Version classiqueVersion mobile
OpenEdition Books

La géoarchéologie française au xxie siècle

Nathalie Carcaud
Gilles Arnaud-Fassetta

Partie II. Les hydrosystèmes fluviaux, entre climat et anthropisation/Section 2. Fluvial hydrosystems, between climate and “anthropogenic processes”

Chapter 7. Fluvial geoarchaeology of the valley floors of the Moselle catchment (France, Luxembourg)

Stéphane Cordier, Henri-Georges Naton, Laurent Brou, Dominique Harmand, Olivier Moine, Pascale Ruffaldi, Gaël Brkojewitsch et Vincent Ollive

Texte intégral

The OSL dating has been performed at the laboratory of the Leibniz Institut for Applied Geophysics (LIAG) in Hannover, Germany, thanks to the head of the laboratory, Manfred Frechen. We also thank Sumiko Tsukamoto, Sabine Mogwitz and Sonja Riemenschneider for their advice and help during the samples preparation and measurements. We thank the two anonymous reviewers for their constructive comments, which allowed the manuscript to be improved. We are finally grateful to Tom White (University of Cambridge, UK) for reading and suggesting amendments to the English manuscript.


1The Moselle River runs through a catchment spanning four countries (France, Luxembourg, Germany and Belgium). The area has been occupied by human populations throughout the Prehistoric period. Holocene and Historical occupation has been especially important: for example, the modern vineyards and the cities of Toul, Metz and Trier were founded during the Roman period. A more or less continuous human presence can hence be assumed since ca. 300 ka and probably earlier (Guillaume, 1982; Guillaume and Janot, 1983; Janot, 1988; Lebrun-Ricalens, 1995).

2Despite the approach that has been largely developed in France over several decades, especially due to the research of Joëlle Burnouf and Jean-Paul Bravard, geoarchaeological research in the Moselle catchment is less developed than in many other areas of Europe. This can be partly explained by a lack of interest by researchers (in particular geographers and archaeologists) working in the catchment, until relatively recently. Furthermore, the morphological and climatic framework is not really favourable to strong environmental dynamics. The area is actually characterised by low tectonic activity, low slope erosion and aeolian activity and low morphogenic influence on the fluvial systems (Cordier et al., 2012) in comparison with other regions such as the Mediterranean Basin. However, geoarchaeological research related to archaeological rescue excavations has been carried out over the past two decades. It has typically focused on fluvial systems, with a long timescale approach (several ka) supported by a chronostratigraphical framework derived from various dating methods (Optically Stimulated Luminescence OSL, radiocarbon, and dendrochronology), in the context of rescue surveys (Carcaud, 1992; Naton et al., 1999; Cordier et al., 2010). The aim of this contribution is to provide an overview of this research in France and Luxembourg, to emphasise the recent advances (understanding of relations between human societies and their environment, methodological improvements, and to identify the prospects for future work in this region).

Study area and methods

3The Moselle River is the main leftbank tributary of the Rhine River, which joins the Rhine at Koblenz (Germany). It rises in the Vosges Massif (Fig. 1), where the bedrock geology consists of Hercynian basement (crystalline) rocks and Permian and Triassic sandstones and conglomerates. Downstream from the Vosges Massif, the Moselle River flows through the cuesta ridges of the Eastern Paris Basin. These ridges developed on hard limestones, leading to a decrease in valley width. In contrast, the valley enlarges in softrocks (marls and clays). This allows the deposition of large quantities of sediment originating from the Vosges Massif in several alluvial basins (e.g., Metz-Thionville, Remerschen-Wintrange). The Moselle River receives discharge from its main tributaries, the rivers Meurthe, Seille, Sûre, and Sarre in the Paris basin. In its lowermost reach, the Moselle River mainly flows across hard rocks (limestones along the German-Luxembourgian border, schists in the Rhenish Massif), and forms wide entrenched meanders.

4Research was conducted at several valley floors sites, in the Moselle valley, such as Remerschen and Vittonville, in the Seille valley at Metz-Mazelle, and in the Alzette valley (a Sûre tributary) at Lorentzweiler (Fig. 1). It included both geoarchaeological (e.g., relations between human societies and their environment) and palaeoenvironmental (reconstruction of the variability of the natural conditions) approaches, based on archaeological data, field work (stratigraphy), sedimentology (grain size, mineralogy), palynological and malacological studies, and absolute dating methods.

Fig. 1. General map of the study area with location of the main sites.

