

286

archaeological micro layers suggest a rapid succession of deposits; indeed, there are no, or

287

very few, biological features. It is possible that continuous anthropogenic activity did not

288

allow time for plant and soil fauna to colonize the area, and soil hardness due to compaction is

289

very likely a factor contributing to this stability. All layers are strongly compacted, and the

290

sub-horizontal orientation of cracks testify to pressure perpendicular to the surface of the

291

deposits together with wetting / drying processes (Jongerius, 1983, p. 114; Pagliai, 1987, p.

292

420; Courty et al., 1989, p. 130; Rentzel et al., 2017, pp. 281-297).

293

In conclusion and in this case study, micromorphological analysis has made it possible to

294

characterize the stratigraphic units identified in the field and to differentiate layers related to

295

construction from those resulting from traffic. Fine sands in a strongly compacted layer seem

296

to be significant of traffic. In other case studies dating to the Roman Period (Famars,

AC C

EP

TE D

M AN U

SC

RI PT

272

ACCEPTED MANUSCRIPT Northern, France, Bezannes, East, France, Paille-Maille road (Metz), see below), the upper

298

traffic layer shows the same enrichment of allochtonous sands, which could be typical of

299

traffic layer construction (Charbonnier and Cammas, 2015).

300

The traffic layers (1040-2, A; 1040-1, A; 1058, A, Table 1) have an homogeneous distribution

301

of coarse components, testifying to the uniformity of sediments (Usai, 2001). This mixture

302

developed while the sediments were moist as the sub-horizontally oriented short cracks (fig.

303

8b: I, j, k, l) result from compaction and desiccation in a moist environment. The abundance

304

of anthropogenic constituents (bones, charcoal, and ceramics (fig. 8a: a, b, e, f, g, h) and the

305

fragmentation of coarse components (fig. 8a: a, b, c, d) at the time of the layer formed suggest

306

that the passage of vehicles and the depression made by wheels in a moist sediment can be a

307

contributing factor to sediment mixing and breakage of coarse components.

308

Thus, the contribution of materials such as sands, compaction of support components, and

309

presence and integration of anthropogenic components (fig. 8a) (ceramic sherds, charcoal,

310

bone fragments) testify to vehicle traffic. More precisely, observation of thin sections made it

311

possible to highlight micromorphological features of Sains-du-Nord street (Figs. 8a and 8b).

312

Photographs show fractured aggregates (fig.8 a: a, b, c and d) and anthropogenic components

313

in thin section 1058: e.g. charcoal in a rut (fig.8a : e,f) or pottery sherd (fig.8 a : g and h).

315 316 317

SC

M AN U

TE D

EP

AC C

314

RI PT

297

3.2. Paille-Maille (Metz) road, intense-traffic axis

318

In the field, the stratigraphy of the Paille-Maille (Metz) road was composed of alternating

319

bedrock and sandy levels. The thin sections considered in this case study were taken from

320

three sandy levels: at the bottom (finely layered deposit, log 2-1), in the middle (layered

ACCEPTED MANUSCRIPT deposit, log 2-2) and toward the top (massive deposit, log 2-3) of the section, to ascertain if

322

they related to construction or traffic.

323

Microscopic observation of the sandy layers shows them to be composed of superimposed

324

centimeter-thick layers (Table 2). The base of the field section (log 2-3, Table 2) corresponds

325

to a rhythmic accumulation of sandy deposits (log 2-3, X, W, R, L, H, F, D, A, Table 2) and

326

runoff deposits alternating with ferruginous crust (Log 2-3, Q, P, O, N, M, J, K, I, C, B, Table

327

2).

328

The sandy deposits are massive and randomly organized. Voids are mostly fissures and

329

elongated sub-horizontal cavities indicating strong compaction (Jongerius, 1983, p. 114;

330

Pagliai, 1987, p. 420; Courty et al., 1989, p. 130). They are probably intentional deposits

331

related to construction of the road. The thinner deposits could be repairs or intermediate

332

layers of a larger backfill.

333

The runoff deposits indicate lapses of time when the surface was exposed to natural agents

334

such as rain or perhaps flooding events (Macphail et al., 1997, pp. 53–88). Sandy and runoff

335

deposits show impregnated redox pedofeatures indicating redoximorphic processes (Courty et

336

al., 1989, p. 181; Gé et al., 1993; Vissac, 2005; Lindbo et al., 2010, pp. 129–147), These

337

features are enhanced by compaction due to traffic and heavy backfills, are typical of traffic

338

axes (Charbonnier and Cammas, 2015; Gebhardt and Langhor, 2015; Nicosia et al., 2017, pp.

339

331-343), and are found in all micro layers of the section. Yellow phosphatic concretions and

340

impregnations (vivianite and para vivianite, McGowan and Prangnell, 2006) are identified in

341

this part of the section. They testify to waterlogged or submerged soils, presence of phosphate

342

(bone and tissue, cess deposits) and iron sources, low sulfide levels and neutral to slighty

343

acidic pH. In the Paille-Maille (Metz) case study, impregnations and nodules are localized at

344

the base of the stratigraphy, near the water table. As no components or features relate to

345

manure or animals, and as the bones are little weathered, phosphatic features are likely related

AC C

EP

TE D

M AN U

SC

RI PT

321

ACCEPTED MANUSCRIPT to the circulation of soil solution enriched in phosphates (Karkanas and Golberg, 2010, pp.

347

521–541). In the sedimentary discontinuities, which are boundaries between the micro layers,

348

ferruginous crusts made of iron impregnations and coatings are identified. In this case study,

349

as in other case studies of Roman streets and roads in northern France, ferruginous crusts

350

correspond to former traffic surfaces (Courty et al., 1989, p. 181; Vissac, 2005; Goldberg and

351

Macphail, 2006, p. 49; Charbonnier and Cammas, 2015).

352

Microscopic study of the sample taken from the middle of the section shows superimposition

353

of three sandy layers (log 2-2, Table 2). The horizontal boundaries are sharp. As with the

354

lower part of the stratigraphy, the sediments of the three layers are randomly organized and

355

strongly compacted, characteristics indicating backfills. The micro layers also show

356

ferruginous pedofeatures similar to the layers of the lower part of the section signifying poor

357

drainage. The intermediate layer (log 2-2, B, Table 2) is 1.5cm thick which suggests it is a

358

repair. This layer shows small V-shaped incisions related to shear stress that may be related to

359

a wheel rut (see above the Sains-du-Nord street with similar features), but as the incisions are

360

small, it could also be a pothole. However, these features, as with strong compaction, testify

361

to vehicle traffic. Angular sands are also observed in this layer. In situ broken sands observed

362

in other Roman roads and streets were very likely broken in situ by wheels (Cammas and

363

Wattez, 2009; Charbonnier and Cammas, 2015). Although diffuse, the sands in the Paille-

364

Maille (Metz) case study are present throughout the upper part of the section (Log 2-2 and

365

Log 2-1).

366

In the sample taken in the upper part of the section (Log 2-1, Table 2), two superimposed

367

layers were identified in thin section. Their characteristics show each to be strongly

368

compacted backfill, and the high degree of compaction very likely indicates vehicle traffic.

369

In contrast to the Sains-du-Nord street case study, anthropogenic components are found in

370

almost all layers of the Paille-Maille (Metz) road study. Consisting mostly of charcoals and

AC C

EP

TE D

M AN U

SC

RI PT

346

ACCEPTED MANUSCRIPT bone fragments, they could have been introduced with the backfill sediments or could be the

372

result of anthropogenic activities in the vicinity of the road.

