Version classiqueVersion mobile

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 6. The Holocene evolution of the Paris basin (France)

Contribution of geoecology and geoarchaeology of loodplains

Jean-François Pastre, Chantal Leroyer, Nicole Limondin-Lozouët, Pierre Antoine, Christine Chaussé, Agnès Gauthier, Salomé Granai, Yann Le Jeune et Patrice Wuscher

Texte intégral

Introduction

1The loodplains of the Paris basin show several rich and varied sedimentary sequences. These sequences allow us to reconstruct the various stages of the regional environmental evolution during the Holocene. The sedimentary fills in the large valleys (Seine, Oise, Marne…) mostly relate to their tributary palaeochannels (backfilled tributary channels next to the present tributary channels) and the sedimentary ills of the main river channels in the small valleys. They give complementary clues that unevenly cover the diferent periods and chronozones. Morphosedimentary (geomorphology, stratigraphy), palynological, and malacological studies allow us to test different explanations concerning palaeoenvironments. hey relate to climatological and anthropogenic evolution and to the general dynamics of the environmental systems in relation to the evolution of the vegetation and land surfaces inluenced by human activities.

2This study concerns the centre of the Paris basin and its surroundings (Fig. 1). The data have been obtained by conservational archaeological work, allowing us to produce large stratigraphic sections with mechanical shovels, and by augering, drilling and coring.The large valleys studied are the Seine, but more speciically the rivers Marne and Oise. These small valleys are drained by several tributaries, which have been studied locally (e.g., Nonette, Serre) or more extensively in large parts of their catchment (Beuvronne, Crould).

Fig. 1. General map of the studied area and sequences.

The sedimentary sequences and environnemental evolution

3The sedimentary fills commonly vary between 6 and 10 m deep and can reach up to 12 m in thickness. In the case of cut-and-filled sequences, the sumation of these layers can reach up to 20 m. The detrital sedimentary units (silt/clay) can correspond to erosion of the upper part of the riverbeds (large valleys), but more specifically indicate slope erosion during the second part of the Holocene (Pastre et al., 1997). Organic sedimentation, well developed in the small valley sequences during the first half of the Holocene, indicates an environnemental stability of the landscape and the absence of significant erosion.

4The floodplains of the large valleys are covered by thin silty-clayey sedimentation dating from the Subboreal and especially from the Subatlantic periods, which is due to sediment transport resulting from slope erosion. The largest quantity of Holocene sedimentation is concentrated in the river palaeochannel infills, which are located along the present main river channels. They are connected to the surrounding Pleniglacial and Lateglacial fluvial deposits through palaeo-riverbanks covered with later sediments. The main palaeochannels are often twice or three times wider than the present ones. They frequently show cut-and-fill sequences, which indicate (i) incision phases correlated more or less clearly to climate changes or (ii) some avulsions of the main river palaeochannels between several phases of abundant sediment yield. After the main channel incision at the beginning of the Holocene, sedimentary sequences can be observed in several palaeochannels at the beginning of the Atlantic, during the Subboreal or the Subatlantic (Pastre et al., 2002a; Fig. 2). These incision phases imply more or less important hiatuses that vary along the channels. The palaeochannels are not very frequent and they cover different periods, with their antiquity generally increasing with their distance from the present river bed. The palaeochannels infilled by peaty sediments date from the first half of the Holocene (Preboreal to Subboreal beginning), those with silty organic sedimentation date from the Subboreal and finally, the last ones with silty or siltysandy sedimentation along the present river channels are Subatlantic in age. It is not rare to find Lateglacial or early Holocene abandoned channels, which started flowing again during the Subboreal (Pastre et al., 2002b).