Pleniglacial to Holocene valley loors evolution

The Pleniglacial complex of the Moselle valley at Remerschen (Luxembourg)

5Geoarchaeological research related to rescue archaeology has been conducted over the past two decades by the Service d’Archéologie Préhistorique at the Musée National d’Histoire et d’Art (MNHA) of the Grand-Duchy of Luxembourg at several sites (Rem I to Rem VI) in the Remerschen-Wintrange alluvial basin, in the Moselle valley. This led to a palaeoenvironmental reconstruction of valley evolution since the Pleniglacial (Naton et al., 2009), made possible by the preservation (unusually for the Moselle catchment) of slope and aeolian deposits above the luvial sediments of the lower terrace M1 (3 m relative height; Cordier et al., 2005, 2009). Seven sedimentary units were identified (Fig. 2):

6- Unit A: the bottom of the sequence consists of 6 m of coarse-grained alluvial deposits (gravels and boulders up to 0.5 m3; Fechner and Langohr, 1994). The sediments are angular and included within a sandy matrix. he deposition of these boulders, originating from Tanus quartzites outcropping a few kilometres upstream, is attributed to ice-rating, as recognised in other sections of the Moselle valley (Cordier et al., 2006). Other sedimentological characteristics suggest that the deposition of this coarse unit A is likely to have occurred in braided channels.

Fig. 2. Stratigraphy and location of the archaeological findings in the sequences of the Remerschen-Wintrange basin.

7- Unit B: this unit (1 to 3 m in thickness) is made up of sandy sediments with trough cross-bedding, similarly attributed to a braided river.

8- Unit C: this unit of varying thickness (between 50 cm and 2 m) is made up of a series of centimetrescale beds of sands and silts. This facies is typical of natural levee deposits brought by a relatively inactive river.

9These three units (A, B, and C) correspond with the sedimentary body of the lower terrace M1. Furthermore, the fining-upward trend is typical of the Moselle fluvial deposits (Cordier et al., 2006).

10- Unit D: above the fluvial sediments, Unit D consists of up to 50 cm of sandy loess at Rem IV and Rem V. Whilst calcareous concretions have been found at the base of the unit, the top of this unit is usually decalcified. Unit D has yielded a terrestrial malacofauna (including Pupilla muscorum, Succinella oblonga, and Trochulus hispidus) typical of Pleniglacial loess and associated with cold, wet environments with minimal plant cover (Moine, 2008). At the Jongerbierg site, silty sediments correlated with Unit D have yielded a more diverse fauna, including Pupilla alpicola, Pupilla loessica and Columella columella. This assemblage is typical of wet conditions and suggests a steppe-like tundra (Moine, 2010). The spectrum is intermediate between the poor assemblages found in Northern France, and the rich spectra recognised in the Rhine area (Moine, 2008; Moine et al., 2011). This molluscan record is richer than those found in Northern France, but poorer than those from the Rhine valley (Moine, 2008). It confirms that an interstadial warm period occurred during the Weichselian Upper Pleniglacial, as previously shown in the surrounding areas (Moine et al., 2008, 2011).

11- Unit E: sediments from units C or D are overlain by marly slope deposits (less than 1 m in thickness), soliflucted from the left bank slope of the Moselle River (Keuper marls). These deposits exhibit various cryoturbation features (involutions, plications, ice-wedges, polygons), allocated to the presence of a discontinuous permafrost (Cordier et al., 2006). Ice-wedges are locally filled with well-sorted reddish sands, originating from the older fluvial terraces preserved on the leftbank or deposited through a resumption of fluvial activity. These sands may be allocated to Unit F.

12- Unit F: this sandy unit is approximately 1 m thick and consists of sands showing alterations resulting from temperate pedogenesis (oxidation, ferro-manganese concentrations, root traces, etc.). The age of this unit remains unclear. Neolithic and Protohistorical archaeological structures have been found at the contact between this unit and Unit G.

13- Unit G: this unit is located just below the present soil. It is ca. 1 m thick and consists of slope deposits. A dark lower layer (subunit G1) can be locally distinguished from a paler upper layer (subunit G2).

14Dating techniques have been applied to sediments from several units. Luminescence dating based on quartz (OSL) and feldspars (IRSL) at Rem VI yielded Weichselian Middle to Upper Pleniglacial ages (approximately MIS 3-2 transition) for the fluvial sands of units B and C (Fig. 2; Cordier et al., 2010). At another site (Rem IV), a Juniperus charcoal fragment recovered from between units C and E yielded an AMS age of 30770± 300 BP (Beta-182248). However, the precise provenance of the charcoal remains unsure as the units are affected by periglacial deformations (wedges, etc.). The charcoal may have been transported with the marly deposits (Unit E) originating from the upper slope. Its size (several mm) suggests a mass-transport, which could be younger than the AMS age obtained. However, this result is consistent with the luminescence ages and allows the deposition of units A to E to be allocated to the Weichselian Middle to Upper Pleniglacial. This reconstruction is consistent with the assumed Lateglacial to Holocene age of the present Moselle floodplain M0 (Carcaud, 1992), and with the chronology for loess deposition in surrounding areas (Northern France, Belgium, Upper Rhine Graben).