373

In conclusion, microscopic observations of the sandy levels indicate successive phases related

374

to construction, repairs, and use of the traffic axis intertwined with phases of disuse due to

375

runoff or flooding at the base of the section (Fig. 9: e, f). Sediment wetness and poor drainage

376

are due to the nearby water table and, as with the Sains-du-Nord case study, are enhanced by

377

sediment compaction due to the weight of successive backfills and traffic. Very few wheel

378

ruts were detected in this sequence. Thus, in these samples, strong compaction, iron features

379

(Fig. 9: a, b, c and d) and coarse sands (Fig. 9: g and h) testify to an arrangement typical of

380

the traffic level.

M AN U

SC

RI PT

371

381 382 383

3.3. The Lieusaint lane

TE D

384

Under the microscope, and in thin sections (Fig. 11), Lieusaint lane is formed of two main

386

components originating from surrounding Luvisols, one from carbonated strata (Bca,

387

carbonate enriched horizon), and the other from a clay silt decarbonated strata (BT horizon,

388

on the Bca horizon in the field) (Table 3).

389

The base of the sequence (P4-F, Table 3) comprises silty sediments with carbonate

390

pedofeatures such as hypo-coatings and calcified root tissues (Fig.10: a and b). The texture

391

and partially carbonated nature of the sediments give them a loessic character, which concurs

392

with deposit types shown on the geological map. The carbonated features indicate a

393

carbonate-enriched horizon (Bca) resulting from an ancient phase of pedogenesis. The

394

presence of a few clay coatings (Fig.10: e to h) superimposed on the carbonate coatings

395

testifies to a second phase of leaching-type pedogenesis. As such, these features characterize

AC C

EP

385

ACCEPTED MANUSCRIPT an in situ Bca horizon, and the superimposition of coatings indicates it is located just under

397

the Bt of a Luvisol.

398

The upper part of P4 and the base of P3 (P4-E, P4-D and P3-C, Table 3) are composed of the

399

same sediments, but they have sub-horizontally oriented voids and cracks (Fig.10: c and d)

400

resulting from compaction perpendicular to the soil surface (Gebhardt and Langohr, 2015, pp.

401

31-38). In the same unit, the presence of rounded calcitic pseudomorphs indicates sediment

402

displacement, possibly a levelling. P4-D and P3-C show little biological activity, suggesting

403

lower anthropogenic pressure on the soils, which might be the result of episodic traffic or

404

show lapse of time between scraping and rolling.

405

P3-B is composed of sediments typical of a Bca horizon. It decomposes into several more or

406

less massive horizontal superimposed units, sometimes delimited by fissures. Here,

407

compaction is moderate, which tends to be an argument in favour of backfill, but the incisions

408

likely testify of at least episodic circulation. In this layer, the washed intercalations indicate

409

runoff processes (Macphail et al., 1997) suggesting short exposition of the layer to rain.

410

At the top of the sequence (P3, A), sub-horizontal cracks are abundant, likely resulting from

411

strong compaction perpendicular to the surface (Rentzel et al., 2017, pp. 281-297). Washed

412

beds testify to runoff (Macphail et al., 1997) and indicate a former surface. Some in situ

413

cracked components indicate strong compaction (Pagliai, 1987, p. 420; Usai, 2001; Cammas

414

and Charbonnier, 2015). The P3, A sequence may be the base of the traffic layer. In

415

conclusion, micromorphological observations indicate that the lane is on a deep horizon. The

416

upper horizon is missing as it had been scraped before use for vehicle traffic (seen also in

417

Sains-du-Nord); here there are no ruts, and typical features related to rolling seem due mainly

418

to strong compaction.

419 420

AC C

EP

TE D

M AN U

SC

RI PT

396

ACCEPTED MANUSCRIPT 421

4. Discussion and conclusion

422 423

4.1. Common microscopic pedofeatures of traffic layers

424 Traffic layers studied in this article are characterized by specific microscopic pedofeatures,

426

which are related mostly to compaction and shearing (Table 4). Table 4 is a synthesis of

427

interpretation of pedofeatures and their processes in soil interpreted in archeological context

428

and known in references.

SC

RI PT

425

429 - Fissures

M AN U

430

In our case studies, fissures are usually thin, long and sub-horizontal, and can form a platy or

432

lamellar microstructure (as in the Lieusaint case study). They testify to strong compaction

433

(Jongerius, 1983, p. 114; Pagliai, 1987, p. 420; Courty et al., 1989, p. 130; Pagliai, 1998),

434

likely related to vehicle traffic (Rentzel et al., 2017, pp. 281–297.)

435

When perpendicular to the surface and grouped closer together towards the top of the layer,

436

they indicate strong compaction perpendicular to the surface which decreases with depth

437

(Pagliai, 1998, pp. 186-196; Charbonnier and Cammas, 2015; Gebhardt and Langhor, 2015,

438

pp. 32-33). Lamellar microstructure indicates very strong compaction of the whole layer, as

439

evident in the Sains-du-Nord street or Lieusaint lane. With regard to the studied traffic axes,

440

layer thickness affected by fissures is up to 5 cm for the Paille-Maille (Metz) road, 4 cm for

441

the Sains-du-Nord street and 1 cm for the Lieusaint lane), and is higher than in trampled

442

floors (more often from a few mm to about 1 cm) (e.g., Gé et al., 1993, p. 153; Banerjea et al.,

443

2015, pp. 1174–1188). In the Paille-Maille (Metz) road, where superimposition of thick and

444

coarse backfill was identified, there are few fissures, and the closed porosity suggests collapse

445

of soil structure (Jim, 1998).

AC C

EP

TE D

431

ACCEPTED MANUSCRIPT 446 447

- Oblique and vertical features Several oblique and vertical features are identified in the 3 cases studies. They are elongated,

449

voids, and / or alignments of sands identified mostly in street and road embankments and

450

backfills. Such features indicate shearing (Cammas, 2003; Hamard, 2017, p. 8-9). In our case

451

studies they delimit incisions probably formed by wheels (ruts). Studies on compaction by

452

tyres in agricultural contexts suggest that the degree of shearing stress is related to tyre

453

thickness. Large tyres tend to compact soils while narrow tyres tend to cut through soils

454

(Hamza and Anderson, 2006). At the microscopic scale, the extent of ruts depends on traffic

455

intensity and the nature of materials used for construction or repair of the traffic layer. Water

456

content of the soil forms another parameter which controls hardness of the soil (Hamza and

457

Anderson, 2006) and consequently the possibility of incision. The thick and coarse material in

458

deep urban street deposits shows very few ruts, possibly because the backfill is too hard for

459

wheels to penetrate (Cammas, 2003).

461

- Iron features

TE D

460

M AN U

SC

RI PT

448

Iron features are very common in road and street deposits. Two main features were observed

463

under the microscope: impregnated nodules and crusts. Impregnated nodules indicate redoxic

464

processes and poor drainage (Courty et al., 1989, p. 181; Gé et al., 1993; Vissac, 2005;

465

Lindbo et al., 2010, pp. 129-147). In the thin sections studied here, redoxic processes benefit

466

from the accumulation of heavy and dense deposits, such as coarse backfills, and compaction

467

due to traffic. Furthermore, moisture is concentrated under watertight backfills, leading to a

468

cover effect similar to that under mats in trampled floors (Gé et al., 1993, pp. 149-163).