5These small valleys generally show Holocene sedimentation nearly completely covering their floodplains. These sedimentary infills indicate a progressive aggradation since the beginning of the Holocene, and cut-and-fill sequences are generally rare. The lower part of the sedimentary sequences shows organic deposits (peat, peaty silt), which are often calcareous (tufa, tufaceous peat) and date from the lower and middle Holocene. The upper part of the sequences is generally represented by clayey-silt deposited during the second half of the Subboreal but mainly during the Subatlantic (Île-de-France). Their thickness often decreases downstream. Peat formation has probably taken place until the present day (e.g., Picardie; Pastre et al., 1997).

Fig. 2. Main phases of cut-and-fill and sedimentation in the large valleys of the Paris basin (after Pastre et al., 2002a).

The rebalancing of the fluvial environments during the Preboreal

6The strong decrease in sedimentary deposition, which follows the Younger Dryas, suggests a rapid incision of the main channels of the large valleys, and an erosion of the valley-floor sediments or a more or less deep incision of the small valleys. This geomorphological change is enhanced by a phase of linear channel incision, which seems to progress downstream. Around 9100 BC, the main river channels of the main valleys seem to have universally reached a maximum level of incision indicating an important lowering of the water surface. Low rates of fluvial instability seem to be demonstrated by some sedimentary inputs along the riverbanks (e.g., Seine valley), some other areas are characterised by early peat deposition again indicating high stability (e.g., downstream part of the Marne valley). During this first half of the Preboreal, the transport capacity of the rivers significantly decreases. In the main channel of the Marne River, this phase is followed by low organic and mineral sedimentation (e.g., Vignely, Jablines). In the small valleys, peat formation can start very early in the bottom of the channels or locally over larger parts of the valleys (e.g., Crould in Bonneuil). Nevertheless, vast areas are still without peat bogs at this time. Pine-woods cover the slopes but locally also the valley bottoms (e.g., Nonette River) while birch and juniper woods and herbaceous heliophitic vegetation persist. The riverbanks were covered by Salix alba species (e.g., Marne valley). During the second part of the chronozone (after 8500 BC), pine-woods remain predominant but more mesophilous formations with hazels and oaks increase at the expense of birches and junipers.

7In the malacological faunas, the climate warming causes the migration of the pioneer species of the Late glacial towards Scandinavian and mountain refuges. As a whole, open land species decrease while the first forest mollusca appear (Limondin and Rousseau, 1991; Limondin, 1995; Limondin-Lozouet and Antoine, 2001). The change of the forest cover during the Preboreal/Boreal, which is shown by the decrease of the conifers to the benefit of broad-leaved trees, is indicated in the malacofaunal composition by the succession of Discus ruderatus faunas followed by Discus rotundatus faunas. This close correlation between the malacological and palynological signal is well kwown in Normandy (Limondin-Lozouet, 2011). The Discus transition is recorded in several sequences of the Somme valley (Limondin, 1995) and at Warluis in the Thérain valley at the beginning of the Boreal (Coutard et al., 2010). During the second part of the chronozone, most of the sequences show a continuity of organic sedimentation but some of them still indicate some detrital sedimentary episodes, which imply some hydrodynamic instability. This might be due to the Preboreal oscillation, ca. 9200 BC (Magny, 1995; Bohncke and Hoek, 2007; Magny et al., 2007) but this correlation has yet to be established. Some sequences also seem to indicate more or less prolonged hydrographic adjustments. Furthermore, the palynological signal of short climate perturbations, which characterise the first part of the Preboreal (Magny et al., 2007), seems to be recorded at Villiers-sur-Seine. The significant thickness of the deposits and the high resolution of the studies show in that area three episodes of pine-trees decrease being replaced by birches, willows, and herbaceous plants, which increase between 9700 and 8900 BC. They can be compared to the three events recorded in the Netherlands during this period (Bohncke and Hoek, 2007).