15A series of AMS dates were also been obtained using remains preserved in structures developed in Unit F (Damblon and Hauzeur, 2009). These provided ages of 6320 ± 50 BP and 6110± 60 BP (e.g., 5230 to 5060 cal BC with a 2σ calibration) for the Early Neolithic structures preserved at Rem I. The dating of two Quercus sp. charcoal fragments sampled in the lower part of Unit F at Rem IV yielded Sub-Boreal ages (Beta 157202: 3770 ± 50 BP; Beta 157203: 4030± 50 BP). Even if an old-wood effect (charcoal originating from the vicinity of the sapwood) was present (especially in the case of oak), these ages are considered to be consistent and provide a terminus post quem for Unit F. At the top of this unit occur cremation urn burials dating from the late Bronze Age at Rem II-Klosbaam. At Rem I, Early Neolithic structures (Atlantic) cut across sediments supposedly analogous to Unit F, suggesting a diachronic age for this unit. It has, however, proved difficult to provide reliable correlations between the archaeological sites Rem I, Rem II and Rem IV. The youngest archaeological structures found within Unit F at Rem I-Schengerwis date from the Iron Age. A human bone from a silo-burial was dated to 2220± 40 BP (Beta 155324) and two beech charcoal fragments were dated to 2155± 45 BP (GrA 23808) and 2145± 40 BP (GrA 23668). Finally, research conducted in the Remerschen-Wintrange area has allowed a tentative reconstruction of palaeoenvironments since the Last Glacial Maximum, and provided a chronostratigraphical framework for the archaeological remains found there (see infra).

The Late Glacial-Holocene Alzette sequence at Lorentzweiler (Luxembourg)

16During the construction of the Luxembourg-Ettelbruck highway, a viaduct was built near Lorentzweiler to cross the Alzette valley. A study of geotechnical borings performed by the Service Géologique du Grand-Duché de Luxembourg (SGL/Administration des Ponts et Chaussées) provided new information about the sedimentary infill of the Alzette valley loor. The Service d’Archéologie Préhistorique of the MNHA and the SGL used this dataset to reconstruct the geometry of the sediments deposited since the LGM, using a similar approach to that applied in the Moselle valley.

17The borings have allowed the recognition of a terrace staircase similar to that of the Moselle River (see supra; Cordier et al., 2005, 2006; Naton et al., 2009), with two alluvial formations in the valley loor (Az0 and the lower terrace Az1) and an older terrace, Az2 (Fig. 3). Peat and organic deposits have been found interbedded with ine-grained sediments (decantation clays and overbank silts). Downstream from the viaduct, highly organic sediments have been found in a core within the coarse valley loor deposits. Palynological analyses suggest allocation to the Bølling period. This result is consistent with those obtained in Lorraine, especially at Marly (Seille valley; Rufaldi, 2000). The pollen sequence indicates an open environment dominated by various steppe herbaceous plants. The only significant trees are birch, willow and shrubs (Juniperus, Hippophae). he Lorentzweiler sequence is the irst Lateglacial record found in Luxembourg since the studies of Coûteaux (1970) in the Echternach area. The Lorentzweiler sequence is also important for the reconstruction of the evolution of regional palaeoenvironments, since it spans times periods not commonly represented in the Moselle valley. Previous research conducted a few kilometres downstream can also be mentioned: in 1953, 5 m of ine-grained sediments were recognised, that had been deposited between the Early Sub-Atlantic and present (Heuertz, 1969). This clearly suggests an increasing sedimentation rates since the Bronze Age.

Valleys and human societies

Human occupation in the Remerschen-Wintrange basin

18The oldest archaeological remains in the Moselle valley date mainly to the Middle Palaeolithic. They have been found on the surface of Triassic rocks or older terrace deposits. These surface sites have yielded artefacts from various lithic industries including flint knapping by-products (cores, levallois flakes) as well as tools (e.g., scrapers, denticulates). The raw material was obtained either from terrace sediments (quartzite gravels) or from local rocks (Muschelkalk lints, Bajocian cherts, quartz; Rebmann et al., 2001). The Mousterian industries were manufactured using recurrent centripetal knapping technique, with a predominance of the discoidal or Levallois methods (Le Brun-Ricalens et al., 2011). Dates for this material range between 300 and 40 ka (Jaubert, 1999; Delagnes et al., 2007).

Fig. 3. Schematic cross-profile of the Alzette valley at Lorentzweiler.