469

Ferruginous crusts indicate iron accumulation under gleyed conditions, implying longer

470

periods of water saturation than for impregnated nodules (Courty et al., 1989, p. 181; Vissac,

AC C

EP

462

ACCEPTED MANUSCRIPT 471

2005; Goldberg and Macphail, 2006, p. 49). While ferruginous crusts can be located under

472

hard backfill, they are also found on the traffic surface.

473

475

4.2. Formation processes of traffic axes (Fig.11)

476

RI PT

474

In this study, micromorphological analysis led to identification of elementary micro facies

478

related to construction and use of roads, streets and lanes. Facies related to construction and

479

maintenance are truncation, levelling, plain backfill and repairs. Elementary facies related to

480

use of space are backfill with ruts and circulation levels which support traffic. Only one

481

period of abandonment or disuse of the studied areas was found, characterized by the presence

482

of runoff or flooding deposits.

483

In the present study, the elementary facies present three different patterns or models according

484

to Gé et al. (1993): monophasic, polyphasic and polycyclic (Fig. 11)

485

The monophasic pattern shows only one phase of construction and traffic (Fig.11 a Sains-du-

486

Nord and b Lieusaint). The two case studies present either a truncation of the original soil,

487

then a levelling (Fig. 11 b Lieusaint) or a construction episode (Fig.11 a Sains-du-Nord)

488

followed by vehicle traffic. Although stratification is thin at both sites, the lane and the street

489

were used for a very long time, unlike the Paille-Maille (Fig. 11 c Metz) road site where thick

490

stratification indicates a long period of use.

491

At the Sains-du-Nord site (Fig.11 a), the thinness of deposits, together with

492

micromorphological results such as the cleanliness of traffic layers and presence of repairs

493

indicate regular maintenance of the traffic area. This can be related to the vicinity of public

494

and religious buildings, and can be a sign of the high status of the street. In contrast, at

495

Lieusaint (Fig.11 b), no real signs of maintenance were identified in the field or in thin

AC C

EP

TE D

M AN U

SC

477

ACCEPTED MANUSCRIPT sections, nor were there any signs of abandonment; thus, the thinness of the deposits seems

497

mostly to be related to permanent use.

498

The polyphasic pattern is layered and presents alternating episodes of construction and

499

occupation. It seems to be characteristic of thick street deposits (as in the Paille-Maille (Metz)

500

road case study). They are found in an urban context, for example the civitas capital where

501

the space devoted to traffic is framed and delimited by rows of houses and buildings. Finally,

502

the polycyclic pattern presents a sequence of construction and use, which alternates with

503

episodes of abandonment. The Paille-Maille road (Fig.11 c Metz) sequence is complex. The

504

bottom of the sequence is polycyclic as layers related to use of the road are covered by thin

505

deposits caused by strong runoff, while the top of the sequence is polyphasic. The thickness

506

of the stratigraphy and vertical variation of the pattern of elementary units can have different

507

origins. They may be related to the vicinity of the water table as runoff deposits suggest

508

flooding events of the Moselle at the bottom of the sequence. In this case, the rise of ground

509

level over the centuries could have been promoted to facilitate drainage and protect roads and

510

buildings from flooding.

513 514

SC

M AN U

TE D

EP

512

4.3. Chronology and duration

AC C

511

RI PT

496

515

It is often very difficult to date traffic axes (road, street or lane) and to evaluate the duration of

516

their use, especially away from towns, as few artefacts are found in these deposits.

517

Micromorpholohgical analysis shows that Lieusaint lane was maintained but not rebuilt,

518

although it was used for a long time. Micromorphological studies identify a sequence of

519

facies related to construction, use of space and abandonment. The current study shows that

520

thickness of the stratigraphic sequence of a road or street is not proportional to duration or

ACCEPTED MANUSCRIPT intensity of use. The Sains-du-Nord street, which was in use for more than two centuries, had

522

a maximum thickness of 0.80 m. In the urban context, street deposits seemed well preserved

523

as the increase in thickness of deposits correlates with an increase throughout the level, except

524

for the religious context where, as we have seen in the case of Sains-du-Nord, regular

525

maintenance of the street limits accumulation of layers, and therefore the thickness of the

526

street. In contrast, roads and lanes away from towns present more discreet amenities (or

527

none), and their conservation is more problematic, particularly because of agriculture

528

practices affecting the soil. Micromorphological analysis enabled identification of the nature

529

of layers, such as repairs or traffic episodes and the breakdown of layers into microscopic

530

units, which refine knowledge of the sequence of events in the street, road and lane. Results

531

of the analysis shows that morphology and thickness are more dependent on rhythms of

532

construction, use, and maintenance, than on duration, and suggest that morphology and

533

thickness are strongly linked to the status of particular areas

M AN U

SC

RI PT

521

535 536

5. Conclusion

TE D

534

Field observations and interpretations together with the results of micromorphological

538

analyses facilitate a better understanding of the construction, maintenance and use of traffic

539

axes. They can be presented as descriptive models of formation processes. A monophasic

540

model seems to result from permanent maintenance due to the status or continuous use of the

541

street or lane. Polyphasic and polycyclic patterns of facies mark rhythmic use and rebuilding.

542

The present study allows us to distance ourselves from the vision of a road or street as a paved

543

area and sheds new light on traffic axes. It shows that they should not be considered as one

544

massive deposit or superimposition of backfills, but should be investigated with the same

545

geoarcheological methods and the same care as other archaeological layers. In the extension

AC C

EP

537

ACCEPTED MANUSCRIPT 546

of the DYNARIF collective research programme, and in the long run, the results of the

547

micromorphological analyses presented here should be compared with a typology of roads

548

and lanes in order to approach the standardization and / or the variation of installations

549

according to width and types of roads which will inform us about the status of these spaces.

RI PT

550 Acknowledgments

552

The authors wish to thank Pascal Neaud, Gaël Brkjojewitch and Pierre Broutin -operations

553

managers- for their advice and assistance at various stages of this research. The authors wish

554

also to express gratitude to AgroParisTech for access to the laboratory where the thin sections

555

were produced, and for helpful discussions with researchers. We also thank Norah Moloney

556

(Institute of Archaeology, UCL) for editing the manuscript, and many thanks to Pr Philippe

557

Baveye (AgroParisTech) for his help.

558

M AN U

SC

551

References

560

Ballet, P., Dieudonné-Glad, N., Saliou, C., 2008. La rue dans l’Antiquité : définition,

561

aménagement, devenir. Proceedings of: La rue dans l’Antiquité : définition,

562

aménagement et devenir, de l’Orient méditerranéen à la Gaule. Held in Poitiers, 7th to 9th

563

November, 2006. Presses Universitaires de Rennes, Rennes.

EP

Banerjea, R. Y., Fulford, M., Bell, M., Clarke, A., Matthews, W., 2015. Using experimental

AC C

564

TE D

559

565

archaeology and micromorphology to reconstruct timber‐framed buildings from Roman

566

Silchester: a new approach. Antiquity, 89 (347), 1174–1188.

567

Brkojewitch, Gaël., 2017. Occupation d’un quartier au Haut Empire et du Bas Moyen Age à

568

nos jours. Rapport final d’opération, Pôle Archéologie Préventive de Metz Métropole.

569

Service Régional de l’Archéologie, Metz.

ACCEPTED MANUSCRIPT 570

Broutin, P., 2013. Route de la ferme de Varâtre, Zac du Carré, Phase 4 : extension Clarion

571

Suites (site B1) et Bureaux de Sénart III (site C1). Rapport final d’opération, INRAP,

572

Service Régional de l’Archéologie, Paris. Bullock, P., Fedoroff, N., Jongerius, A., Stoops, G., Tursina, T., Babel, U., 1985. Handbook

574

for soil thin section. Waine Research Publications, Wolverhampton.