Stabilisation in the Boreal and the hydrodynamic changes during the lower Atlantic

8The beginning of the Boreal is marked by an important stabilisation of the fluvial enviroments and the valleys are characterised by a significant extension of peat formation. These peats aggradate and progradate from the lower part of the flood plains to the riverbanks. They rapidly occupy the majority of the floodplains of the small valleys and progressively colonise the riverbanks of the tributary channels in the main valleys. At the same time, organic and minerogenic units rich in organic matter and carbonates issued from chalk of the upstream catchment of the Seine (e.g., Marne River) and progressively fill the river channels. The growth of tufaceous, bedded peat indicates a strong contribution of organic material from the riverbanks or from the aquatic flora (fragments of herbaceous plants, moss, abundant woods, humified organic matter). Most of the carbonates derive from the erosion of the upstream part of the catchments (headward erosion). The bedded sedimentation, which often shows a seasonal alternation of organic matter and carbonates, indicates the calm and regular regime of the river flows. In the most carbonate-rich environments, the sedimentation also shows the development of tufas with algal balls (e.g., Marne River). These conditions seem common during the whole of the Boreal. During this period, large hazel woods grow at the expense of pine-woods and oak whereas elm woods progressively expand (Leroyer et al., 2011).

9The sedimentary sequences of the small valleys and those of the palaeochannels in the large valleys are characterised by continuous peat formation until the recent Atlantic, indicating a great stability in the rate of the sedimentation. Tufaceous peats facies are frequent in some valleys (e.g., Crould, Beuvronne), however, the partial truncation of the Boreal and the local supply of silt revealed in several cores of the Marne and Oise valleys imply an increase in hydrodynamic activity at the Boreal/Atlantic transition and/or during the lower Atlantic.

10Sedimentological data obtained in the sequences of Vignely and Neuilly-sur-Marne in the Marne valley, and from Boulages in the Aube valley, demonstrate the importance of the hydro-sedimentary changes during this period. In Vignely, the rapid transition of the tufaceous bedded peat in the main river channel to marly silt indicates a first increase of the detrital sediment transport ca. 7469-6861 BC. Then, a unit of carbonate-rich silt terminates organic sedimentation ca. 6999-6448 BC. These carbonate-rich units, which issued from the upstream chalk, seem to be the result of a linear erosion phase upstream and are not related to erosion of the interfluves covered by Pleistocene loams. The instability of the fluvial flows continues during a part of the lower Atlantic, leading to channel incision of some main river channels (Marne and Oise). In Boulages, a short gravely strata is intercalated with organic sedimentation of a channel fill of the Aube River between 7499-7055 and 7302-6652 BC. These phenomena can be correlated with a meandering river pattern in Noyen-sur-Seine between 7289-6645 and 7083-6598 BC (Marinval-Vigne et al., 1989). This hydrodynamic episode seems to be correlated with wet conditions and increased hydrodynamic activity as reported by Starkel et al. (1996) in several fluviatile sequences observed in Europe, especially in the Vistula catchment between 7550 and 6650 BC and in the Mark catchment just before 6850 BC (Bohncke et al., 1987). If the period ca. 6800 BC does not directly correspond to a precise climate deterioration, the observed phenomena could constitute a slightly delayed repercussion of the cooling that occurred during the second part of the Boreal. This cooling is illustrated by the increase of the Alpine glaciers (Bortenschlager, 1977) and increased levels of the Jura lakes during the phase of “Joux” (Magny, 1992, 2004), and the lowering of the timberline in the Swiss Alps ca. 6930 BC (Burga, 1987). If these changes take place earlier than the ‘8200 cal. BP event’ (ca. 6250 BC) as recorded in the Greenland cores (Alley et al., 1997), and the hiatus of the lower Atlantic shown by numerous sequences takes place at the same time we need to consider its meaning in terms of regional fluvial dynamics.