19Upper Palaeolithic industries have only been found in tributary valleys and belong to the recent Aurignacian (Brou et al., 2009) and the Gravettian. Late Glacial industries are sparsely represented, which is surprising considering their abundance in surrounding areas. In the Remerschen-Wintrange basin (at Raederbierg; Fig. 2), an epi-Ahrensbourgian industry (late Younger Dryas-Early Preboreal) and an Iron Age structure have been found within the slope deposits of Unit G and in the underlying Unit F, respectively (Brou, 2001). We assume that the slope deposits were transported only a short distance and are related to soil erosion during Roman times. Finds related to the Mesolithic period are sparse and located on the surface. The lack of Mesolithic artefacts can be interpreted as a consequence of natural and/or anthropogenic slope erosion. In contrast, several well-preserved Neolithic sites have been recognised in the Remerschen-Wintrange basin. An Early Neolithic village (upper and late Linear Pottery Culture) at Rem I-Schengerwis has been studied since 1993 (Hauzeur, 2006). The structures were preserved at the top of clayey sand deposits allocated to Unit F. Anthracological (charcoal) analysis has provided an improved knowledge of the forest landscape and its use by humans. Six types of vegetation were identified: shoreline vegetation, alluvial forest, ravine and lower slope forest, forest edge, living hedges and clearings with pioneer stages of the floral succession (Damblon et al., 2007; Damblon and Hauzeur, 2009). The predominance of oak is consistent with an Atlantic climate but may also be the result of human preferences (for building purposes). However, no direct anthropogenic influence on the fluvial dynamics of the Moselle River has been clearly recognised. In the Metz area (La Maxe and Maizières-lès-Metz sites), human influence has also been recognised at the beginning of the 5th millennium BC, indicated by the presence of Cerealia and ruderal plant pollen (Plantago lanceolata, Plantago major/media, Chenopodiaceae, Artemisia, etc.). This human influence is also limited and largely dependent on population densities and the distance between the settlements and the positions of boreholes (Ruffaldi, 2000).

20With the exception of small isolated objects, no signs of human occupation during the Late Neolithic and the Late Bronze Age are known. However, a cremation necropolis attributed to the latter period has been found at Rem II-Klosbaam near the Neolithic settlement (Nicolas and Le Brun-Ricalens, 2010). The structures were found in sandy-clay sediments similar to those on the same surface that the Early Neolithic and second Iron Age structures were discovered.

21Increasing human occupation during Gallo-Roman times is attested by numerous finds. Several clusters of buildings and a large villa have been found. Downstream from the Remerschen-Wintrange basin, between the villages of Greiveldange and Machtum (11 km), a villa was recognised every 1.5-2.5 km (Wagner, 1987). In the Remerschen-Wintrange area, there is evidence for the development of farming (cereal cultivation, vineyards) and, as a consequence, increased soil erosion. This led to the destruction of many features from earlier periods and to the levelling of the microreliefs associated with previous fluvial evolution (progressive filling of small fluvial depressions by slope deposits – a process that may have continued after the Roman period until today). The banks of the Moselle River were also stabilised, and a portion of an ancient Roman road has been found at Schengerwis, 100 m away from the present riverbank.

22To conclude, this overview shows that human occupation in the Remerschen-Wintrange basin has been more or less continuous since the Middle Palaeolithic, with a hiatus recognised between the Early Neolithic and the Late Bronze Age. This hiatus is probably due to the presence of many gravel pits in the area, which destroyed about 30% of the land surface before geoarchaeological research could be performed Humanly-induced erosion, local topography and the type of structure (deep pits or postholes) also explain the uneven preservation of these sites. Erosion may have started during the Late Bronze Age (as was shown for the Alzette valley, see supra) and increased during the Iron Age, culminating during the Roman period. The acceleration of soil erosion is confirmed by research conducted in the German Moselle valley (Trier area). Here, fluvial sedimentation rates increased from 0,14 mm yr-1 before the Roman conquest to 2.6 mm yr-1 during the Roman period (Zolitschka and Löhr, 1999). Accordingly, the slope deposits from subunits G1 and G2 may be attributed to the pre-Roman and Roman/post-Roman periods, respectively.

Geoarchaeological question and the Antic site of Metz-Mazelle

23An archaeological rescue survey was conducted in 2009 in the city of Metz, in the Place Mazelle, situated on the floor of the Seille valley close to its confluence with the Moselle River. This research aimed to elucidate the nature of human occupation is this poorly-understood area in the vicinity of the Gallo-Roman settlement Divodurum, and to reconstruct its development. Up to 7 m of sediments have revealed a complex stratigraphy (Cordier et al., 2012; Champougny, 2013). Four main occupation phases have been recognised, spanning late Antiquity to the Modern period. The first unit (1.3 m in thickness) corresponds with the floodplain deposits of the Seille River (alternating sands and silts). This suggests that the Mazelle site was located some distance from the river channel(s). Typological and chronological analyses of the archaeological remains found in this unit, including a radiocarbon date, indicate that they were deposited at the end of the 2nd c. AD and during the 3rd c. AD. This age is consistent with an OSL age obtained from a sandy layer located in the middle of the unit (1685± 75 BP, i.e. AD 250-400). Subsequently, a large amount of refuse was deposited at the site, which was used as a dump for the ancient city. The midden material is ca. 30 cm in thickness and includes building materials (rubbles, tiles), decorative features (painted plasters, plinths, marble pieces), butchery wastes (cattle bones) and more than 5000 pieces of broken ceramics. Dating of the archaeological remains shows that this phase of deposition took place at the end of the 3rd c. AD.