575

Cammara Munoz, A., Revuelta Pol, B., 2016. Roman engineering. « that the greatness of the

576

empire might be attended with distinguished authority in its public building ». Held in

577

Segovia from 9th to 11th November 2012 and organized jointly by the UNED and the

578

Fundacion Juanelo Turriano. English edition, 2016.

SC

RI PT

573

Cammas, C., 2003. L’architecture en terre crue à l’âge du Fer et à l’époque romaine : apports

580

de la discrimination micromorphologique des modes de mise en œuvre. In: Chazelles, C.-

581

A. de, Klein, A., Acetta, A. (Eds), Echanges transdisciplinaires sur les constructions en

582

terre crue 1. Terre modelée, découpée ou coffrée. Matériaux et modes de mise en œuvre,

583

Proceedings of a Conference Held 17-18 Novembre 2000, Montpellier, Editions de

584

l’Espérou, pp. 33-53.

TE D

585

M AN U

579

Cammas, C., 2015. L’anthropisation des systèmes pédo-sédimentaires aux périodes protohistoriques et historiques : Une recherche sur les sols urbains (géoarchéologie et

587

micromorphologie). Ph.D. Thesis, AgroParisTech, France.

AC C

EP

586

588

Cammas, C., Wattez, J., 2009. L’approche micromorphologique : méthode et applications aux

589

stratigraphies archéologiques. In: Ferdière, A. (Ed.) La géologie : les sciences de la Terre

590

appliquées à l’Archéologie. Collection Archéologiques, Errance, pp. 181-216.

591

Charbonnier, M.-C, Cammas, C., with the collaboration of Clotuche, R., Guyard, L.,

592

Teysseire, G., Clerget, J., 2015. Apport de la géoarchéologie à l’étude des espaces de

593

circulation dans les villes antiques (Nord de la France). Revue du Nord, 23, Hors-série

594

Collection Art et Archéologie, 133-146.

ACCEPTED MANUSCRIPT 595

Chevallier, R., 1997. Les voies romaines. Armand Colin, Paris.

596

Courty, M.-A., Goldberg, P., Macphail, R.-I., 1989. Soils and micromorphology in

597 598

archaeology. Cambridge University Press, Cambridge. Courty, M.A., 2001. Microfacies analysis assisting archaeological stratigraphy. In: Golberg, P., Holliday, VT., Ferring., C.R., (Eds.), Earth sciences and archaeology, Kluwer

600

Academic/Plenum, New York, pp. 205-239.

601

RI PT

599

Deak., J., Gebhardt, A., Lewis, H., Usai, M-R., Lee, H., 2017. Soils disturbed by vegetation clearance and tillage. In: Nicosia, C., Stoops, G., 2017. Archaeological soils and

603

Sediment Micromorphology. John Wiley & Sons, Ldt, Chichester, pp. 233-259.

SC

602

Duval, P-M., 1989. Construction d’une voie romaine d’après les textes antiques. In: Travaux

605

sur la Gaule (1946-1986) Rome, Ecole Française de Rome, pp. 757-764 (Publications de

606

l’Ecole Française de Rome, 116).

607

M AN U

604

Fedoroff, N., Courty, M.-A., 1994. Organisation du sol à l’échelle microscopique. In: Bonneau, M., Souchier, B. (Eds), Constituants et propriétés du sol. Pédologie vol 2.,

609

Masson, Paris, pp. 349-375.

TE D

608

Fitzpatrick, E.A., 1993. Soil microscopy and micromorphology. Wiley, New York.

611

Gé, T., Courty, M.-A., Matthews, W., Wattez, J., 1993. Sedimentary formation processes of

EP

610

occupation surfaces. In: Goldberg, P., Nash, D.T., Petraglia, M.D. (Eds.), Formation

613

Processes in Archaeological Context. Monographs in World Archaeology, 17, Prehistory

614

Press, Madison, Wisconsin, pp. 149-163.

615 616 617 618

AC C

612

Gebhardt, A., Langohr, R., 2015. Traces de roulage ou de labour ? Le diagnostic micromorphologique. ArcheoSciences 39, 31-38. Goldberg, P., Macphail., R.I., 2006. Practical and theorical geoarchaeology. Wiley-Blackwell Publishing, Oxford.

ACCEPTED MANUSCRIPT 619 620 621 622

Grenier, A., 1985. Manuel d’archéologie gallo-romaine. Tome 2, l’archéologie du sol. Picard, Paris. Guilloré, P., 1983. Méthode de fabrication mécanique et en série des lames minces. 2nd Edition. Thiverval-Grignon. Hamard, E., Cammas, C., Fabbri, A., Razakamanantsoa, A., Cazacliu, B., Morel, J.-C., 2016.

624

Historical Rammed Earth Process Description Thanks to Micromorphology. International

625

Journal of Architectural Heritage, DOI: 10.1080/15583058.2016.1222462.

628

SC

627

Hamza, M.A., Anderson, W.K., 2005. Soil compaction in cropping systems. A review of the nature, causes and possible solutions. Soil & Tillage Research 82 (2005), 121–145. Huisman, H., Milek, K., 2017. Turf as constructional material. In: Nicosia, C., and Stoops, G.

M AN U

626

RI PT

623

629

(eds) Archaeological Soil and Sediment Micromorphology. John Wiley and Sons, Ltd,

630

Chichester, pp. 113-119.

632

Jim, C.Y., 1998. Physical and chemical properties of a Hong Kong roadside soil in relation to urban tree growth. Urban Ecosystems, 2, 171-181.

TE D

631

Jongerius, A., 1983. The role of micromorphology in agricultural research. In: Bullock, P.,

634

Murphy, C.P. (Eds). Soil Micromoprhology. Volume 1, Proceeding of the International

635

Working Meeting on Soil Micromorphology held in August 1981. AB Academic Press,

636

Berkhamsted, pp. 111-138.

Karkanas, P., Goldberg, P., 2010. Phosphatic features. In: Stoops, G., Marcelino, V., Mees,

AC C

637

EP

633

638

F., (eds.), Interpretation of Micromorphological Features of Soils and Regoliths. Elsevier,

639

Amsterdam, pp. 521–541.

640

Lindbo, D.L., Stolt, M.H., Vepraskas, M.J., 2010. Redoximorphic features. In Stoops, G.,

641

Marcelino, V., et Mees, F., (eds.), Interpretation of Micromorphological Features of Soils

642

and Regoliths. Elsevier, Amsterdam, pp. 129–147.

ACCEPTED MANUSCRIPT 643

Macphail, R. I., Courty, M. A., Hather, J., Wattez, J, with contribution of Ryder, M.,

644

Cameron, N., Branch, P., 1997. The soil micromorphological evidence of domestic

645

occupation and stabling activities. In: Maggi, R., (ed.), Arene C.. A Functional and

646

Environmental Assessment of the Holocene Sequence (Excavations Macphail, R. I., Goldberg, P., 2010. Archaeological materials. In: Stoops, G., Marcelino, V.,

RI PT

647 648

Mees, F., (eds.), Interpretation of Micromorphological Features of Soils and Regoliths.

649

Elsevier, Amsterdam, pp. 589-622.

652

SC

651

McGowan, G., Prangnell, J., 2006. The significance of vivianite in archaeological settings. Geoarchaeology 21, 93–111.

Neaud, P., with the collaboration of Charbonnier. M-C., Willems, S., Yvinec, J-H., Jagou,

M AN U

650

653

B., Cammas, C., Coubray, S., under press. Sains-du-Nord (Nord) : une agglomération de

654

la cité des Nerviens. In: Gallia, under press.