11The parallel influence of Mesolithic human activities on the environment, although perceptible, is, however, been very limited. In Noyen-sur-Seine, the influence of the last groups of hunter-gatherers between 7236-6635/6362-6017 BC and 6048-5747/5374-5007 BC is reflected in the pollen spectra by an increase of the ruderal species and a decrease of the woods but without a general decrease of arboreal pollen, and only next to the sites (Leroyer, 2004). The frequency of charcoal in the contemporaneous sediments does not allow us to specify their origin: domestic activities, burnings, anthropogenic or natural fires. The palynological spectra show the contemporaneous change of the regional vegetation (Leroyer, 1997). After a decrease in hazel to the benefit of oak and elm trees from 7550 BC, the lower Atlantic shows a slight increase of pine whereas oaks continue to increase and lime appears. Rather than a slight openness of the natural environment, this pattern may indicate changes in the oceanic or continental trend of the climate. In England, the hydrodynamic changes, also observed during this period, are imputed to an increase of the oceanic nature of the climate (Brown, 1997). A strong hydrodynamic activity is also confirmed during this period in the Vistula catchment (Starkel, 1995).

12During the Boreal and the lower Atlantic, molluscs living in forested environments become predominant and constitute generally less than 25% but up to 60% of the populations. The other gastropods have broader environmental requirements and can also adapt locally to forested environments. Except for the specific aquatic environments, the aquatic faunas are generally residual and the malacological populations indicate a well-drained environment (Limondin and Rousseau, 1991; Limondin-Lozouet and Preece, 2004).

13During the second half of the lower Atlantic, after 5800 BC, organic sedimentation increases and seems to indicate some hydrodynamic stabilisation. In the main water channels of the large valleys, this sedimentation is marked by renewed peat formation, and the deposition of peaty silt, and organic and minerogenic silts. In the small valleys, peat and tufaceous peat deposits still accumulated without significant change.

The relative stability of the upper Atlantic

14During the upper Atlantic (4880-3500 BC), valley deposition of organic matter is maintained at a significant rate. The homogeneity or the bedding of the deposits mostly indicates quiet flows and low discharge variability. The example of the sequence of Armancourt in the Oise valley, which is characterised by fine organic sedimentation ca. 4655-4166 BC, illustrates these trends (Fig. 3). The upper Atlantic also shows a widespread development of tufas as in the Seine valley (Chaussé et al., 2008) or in the Marne valley (e.g., core C11 of the Haute-Ile site in Neuillysur-Marne). However, the observed increase of clay and silt probably shows the subtle erosive impact of lower and middle Neolithic agricultural activities. In the small valleys, peat and tufaceous peat continue to deposit without particular changes and the example of the Biberonne valley in Compans (Beuvronne catchment, tributary of the Marne River, Seine et Marne department) illustrates this type of sedimentation (Fig. 4). Infrared spectroscopic mineralogical analysis does not show micro-quartz occurrence, which would imply the erosion of the Pleistocene loams on the slopes. The carbonate-rich content, derived from tufas and the organic matter of the peat formations, indicates autochtonous biogenic sedimentation without detritic inputs related to human activities.

Fig. 3. Evolution of the sedimentation of the main palaeochannel of the Oise River in Armancourt (Oise). 1: loams and stones (embankment); 2: mixed loams (by coring); 3: bedded organic clay with ligneous fragments; 4: beige silty clay; 5: bedded dark grey organic and minerogenic clay; 6: grey-black organic clay; 7: grey silty clay; 8: dark grey organic and minerogenic clay; 9: gravels and small peebles (silex); 10: dark grey silty clay; 11: bedded organic and minerogenic clay with ligneous fragments; 12: fine grey sand; 13: organic and minerogenic silt; 14: bedded organic clay; 15: sand and gravels.