24The dump is overlain by 2 m of backfill, which was stripped without any investigation (by order of the Service Régional de l’Archéologie), so that age and origin remain unknown. Finally, the remains of a canal, found at the southern edge of the Place Mazelle are thought to correspond with the building of a Mediaeval enclosure. Archaeological investigations show that this canal was renovated several times during the 12th and 13th c. AD. In the middle section, oak stakes were deposited in a hole at the bottom of the canal. These have been dated using dendrochronology, indicating that the trees were felled in about AD 1441-1444 (Willy Tegel, Dendronet, Labor für Holzanalyse, Bohlingen, Germany). This date corresponds with the restoration of the city ditches before it was besieged in 1444. The canal was in use until 1739, before being filled for the construction of the Place Mazelle.

25The study of the Metz-Mazelle has led to greater understanding of the historical evolution of this part of the city of Metz. However, the Roman sediments present a major problem; these are preserved at the bottom of the sequence, at the contact with the marly bedrock located at 163 m asl. This is surprising, since previous research made by Carcaud (1992) demonstrated that the present floodplain of the Moselle River and its main tributaries was of Lateglacial to Holocene age. Even if the location of the Mazelle site some distance from the Seille channel(s) is taken into account, the absence of pre-Roman fluvial sediments suggests that the Seille River incised into its older deposits (ca. 5m in thickness). No evidence has been found to correlate this erosion to a climatic event. Taking into account the chronological framework and the regional context, an anthropogenic origin can be assumed. However, in-situ clearing of the fluvial sediments seems unlikely due to the use of the Mazelle site. We therefore propose that a period of regressive erosion affected the lower Seille valley and thus the Mazelle site (Cordier et al., 2012). This erosion might be related to the construction of major infrastructure, related to the Antic city of Divodurum. An embankment and a pontoon (dated by dendrochronology to AD 119-249) were found 1 km downstream from the Place Mazelle (Rohmer, 1999). However, further research focusing on ancient Divodurum and on the historical evolution of the Seille valley is necessary to confirm this hypothesis.

26The investigations conducted Place Mazelle complement previous research related to antic Divodurum. This main city of the Mediomatrices had undergone notable development from the 1st c. AD, as indicated by the numerous monumental remains such as the amphitheatre, vast baths (partially preserved) and, a few kilometres southwards, the impressive Gorze aqueduct. Recent research performed in the Modern city centre (Arsenal-Place de la République area) showed that the Modern orthogonal road network was inherited from the Roman period. It also demonstrated that the city was more extended than previously expected, covering a surface of ca. 120 ha (Alix et al., 2009). In particular, the urban area extended towards South-East, as indicated by the remains found in the Porte des Allemands area (Ilots Turmel; Gama, 2001). These results contrast with those concerning the Place Mazelle, where fluvial instability is likely to have limited urbanisation before the building of the Mediaeval enclosure. Like the research in the Remerschen-Wintrange area, the study of the Metz-Mazelle site confirms that environmental dynamics were very important during the last thousand years, and it underlines the complexity of the interactions between human societies and their environment, both in rural and urban areas (Tab. 1).

Historical evolution of the Moselle fluvial pattern upstream from Metz

27Archaeological survey conducted in the Moselle floodplain upstream from Metz has also provided useful information on changes of the historical channel and changes of fluvial pattern. This is particularly the case near Vittonville (between Pont-à-Mousson and Metz). The study area is located on the eastern part of the valley floor and covers an area of 105,883 m². Micro-topographical analyses allowed the recognition of several south-north orientated straight palaeochannels, ranging in width from ten to a few tens of metres. These were filled during various historical periods, as shown by the archaeological remains found within them. In the eastern part of the area, a channel cut through the Lateglacial to Holocene coarse sediments is filled with silts and clayey sands. However, sparse gravels are also preserved. These gravels originate from the neighbouring north-south Roman road from Metz-Divodurum to the ancient vicus of Scarpone (near Dieulouard; Mocci, 1998). This provides a terminus ante quem for the channel abandonment. In a more northern palaeochannel, a wooden board was found at the base of oxbow-lake deposits and is thought to be part of a small boat (Galland et al., 2011). Dendrochronology has indicated a Modern age (AD 1632± 10; Willy Tegel, Dendronet, Labor für Holzanalyse, Bohlingen, Germany) for this artefact, and thus also for commencement of the filling of the channel. Comparison with old maps suggests that the number of channels decreased during more recent periods, leading to the concentration of water flow in a single meandering channel, as it is the case at present. However, the origin (climatic or anthropogenic) and timing of this change remains unclear: observations in the Scarpone area have shown that several active channels persisted into the 18th c. This underlines the importance of sedimentological analyses and numerical dating in association with the study of historical archives, to provide a better interpretation of recent valley floor evolution (Carcaud, 1992; Bonnefond and Carcaud, 1997; Ollive, 2007).