Neaud, P., 2014. Développement et abandon d’un sanctuaire au IIIe siècle à Sains-du-Nord

656

chez les Nerviens. In: Van Andringa W., (Ed.), La fin des dieux : les lieux de culte du

657

polythéisme dans la pratique religieuse du IIIe au Ve s. ap. J.-C. (Gaules et provinces

658

occidentales), Gallia, 71.1, 81-95.

Neaud, P., 2015. Sains-du-Nord, 13 rue du Docteur Chevalier. Une rue de l’agglomération

EP

659

TE D

655

antique de Sains-du-Nord, Rapport final d’opération de fouille, INRAP, Service Régional

661

de l’Archéologie, Lille.

662 663 664

AC C

660

Nicosia, C., Stoops, G., 2017. Archaeological Soils and Sediments Micromorphology. John Wiley & Sons, Ldt, Chichester. Nicosia, C., Devos, Y., Macphail, R., 2017. European Dark Earths. In: Nicosia, C. and

665

Stoops, G. (eds): Archaeological Soil and Sediment Micromorphology. John Wiley and

666

Sons, Ltd, Chichester, pp. 331–343.

ACCEPTED MANUSCRIPT 667

Pagliai, M., 1987. Effects of different management practices on soil structure and surface

668

crusting. In: Fedoroff, N., Bresson, L.-M., Courty, M.-A. (Eds.), Micromorphologie des

669

Sols, Soil micromorphology. AFES, Paris, pp. 415-421.

670

Pagliai, M., 1998. Changes of pore system following soil compaction. In: Van den Akker, J.J.H., Arvidsson, J., Horn, R. (Eds.), Proceedings of the 1st Workshop of the Concerted

672

Action on Subsoil Compaction. Experience with the Impact and Prevention of Subsoil

673

Compaction in the European Community, Part 2, 28–30 May, 1998. Wageningen, The

674

Netherlands, pp. 186–196.

SC

676

Quérel, P., 2008. Chemins, gués et établissements routiers dans l’ouest de la Gaule Belgique. Revue archéologique de Picardie N°3/4., 85-122.

M AN U

675

RI PT

671

Rentzel, P., Nicosia, C., Gebhardt, A., Brönniman, D., Pümpin, C., Ismail-Meyer, K., 2017.

678

Trampling, poaching and the effect of traffic. In: Nicosia, C., and Stoops, G. (eds)

679

Archaeological Soil and Sediment Micromorphology. John Wiley and Sons, Ltd,

680

Chichester, pp. 281–297.

683 684 685 686 687

archéogéographes et archéologues en Ile-de-France, FERACF, supp n°52, Tours. Stoops, G., 2003. Guidelines for Analysis and Description of Soil and Regolith Thin Sections.

EP

682

Robert, S., Verdier, N., 2015. Dynamique et résilience des réseaux routiers :

Soil Science Society of America, Madison, Wisconsin. Stoops, G., Marcelino., V., Mees., F. (Eds), 2010. Interpretation of micromorphological

AC C

681

TE D

677

features of soils and regoliths. Elsevier, Amsterdam. Usai, M.R., 2001. Micromorphology of soils/sediments from Adwick-le-Street, Roman Ridge,

688

Doncaster, South Yorkshire. Reports from the Environmental Archaeology Unit, York,

689

43.

690 691

Van Lanen, R-J., Jansma, E., Van Doesburg, J., Groenewoudt, B-J., 2016. Roman and early-medieval long-distance transport routes in North-Western Europe:

ACCEPTED MANUSCRIPT 692

Van Lanen, R-J., Pierik, H-J., 2017. Calculating connectivity patterns in delta landscapes: Modelling Roman and early-medieval route networks and their stability in dynamic

694

lowlands. Quaternary International (2017), http://dx.doi.org/10.10.16/j.quaint.03.009.

695

modelling frequent-travel zones using a dendroarchaeological approach. J. Archaeol. Sci.

696

73, 120-137.

697 698

RI PT

693

Vissac, C., 2005. Study of a historical garden soil at the Grand-Pressigny site (Indre-et-Loire, France): evidence of landscape management. Journal of Cultural Heritage 6, 61-67.

AC C

EP

TE D

M AN U

SC

699

ACCEPTED MANUSCRIPT Tables captions

AC C

EP

TE D

M AN U

SC

RI PT

Table 1. Main micromorphological pedofeatures of Sains-du-Nord street. Table 2. Main micromorphological pedofeatures of Paille Maille (Metz) road. Table 3. Main micromorphological pedofeatures of the lane of Lieusaint. Table 4. Typical micromorphological features linked to passage and vehicle traffic.

ACCEPTED MANUSCRIPT Figure captions

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 1. Ideal model of road showing the different characteristics that can be highlighted, © EHESS-CRH/GGH-Terres: S.Robert, D.Carron. Fig. 2. Location of the archaeological excavations of: Sains-du-Nord (on an extract of the geological map 1/50000, BRGM, n°39, Trelon, © BRGM 2001.) ; Paille-Maille (Metz) road (1/10 000), © Metz Métropole, Y.Daune (Geological map (1/10 000), © Metz Métropole, Y.Daune.) and of Lieusaint (Geological map, Inrap) ; Fig. 3a. Spatial location of micromorphological samples of Sains-du-Nord, © Inrap, Rudy Debiak, Bruno Vanwalscappel, Pascal Neaud. Fig. 3b. Stratigraphic survey of section 5 and location of the studied samples (US 26 = US 1040/1041; US 42 = US 1058) at Sains-du-Nord, © Inrap, Pascal Neaud. Fig. 4. Location of the archaeological area of Paille-Maille (Metz) road ; Spatial location of the micromorphological samples. Fig. 5. Stratigraphic section of the Paille-Maille (Metz) road, © G. Brkojewitch. Fig. 6. Topographical plan of the excavation of Lieusaint (1/500) © Inrap ; Spatial location of micromorphological samples, © Inrap, Marie-Caroline Charbonnier Fig. 7. Thin sections of Sains-du-Nord, © Marie-Caroline Charbonnier. Fig. 8a. Micromorphological features of Sains-du-Nord street: a and b (1040-1) – fractured aggregate (PPL and XPPL); c and d (1040-2) – fractured aggregate (PPL and XPPL); e and f (1058) – charcoal in a rut (PPL and XPPL); g and h (1058) – pottery sherd in a rut (PPL and XPPL), © Marie-Caroline Charbonnier, Cécilia Cammas. Fig. 8b. Micromorphological features of Sains-du-Nord street: i to l (1040-1) – fissural microstructure (PPL and XPPL), © Marie-Caroline Charbonnier, Cécilia Cammas. Fig. 9. Micromorphological features of Paille-Maille (Metz) road; a and b – ferruginous crust (PPL and XPPL); c and d – ferruginous impregnations (PPL and XPPL); e and f – thin runoff deposit (PPL and XPPL); g and h – sandy backfill (PPL and XPPL), © Marie-Caroline Charbonnier, Cécilia Cammas. Fig. 10. Micromorphological features of Lieusaint lane; a and b – calcitic hypocoating (PPL and XPPL); c and d – cavities and vughs (PPL and XPPL); e to h – clay coatings (PPL and XPPL), © Marie-Caroline Charbonnier, Cécilia Cammas. Fig. 11. Formation processes of Sains-du-Nord street, Paille-Maille (Metz) road and Lieusaint lane. © Marie-Caroline Charbonnier, Cécilia Cammas.

ACCEPTED MANUSCRIPT

Quartz and carbonated silt Crystallitic b-fabric (calcitic)

B

Pack of calcitic and clay pedofeature sand more / less massive sub horizontal or lenticular washed units of fabric, unit of fabric with obliqueo r sub-horizontal orientation of voids, limits with v incisions delimited by elongated voids, change of microstructure, few fissures c/f: single to double porphyric

Quartz and carbonated sil t Crystallitic b-fabric (calcitic)

P4

D E

F

11-12

0-2 cm 2 cm

Channel microstructure, few

Quartz and carbonated sil t Crystallitic b-fabric (calcitic) c/f: single to double porphyric

TE D

C

Charcoals

Sub-horizontal spongy unit of fabric, few flattened cavities c/f: single to double porphyric pattern

Quartz and carbonated silt Crystallitic b-fabric (calcitic)

2-20 cm Channel to massive microstructure, few vughs c/f: single to double porphyric pattern

Quartz and carbonated sil t Crystallitic b-fabric (calcitic)

Pedofeatures

Interpretation

Very abundant rounded and fissured calcitic features, hypocoatings and roots pseudomorph swith thick clay coatings, sub-horizontal washed intercalations Very abundant rounded and fissured calcitic features, hypocoatings and roots pseudomorphs with thick clay coatings, irregular silty (washed) intercalations

Strong compaction Vehicle traffic

RI PT

Fissural microstructure, some cavities c/f: single to double porphyric

1-11 cm

Inclusions

SC

Groundmass

M AN U

Microstructure and porosity, C/F pattern

EP

P3

Micro Depth stratigraphic unit A 0-1 cm

AC C

TS

Abundant rounded calcitic hyp-o Deposit of Bca horizon cm coatings and roots Leveling pseudomorphs Fairly abundant ferruginous nodules Abundant ferruginous nodules

Fabric units from Bca horizon Levelling and vehicle traffic

vughs, few fissures Poor drainage

Shear stress Displacement of sediment in mud state over short distance Scraping of upper horizons

Abundant calcitic hypo-coatings Bt horizon and roots pseudomorph s Bca horizon, base of a Luvisol Few and thin yellow to very dark Poor drainage brown clay coatings, few ferruginous nodules, locally periglacial features (ice lenses)

ACCEPTED MANUSCRIPT Pedofeatures

Process in soils

References

Interpretation in archaeological context

Fissures, planar voids, platy microstructure

Compaction

Strong compaction, vehicle traffic

Sub-parallel joint structure

Collapse of microstructure under rotary cultivator, humectation and compaction

Jongerius, 1983, p. 114; Pagliai, 1987, p. 420 (cultivated soils); Courty et al., 1989, p. 130 (under grazing); Gé et al., 1993, p. 153; Pagliai, 1998, pp. 186-196; Banerjea et al., 2015, pp. 11741188; Gebhardt and Langhor, 2015, pp. 32-33; Rentzel et al., 2017, pp. 281-297 Pagliai, 1987, p. 419

Elongated oblique / sub vertical vughs Oblique / sub vertical lines of sands or fine gravels

Shear stress

Redoximorphic features: Fe impregnative oxide nodules

Alternation of oxiding / reducing conditions

Redoximorphic features: ironpans, ferricrust

Cementation by iron (under gleyed conditions)

RI PT

Cammas, 2003; Hamza and Anderson, 2006; Hamard et al., 2017, p. 8-9; Deak et al., 2017, pp. 233-259; Huisman and Milek, 2017, pp. 113-118 Courty et al., 1989, p. 181; Gé et al., 1993; Vissac, 2005; Lindbo et al., 2010, pp. 129-147

Oblique / vertical mixing by wheels, rut

Courty et al., 1989, p. 181; Goldberg et Macphail, 2006, p. 49; Vissac 2005

Hydromorphy due to covering by high watertight backfill and strong compaction

Hydromorphy due to covering by watertight backfill and compaction

SC

M AN U TE D EP AC C

Strong compaction, vehicle traffic

ACCEPTED MANUSCRIPT

Groundmass

Inclusions

A

0-3 cm

Fissural microstructure, random pattern of fissures, cavities c/f: double space to open porphyric

Grey sandy silt Crystallitic to indifferentiate bfabric

Random distribution of often Ferruginous impregnations fissured component s Concrete with tiles fragments, heterogeneous and abundant pottery fragments, bones fragments, charcoals, carbonized and decayed plant fragments, limestone aggregate swith more or less abundant bioclast s

Levelling and mechanical mixing by vehicle traffic Poor drainage Inside of a rut

3 cm

Abrupt wavy limit

3-6.5 cm

Dense packing of lithorelics Fissural microstructure c/f: open prophyric

Clay, few silts Crystallitic b-fabric

Random distribution ando blique Ferruginous impregnation s lines of schist fragments, smaller fragments than the laye rbelow

6.5 cm

Abrupt oblique limit

Thin backfill, levelling Compaction Poor soil drainage Shearing (oblique lines) Vehicle traffic, rut No surficial horizon: scraping

Centimeters to pluri centimeter Clay coatings and intercalations schist fragments, oblique lines of parallel schist plates, in situ broken schist plates Coarse components such as ceramic broken in situ

Substratum, truncated soil Compaction Vehicle traffic, rut (left), edge of the rut (right) Coarse components broken by the traffic Levelling and mechanical mixing by vehicle traffic

Sub angular pottery fragments, Ferruginous impregnation s fresh bones fragments, rounded aggregates, in situb roken aggregates

Backfill of anthropogenic components strongly compacted from the top by vehicle traffic Poor soil drainage

Very small rounded schist fragments, limestone fragments embedded in the ground ma ss Mixture of small, disorganized Few clay coatings and sub rounded schist plates, oblique oriented schist plate s

Thin backfill of clayey silt, schist and limestone Levelling Rut in a geological layer (schist) No surficial horizon at the top, scraping of the soil

SC

6.5-11 cm Dense packing of lithorelics Fissural microstructure c/f: open prophyric 0-0.5 cm Massive microstructure c/f: double space to open porphyric

Clayey silt Crystallitic to indifferentiate bfabric

B

0.5-2 cm

Fissural to platy microstructure, plant pseudomorphos is c/f: open prophyric

Grey silts, dusty dark mass with fine iron impregnation s Crystallitic to indifferentiate bfabric

C

2-4 cm

Fissural microstructure c/f: open prophyric

D

4-8 cm

Fissural microstructure c/f: open prophyric

A

1040-1

TE D

C

M AN U

B

Pedofeatures

RI PT

Microstructure - Porosity - c/f pattern

Grey sandy silt Crystallitic to indifferentiate bfabric

EP

1058

Micro Depth stratigraphic unit

AC C

TS

Clayey silt Crystallitic to indifferentiate bfabric Clayey silt Crystallitic to indifferentiate bfabric

Interpretation

ACCEPTED MANUSCRIPT

A

0-0.5 cm

1040-2

Fissural microstructure, short curved fissures, planar voids c/f: double spaced porphyri c

Sandy silt Crystallitic b-fabric

Components broken in situ, fine charcoals surrounded by organ o phosphatic impregnatio n

0.5- 3 cm

Fissural microstructure, short curved fissures, planar voids c/f: double spaced to open porphyric

Clayey silt Crystallitic b-fabric

Heterogeneous coarse pottery sherds, coarse charcoals fragments, limestone fragments, small rounded schist fragments

C

3-4.5 cm

Stack of schist fragments Fissural microstructure c/f: double space to open porphyric