After Carcaud (1992), Fechner and Langohr (1994), Zolitschka and Löhr (1999), Ruffaldi (2000), Brou (2001), Damblon et al. (2009), Naton et al. (2009), and Cordier et al. (2012).
Tab. 1. Geoarchaeological synthesis derived from the main studied sites.


28Geoarchaeological research has been steadily expanding in the Moselle catchment for several years, with a particular emphasis on fluvial systems. These are the main agents of regional morphogenesis, and provide detailed records of palaeoenvironmental changes. Increasingly, research is incorporating new dating methods (such as OSL dating) to confirm the presence of archaeological sites spanning various Prehistoric and Historic periods. A scientific network, bringing together researchers from various disciplines (archaeology, geography, geosciences) is being created to continue this work. This overview should therefore be seen as a milestone in this ongoing evolution. Several areas for future work have been identified. In the Luxembourgian Moselle valley, research will focus on the description and dating of Unit F, in order to improve our understanding of human influences on the landscape. Further research on the Alzette valley floor should similarly allow a better understanding of the climatic influences that have been responsible for the evolution of this river since the Last Glacial Maximum. Finally, study of the Holocene to historical fluvial response to climate change and anthropogenic forcing (including the interactions between climate and human influences) will be the main focus of continuing research in the French Moselle catchment, especially in the Metz area.



Alix S., Gébus L., Gama F., Georges-Leroy M., Thion P., « Quartiers périphériques de Metz antique. L’apport des recherches récentes », Archéopages, no 24, 2009, p. 6-13.

Brou L., « Découverte d’une occupation Épipaléolithique ou Mésolithique ancien à Remerschen-Enner dem Raederbierg (Grand-Duché de Luxembourg). Présentation et implications », Bulletin de la Société préhistorique luxembourgeoise, no 20-21, 2001, p. 197-223.

Brou L., Le Brun-Ricalens F., Löhr H., Ziesaire P., Griette M., «Quid de l’Aurignacien entre les Vosges et l’Ardenne-Eifel?», in Bodu P., Chehmana L., Klaric L., Mevel L., Soriano S., Teyssandier N. (dir.), Le Paléolithique supérieur ancien de l’Europe du nordouest: réflexions et synthèses à partir d’un projet collectif de recherche sur le Paléolithique supérieur ancien du Bassin parisien, Actes du colloque de Sens, 15-18 avril 2009, Mémoire de la Société préhistorique française, no 56, sous presse.

Bonnefond J.-C., Carcaud N., « Le comportement morphodynamique de la Moselle avant ses aménagements », Géomorphologie: relief, processus, environnement, no 4, 1997, p. 339-353.

Carcaud N., Remplissage des fonds de vallée de la Moselle et de la Meurthe en Lorraine sédimentaire, thèse de doctorat, université Nancy 2, 1992.

Champougny A., «Étude géoarchéologique du site de la place Mazelle» in Brkojewitsch G., Marquié S., Sedlbauer S., Place Mazelle – Bassin retenue pollution, Rapport final d’opération, SRA Lorraine, 2013, p. 313-321.

Cordier S., Frechen M., Harmand D., Beiner M., «Middle and upper Pleistocene fluvial evolution of the Meurthe and Moselle valleys in the Paris basin and Rhenish massif», Quaternaire, no 16, 2005, p. 201-215.

Cordier S., Harmand D., Frechen M., Beiner M., «Fluvial system response to Middle and Upper Pleistocene climate change in the Meurthe and Moselle valleys (Eastern Paris basin and Rhenish Massif», Quaternary Science Reviews, no 25, 2006, p. 1460-1474.

Cordier S., Frechen M., Harmand D., «The Pleistocene fluvial deposits of the Moselle and Middle Rhine valleys: new correlations and compared evolutions», Quaternaire, no 20, 2009, p. 35-47.

Cordier S., Frechen M., Tsukamoto S., «Methodological aspects on luminescence dating of fluvial sands from the Moselle basin, Luxembourg», Geochronometria, no 35, 2010, p. 67-74.

Cordier S., Lauer T., Harmand D., Frechen M., Brkojewitsch G., «Fluvial response to climatic and anthropogenic forcing in the Moselle drainage basin (NE France) during historical periods: evidence from OSL dating», Earth Surface Processes and Landforms, no 37, 2012, p. 1167-1175.

Coûteaux M., « Étude palynologique des dépôts quaternaires de la vallée de la Sûre à Echternach et à Berdorf et de la Moselle à Mertert », Archives de la Section Sciences naturelles, physiques et mathématiques de l’Institut Grand-Ducal de Luxembourg, no 34, 1970, p. 297-336.

Damblon F., Buydens C., Hauzeur A., «Contribution de l’anthracologie à la connaissance de l’environnement de l’Homme néolithique au Grand-Duché de Luxembourg», in Le Brun-Ricalens F., Valotteau F., Hauzeur A. (dir.), Relations interrégionales au Néolithique entre Bassin parisien et Bassin rhénan, Actes du 26e colloque interrégional sur le Néolithique, Luxembourg, 2003. Mersch: Musée National d’Histoire et d’Art de Luxembourg, Service Régional d’Archéologie de Lorraine, Landesdenkmalamt des Saarlandes, Archaeologia Mosellana, no 7, 2007, p. 79-81.

Damblon F., Hauzeur A., Buydens C., « Étude anthracologique des occupations rubanées et protohistorique du site de Remerschen-«Schengerwis » (Grand-Duché de Luxembourg): utilisation du bois, environnement et chronologie», Bulletin de la Société préhistorique luxembourgeoise, no 27-28, 2009, p. 61-117.

Delagnes A., Jaubert J., Meignen L., «Les techno-complexes du Paléolithique moyen en Europe occidentale dans leur cadre diachronique et géographique», in Vandermeersch B., Maureille B. (dir.), Les Néandertaliens, biologie et cultures, Paris, Comité des Travaux Historiques et Scientifiques, 2007.

Fechner K., Langohr R., « Sols anthropiques et alluvions anciennes sur le site de Remerschen-Schengerwis: une longue histoire faite d’événements naturels et humains, état de la question », Bulletin de la Société préhistorique luxembourgeoise, no 15, 1994, p. 99-113.

Galland S., Ollive V., Antoine X., Dohr S., Glad A., Siafi S., Dolata J., «Vittonville (Meurthe-et-Moselle), La Lise, En Prêle, Le gravelot et la Morte, Les Gravelots», Rapport intermédiaire: Indices de circulation fluviale et terrestre dans la vallée de la Moselle depuis l’Antiquité, Inrap-Metz, Service Régional d’Archéologie de Lorraine, 2011.

Gama F., Metz, «Ilot Turmel, 16/24 rue du Wad-Billy», Bilan Scientifique 2001, Direction Régionale des Affaires Culturelles de Lorraine, 2001, p. 78-79.

Guillaume C., « Les gisements du Paléolithique inférieur et moyen de Lorraine », Bulletin de l’Association française pour l’étude du Quaternaire, no 19, 1982, p. 135-146.

Guillaume C., Janot A., « Stations de surface du Paléolithique inférieur et moyen de la vallée de la Meuse en Lorraine », Revue archéologique de l’Est et du Centre-Est, no 34, 1983, p. 5-17.

Hauzeur A., « Le Rubané au Luxembourg. Contribution à l’étude du Rubané du Nord-Ouest européen », Études et Recherches Archéologiques de l’Université de Liège, Dossiers d’Archéologie X-MNHA, no 114, 2006, 669 p.

Heuertz M. « Documents préhistoriques du territoire luxembourgeois. Le milieu naturel. L’homme et son œuvre », Musée d’histoire naturelle et Société des naturalistes luxembourgeois, no 1, 1969, 296 p.

Janot A. « Paléolithique inférieur et moyen de Lorraine, originalité des industries à quartzites », Bulletin de la Société préhistorique française, no 85, 1988, p. 291-303.

Jaubert J., « Chasseurs et artisans du Moustérien », Paris, Maison des roches, 1999, 152 p.

Lebrun-Ricalens F., «Le Paléolithique ancien du Grand-Duché de Luxembourg», Bulletin de la Société préhistorique luxembourgeoise, no 16, 1995, p. 17-31.

Le Brun-Ricalens F., Gazagnol G., Rebmann T., Thill-Thibold G., Thill-Thibold G., «La station moustérienne de Lellig-«Mierchen-Mileker», commune de Manternach (G.-D. de Luxembourg). Approche litho-techno-typologiques», Bulletin de la Société préhistorique luxembourgeoise, no 32, 2011, p. 57-94.

Meyrick R., «The development of terrestrial mollusc faunas in the ‘ Rheinland region’(western Germany and Luxembourg) during the Lateglacial and Holocene», Quaternary Science Reviews, no 20, 2001, p. 1667-1675.

Mocci L., « Arry (Moselle), Derrière le Moulin », Bilan Scientifique Régional, Inrap-Metz, Service Régional d’Archéologie de Lorraine, 1998.

Moine O., «West-European malacofauna from loess deposits of the Weichselian Upper Pleniglacial: compilation and preliminary analysis of the database», Quaternaire, no 19, 2008, 11-29.