Clayey silt Crystallitic b-fabric

D

4.5-7 cm

Stack of schist fragments Fissural microstructure, some cavities c/f: double space to open porphyric

Clayey silt to silt Crystallitic b-fabric

Random distribution of rounded micro-sherds of pottery (tiles, burned bricks), sub rounded schist fragments, sub angular schist plates Oblique oriented schist plate s

AC C

EP

TE D

M AN U

SC

RI PT

B

Few clay coatings

Base of the levelling, mechanical mixing by vehicle traffic Poor drainage Strong compaction Abundant construction materials Strongly compacted backfill Sediments with anthropogenic components related to the construction of the porticus First phase of vehicle traffic Rut in a geological layer (schist) No surficial horizon, truncation of the soil

ACCEPTED MANUSCRIPT

log 2.3

B

10.5-13 cm

Complex microstructure, Dark grey sandy silt, some massive and fissural fabric units, angular sands (calcitic) units c/f: porphyric, enaulic

A

0-5 cm

Fissural microstructure, random Sand to sandy silt, some angular short and partially sands accommodated faces, cavitie s c/f: Crystallitic b-fabric (calcitic) single to double grain porphyric

B

5-6.5 cm

Massive to fissural Sand to sandy silt microstructure, random short and Crystallitic b-fabric (calcitic) partially accommodated faces, cavities, open V-shape lines of gravels with sandy infilling c/f: single to double grain porphyric

Charcoals, few rounded bones Localized ferruginous Compacted backfill Poor fragments, some silt aggregates, impregnations in the thickness of drainage Compaction and rounded pluri centimetric the micro layer, continuous poor drainage due to dense gravels ferruginous features at the top of and heavy superimposed the layer (impregnations and backfill and vehicle traffic coatings) Few rounded bones, very few Abundant ferruginous silt Thin backfill, repair aggregates impregnation and coating s Redox processes between layers (discontinuities of the soil) Poor drainage Poor vehicle traffic, small rut or pothole

C

6.5-11.5 cm

Fissural microstructure, random Sand short and partially Crystallitic b-fabric (calcitic) accommodated faces, cavitie s c/f: single to double grain porphyric

Burned bone fragment, abundant Ferruginous features, Backfill related to road continuous construction charcoals, rounded gravels < 2 cm ferruginous impregnation and coatings at the Poor drainage top of the layer

Inclusions Pedofeatures pattern

Charcoals, sub angular fissural Few ferruginous impregnation s sandy silt aggregates, abundant rounded pluri centimetric gravels, few rounded oolithic calcareous gravels Burned bone fragments, Large ferruginous impregnation s abundant charcoal s intergrain microaggregate fabric Crystallitic b-fabric

RI PT

Sand, some angular sands, little silt Crystallitic b-fabric (calcitic)

Interpretation

Dense coarse backfill, no rut Repair Poor drainage

Compacted backfill, no rut Vehicle traffic? Poor drainage

M AN U

0-10.5 cm Pellicular grain structure c/f: Chitonic to gefuric

Groundmass

SC

Microstructure and porosity- c/f

TE D

log 2.2

Depth

EP

log 2.1

Micro stratigraphic unit A

AC C

TS

A 0-1.5 cm Pellicular grain to bridged grain Sandy silt Charcoals Abundant ferruginous impregnation s Poor drainage c/f: chitonic

Sandy backfill haut microstructure

Crystallitic b-fabric (calcitic)

ACCEPTED MANUSCRIPT

B

1.5-2 cm

Massive microstructure, random short fissures with partially accommodated face s c/f: open porphyric

Brownish sandy silt (colour due Few charcoals to iron impregnation )

C

2-3 cm

See below

See below

D

3-4.9 cm

Microlayered, random short Sandy silt fissures with partially Crystallitic b-fabric (calcitic) accommodated faces, sub horizontal fissures, elongated cavities c/f: single grain to open porphyric

Rounded bone fragments, charcoals

E

4.9-5 cm

Micro-layered microstructure c/f: Silt open porphyric Crystallitic b-fabric (calcitic)

Abundant fine charcoals with sub Few sub horizontal lenses of fine Thin runoff deposit horizontal patter n sands

F

5-5.4 cm

Spongy microstructure, Silty sands polyconcave voids, few channels Crystallitic b-fabric (calcitic) c/f: single grain porphyric to monic

Abundant charcoals and yellow Large ferro-phosphatic phosphatic nodules and aggregate impregnation s s Bones fragments, siliceous plants residues

G

5.4-5.5 cm

Micro-layered microstructure c/f: Silt open porphyric Crystallitic b-fabric (calcitic)

Abundant fine charcoals with sub horizontal patter n Siliceous plants residues

H

5.5-6 cm

Single grain to bridged grain microstructure c/f: porphyric to monic

EP

TE D

AC C

Sand Crystallitic b-fabric (calcitic)

Ferruginous crust Redox processes between layers ( discontinuities of the soil) Compaction and poor drainage due to above vehicle traffic and heavy backfill

RI PT

M AN U

SC

Few charcoals

Continuous sub horizontal ferruginous impregnatio n

Ferruginous coatings and hypo coatings

Poor drainage

Ferruginous impregnations, Yellow phosphatic nodules , Intercalation of more silty / sandy sub horizontal lense s

Compacted backfill, no rut Compaction and poor drainage due to above vehicle traffic and heavy backfill

yellow Thin compacted backfill Anthropogenic refuse (nightsoils)

Subhorizontal lenses of fine sands Thin runoff deposit Anthropogenic refuse

Abundant charcoals, burned seed Ferruginous impregnations, lenses Thin compacted backfill, no rut of fine sands Poor drainage Compaction and poor drainage due to dense and heavy above backfill and vehicle traffic

ACCEPTED MANUSCRIPT

6-6,5 cm

Micro-layered, massive Silt to sandy silt microstructure, few vesicles, Crystallitic b-fabric (calcitic) from top to bottom: ferruginized calcitic bed, calcitic bed, ferruginized calcitic bed

Charcoals, bone fragments, a Ferruginous crust at the top and small ferruginized wood at the bottom, intercalations of fragment fine lenses of silt or fine sands

L

6.5-8 cm

Single grain microstructure c/f: monic

Heterometric sand Crystallitic b-fabric (calcitic)

Charcoals

MNOPQ

8-9 cm

Microlayered, from top to bottom: iron crust, silty layer, silty sand layer, silty ferruginous layer c/f: porphyric

Silt to sand Crystallitic b-fabric (calcitic)

Charcoals

R

9-10.5 cm

Single grain microstructure c/f: monic

Sand Crystallitic b-fabric (calcitic)

S

10.5-11 cm

Micro-layered microstructure c/f: Sandy silt single to open porphyric Crystallitic b-fabric (calcitic)

T

11 cm

0-1 cm

Log 23bas

V

1

Sandy backfill, repair

Ferruginous impregnations

Stack of thin runoff deposit Poor drainage Redox processes between layers (discontinuities of the soil)

Charcoals

Yellow phosphatic nodules, lenses and intercalations of fine to coarse sand

Sandy backfill, repair

Rounded bone fragmented in situ

Intercalation of sandy lenses

M AN U

SC

Yellow phosphatic nodules, intercalations of fine to coarse sands

TE D

Organic runoff deposit, medium competence Poor drainage Compaction and poor drainage due to vehicle traffic Thin bed of microlayered silt Brownish red silt Fine charcoals Continuous sub horizontal Organic runoff deposit, low competence c/f: open porphyric Low crystallitic b-fabric ferruginous impregnation s Redox processes between (calcitic) layers (discontinuities of the soil) -1.5 cm Micro-layered microstructure, Silty sand Charcoals in a sub horizontal Some ferruginous Runoff / washed deposit, curved, subhorizontal and Crystallitic b-fabric (calcitic) pattern impregnations,i ntercalation of vertical variation of subvertical short fissures with sandy lenses competence partially accommodated faces , c/f: open porphyric