Moine O., «Rapport sur la malacofaune des lœss du Weichsélien supérieur de Jongerbierg, Remerschen (Luxembourg)», 2010, 10 p.

Moine O., Antoine P., Rousseau D.-D., «The impact of Dansgaard-Oeschger cycles on the loessic environment and malacofauna of Nussloch (Germany) during the Upper Weichselian», Quaternary Research, no 70, 2008, p. 91-104.

Moine O., Antoine P., Deschodt L., Sellier-Segard N., «Enregistrements malacologiques à haute résolution dans les lœss et les gleys de toundra du Pléniglaciaire weichselien supérieur: premiers exemples du nord de la France», Quaternaire, no 22, 2011, p. 307-324.

Naton H.-G., Cordier S., Brou L., Damblon F., Frechen M., Hauzeur A., Le Brun-Ricalens F., Valotteau F., «Fluvial evolution of the Moselle valley in Luxembourg during late Pleistocene and Holocene: palaeoenvironment and human occupation», Quaternaire, no 20, 2009, p. 81-92.

Nicolas T., Le Brun-Ricalens F., « Une nécropole à incinération de l’âge du Bronze final à Remerschen-Klosbaam (Grand-Duché de Luxembourg) : premiers résultats », Empreintes, no 3, 2010, p. 18-24.

Ollive, V., «Dynamique d’occupation anthropique et dynamique alluviale du Rhin au cours de l’Holocène: Géoarchéologie du site d’Oedenburg (Haut-Rhin, France)», thèse de doctorat, université de Bourgogne, 2007.

Rebmann T., Le Brun-Ricalens F., Stead-Biver V., «Inventaire et déterminations préliminaires des matières premières siliceuses des stations moustérienne de Lellig-«Mierchen-Mileker» (Grand-duché de Luxembourg)», Bulletin de la Société préhistorique luxembourgeoise, 20-21, 2001, p. 77-144.

Rohmer P., «Metz-Boulevard Paixhans», D.F.S. de fouille d’archéologie préventive, Service Régional d’Archéologie de Lorraine, 1999.

Ruffaldi P., « Premières traces polliniques de Néolithisation des zones de base altitude de Lorraine (France) », Quaternaire, no 10, 1999, p. 263-270.

Ruffaldi P., « Nouvelles données sur l’évolution du couvert végétal au Tardiglaciaire et au début de l’Holocène dans les zones de basse altitude de Lorraine », Bulletin de la Société Préhistorique Luxembourgeoise, no 19, 2000, p. 31-40.

Wagner R., «Römische Siedlungsspuren bei Dreiborn «Laumeringen»», Hémecht, no 39, 1987, p. 239-254.

Zolitschka B, Löhr H., «Geomorphologie der Mosel-Niederterrassen und Ablagerungen eines ehemaligen Altarmsees (Trier, Rheinland-Pfalz): Indikatoren für jungquartäre Umweltveränderungen und anthropogene Schwermetallbelastung», Petermanns Geographische Mitteilungen, no 143, 1999, p. 401-416.

Table des illustrations

Légende Fig. 1. General map of the study area with location of the main sites.
Fichier image/jpeg, 840k
Légende Fig. 2. Stratigraphy and location of the archaeological findings in the sequences of the Remerschen-Wintrange basin.
Fichier image/jpeg, 478k
Légende Fig. 3. Schematic cross-profile of the Alzette valley at Lorentzweiler.
Fichier image/jpeg, 230k
Légende After Carcaud (1992), Fechner and Langohr (1994), Zolitschka and Löhr (1999), Ruffaldi (2000), Brou (2001), Damblon et al. (2009), Naton et al. (2009), and Cordier et al. (2012).Tab. 1. Geoarchaeological synthesis derived from the main studied sites.
Fichier image/jpeg, 333k


Associate Professor, University of Paris Est-Créteil (Paris 12), Mixed Research Unit (UMR 8591) CNRS/Universities of Paris 1 & Paris 12/INRAP (Laboratory of Physical Geography: Present and Quaternary Environments–LGP), Meudon, France (

Scientific Assistant, National Centre of Archaeological Research, National Museum of History and Art of Luxembourg, Bertrange, Luxembourg (

Professor, University of Lorraine, University Research Team (EA 1135; Studies and Research Centres on Landscapes–CERPA), Nancy, France (

Researcher, National Centre for Scientific Research (CNRS), Mixed Research Unit (UMR 8591) CNRS/Universities of Paris 1 & Paris 12/INRAP (Laboratory of Physical Geography: Present and Quaternary Environments – LGP), Meudon, France (

Associate Professor, University of Franche-Comté, Mixed Research Unit (UMR 6249) CNRS/University of Franche-Comté (Chrono-Environment Laboratory – LCE), Besançon, France (

Archaeologist, Centre of Preventive Archaeology of Metz Metropolis, Metz, France (

© CNRS Éditions, 2015

Conditions d’utilisation :