EP

11-12 cm

Thin runoff deposits Redox processes between layers (discontinuities of the soil)

AC C

U

RI PT

IJK

Massive microstructure, random Yellow to red silty sand short fissures with partially Crystallitic b-fabric (calcitic) accommodated faces, few cavities c/f: open porphyric

Abundant charcoals, ferruginous Continuous sub horizontal pseudomorphs of plant residues, ferruginous impregnatio n bone fragment s

Organic runoff deposit, low competence Redox processes between layers (discontinui ties of the soil) Compaction and poor drainage due to dense and heavy above backfill and

ACCEPTED MANUSCRIPT

vehicle 2-4 cm

Microlayered single grain to pellicular grain microstructure, V -shape incisions at the base of the layer c/f: monic

Sand, washed sands at the base, heterogeneous particle siz e Thin layer of fine sands at the base Crytallitic b-fabric (calcitic)

Abundant charcoals, rounded aggregate s

Fine charcoals, ferruginized and Almost continuous ferruginous Deposit of sandy materials charred plants residues, rounded and phosphate impregnations of Thin compacted backfill quartz < 2cm the groundmas s Poor drainage Ferruginous crust at the to p

SC

W

RI PT

traffic

Few ferruginous coatings Sandy backfill, part of the street construction Poor drainage Runoff at the base (washe d layer) with mechanical disturbance

4-5.5 cm

Massive microstructure Small and short curved fissures, short accommodated fissures, elongated cavitie s c/f: porphyric

Sand Crystallitic b-fabric (calcitic)

Y

5.5-6.5 cm

Massive to fissural microstructure, random short fissures with partially accommodated faces, cavitie s Upper limit with V-shape incision c/f: single to open porphyric

Sand to silty sand Rounded heated bone fragme nt Some ferruginous and Some bioturbation phosphatic impregnation s Ancient surface Dusty groundmass (fine organic Poor drainage particles) Crystallitic b-fabric (calcitic)

AC C

EP

TE D

M AN U

X

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

b

d

AC C

EP

TE D

c

M AN U

SC

a

RI PT

ACCEPTED MANUSCRIPT

e

f

g

h

ACCEPTED MANUSCRIPT

Geological layer

Troncature of the soil

Construction

Troncature of the soil

Leveling

Vehicle traffic

Backfill

EP

Runoff deposit

AC C

Ancient backfill

Runoff deposit

TE D

c METZ

M AN U

Geological layer

Repair

SC

b LIEUSAINT

RI PT

a SAINS DU NORD

Construction and vehicle traffic

Vehicle traffic

Compaction

Formation of a ferruginous crust

Sandy deposits

ACCEPTED MANUSCRIPT

Recent and current fluvial deposits

Ancient alluvium of the lower terraces (5-8 m)

Marnes to Amaltheus margaritatus (Lower Domerian

Marno-limestone and dolomitic ("Gray layers" and "W (medium Muschelkalk)

1:10 000

0

200

400

c Geological map of Metz (1/10 000), © M

AC C

EP

TE D

M AN U

SC

RI PT

a Localization of the archaeological intervention of Lieusaint at 25,000 (©IGN, SCAN 25), Inrap.

Localization of the archa

Localization of the archa

Localization of the archa

X =772290

X =772290

0

10 m

ACCEPTED MANUSCRIPT

Figure 3a : Spatial location of micromorphological samples of Sains-du-Nord US 1

US 2

E

US 3 O 201,25 m NGF US 1 US 31 US 33

US 2

US 15

US 3

US 30 US 45

US 44 US 13 US 20

US 12

US 67

2

US 68

1

US 47

RI PT

US 29

US 26

US 43 US 42

US 48

US 46

Blocks of blue stones

0

US 22

US 23 US 28 US 27

US 21

5m Studied samples: 1 = US 1040/1041 ; 2 = US 1058

AC C

EP

TE D

M AN U

SC

Figure 3b : Stratigraphic survey of section 5 and location of the studied samples (US 26 = US 1040/1041; US 42 = US 1058) at Sains-du-Nord

Ponti Pontiffroy

SE LL E

N

MO

ACCEPTED MANUSCRIPT

170

180

Metz-Trèves road

others a

RI PT

la Meu

carp on

ne

rthe

Moselle

LE IL SE

Road le

Amphithéâtre

0

500 m

1: Hotel de P 2 : Pontiffroy (Gama et al.

e Voie d

Voie

de S

1

M AN U

SC

The Pontiffroy district in Metz-Divodorum at the end of the High Empire.

3

3

TE D

3

Figure 7b

To MetzDivodorum

3

Of the diverticulum at 100 m in the extension

AC C

6

Recent an

EP

2

2

2

Ancient a

Marnes to

Marno-lim (medium M

Caption :

4

7

Pole related to a building: Antique wall recognized by excavation

5

Main road recognized in excavation Secondary road recognized in excavation restitution Excavation limit Localization of the survey

0

50 m Ech. 1/1000

Figure 7a

1: Hotel de Police (WATON 1985) 2 : Pontiffroy, rue Belle-Isle et de la caserne (GAMA et al. 2004) 3 : Cité administrative (WATON 1983 ; 1984) 4 : Jardin du Mail (GEORGES et al. 1988) 5 : Salle de réunion du Conseil Régional (DAUTREMONT 1987) 6 : Résidence St-Vincent (BOUCHET 2004) 7 : Rue St-Vincent (GÉBUS1997)

1:10 000

0

20

RI PT

167,00

168,00

ACCEPTED MANUSCRIPT

s-o

1006

1005

fo994

1009 1011

1017

1025

1019

1023

1028

993 999

998

1024

2984

1002 1001

Road I

2

3

1041

4

Sidewalk

5

6

Material recovery pit Flood levels

Gallo-Roman & Medieval times

Medieval times

Road V

Road VI

Medieval road

Sidewalk

Sidewalk

Modern remains

Gutter

7

8

9

Road III Road III

Sidewalk ditch

Levelling

Road V

1042

1043

Road II

AC C

Road Ib Sidewalk

1038

1029

Road II

Road Ia

993

1037 1040

1039

TE D

1m

1033

fo1030

EP

0

1032

fo1026 1016

1018

992 165,00

1008

1010

M AN U

166,00

1007 989

SC

976

ACCEPTED MANUSCRIPT

616000

M AN U

SC

102000

LIEUSAINT

RI PT

102000

614000

SAINT-PIERRE-DU-PERRAY

SAVIGNY-LE-TEMPLE

616000

P2

AC C

EP

TE D

614000

Study area

286

395

535

46 1 ornières 234

excavation limit

S/E

Medieval times : Merovingian Carolingian Capetian

0

10

20

30

40

50 m.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

1 cm

1 cm 1 cm

a Thin section 1040.1

b Thin section 1040.2

c Thin section 1058

b

M AN U

SC

a

RI PT

ACCEPTED MANUSCRIPT

d

AC C

EP

TE D

c

e

f

g

h

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

i

j

k

l

b

d

AC C

EP

TE D

c

M AN U

SC

a

RI PT

ACCEPTED MANUSCRIPT

e

f

g

h