15During the lower and middle phases of Neolithic, the anthropogenic effects recorded by palynological data are only highlighted in the sequences that come from a major archaeological context. The stratigraphic section of Pont-Sainte-Maxence in the Oise valley clearly shows three occupations of the site. The sites in the downstream part of the Marne River show a strong influence of the Villeneuve-Saint-Germain (VSG) settlements of farmers and a decrease in anthropogenic markers during the middle Neolithic (Fig. 5). Meanwhile, the region of La Bassée (Seine valley), where the lower Neolithic sites are rare, still exhibits very low anthropogenic impacts. However, the region will be characterised by anthropogenic signals during the Middle Neolithic in connection with the numerous sites from this period (Leroyer, 1997). Apart from agricultural and pastoral activities, the sequences related to archaeological sites show the local land clearing of alders, oaks and lime trees with sometimes a preferential use of oak. However, the influence of these human groups remains rather limited and does not produce any appreciable increase of the anthropogenic impact on the vegetation during the middle Neolithic (Leroyer, 1998, 2006a). Sedimentological and palynological data thus converge to define a relatively closed forested landscape with active regeneration and without massive deforestation of the landscape. In the small valleys, the formation of significant tufa beds continues until ca. 2300 BC (Limondin-Lozouet and Preece, 2004). Forest malacofaunas prevail and indicate closed environments and low anthropogenic influence (Fig. 6). At the same time, the global curves of diversity indices show a close correlation between malacofaunas destabilisation and the local phenomena of river flooding (Rousseau et al., 1993).

Fig. 4. Evolution of sedimentation in the Biberonne valley (Beuvronne catchment) in Compans (after Orth, 2003, modified). 1: present soil; 2: palaeosoil; 3: silt; 4: organic and minerogenic silt; 5: humified black peat; 6: russet coloured peat; 7: brown peat; 8: organic and minerogenic peat; 9: tufa; 10: interbedded silt, peat and tufa; 11: peaty silt; 12: sandy silt; 13: silty sand; 14: sand; 15: clay, sand and gravels; 16: sand of Beauchamp; 17: marly limestone of Saint-Ouen; 18: calibrated radiocarbon date BC; 19: stratigraphic unit.

Fig. 5. Pollen diagrams of anthropogenic influence (anthropisation) in several Neolithic sequences from the Paris Basin. The criteria for anthropogenic impact or “anthropisation” is indicated by the percentage of ruderals, Plantago lanceolata and Cerealia present in each palynological spectra. 1: local occupation; 2: occupation in the surroundings; 3: calibrated radiocarbon date in BC; 4: cereals; 5: Plantago lanceolata, 6: other ruderals.

Fig. 6. Biozonation of the Holocene tufa of Saint-Germain-le-Vasson (Calvados) after the malacological faunas (modified after Limondin-Lozouet and Preece, 2004). (a) occurrence of the species in the stratigraphical unit associated with the main and lateral section and the core 2; (b) occurrence of the species in the core 1 that comprises the whole part of the tufa; (c) correlation after the malacofaunas. The species Discus rotundatus, for which the increase corresponds to that of the hazel, characterises the occurrence of the broad-leaved tree forests in the Holocene (Limondin-Lozouet et al., 2005) and offers the basis of biostratigraphical correlation. The appropriateness of this choice is corroborated by the coherence in the spectra of the other key species. The sequence of Saint-Germain-le-Vasson gives the malacological succession of reference for the first half of the Holocene. The six malacological zones are characterised by key taxa remarkable for their constrained ecological demands and/or their limited geographical distribution. The malacofaunas indicate the development of swampy open-lands at the beginning of the Holocene (SGV 1), then the development of a forested environnment since the zona SGV 2, be marked by a decrease of wetness all along the Atlantic (SGV 3 to 5), and which finally becomes clearer at the end of the climatic optimum. The dates give a precise chronological framework to this succession and permit precise comparisons with other indicators (Limondin-Lozouet and Preece, 2004). The dates in bold are in years BP and the dates in normal letters are in years BC.

Climatic and anthropogenic changes during the Subboreal

16The beginning of the Subboreal exhibits an increase of fluvial activity and the first and typical elements of a significant environmental change. In the main channels of the rivers Oise and Marne, an increase of the silty-clayey or sandy beds is locally observed ca. 3500 BC, e.g. in Armancourt (Oise) between 3778 and 3103 BC (Fig. 3) and Annet-sur-Marne (Marne) between 3646 and 3364 BC (Pastre et al., 1997, 2002 a and b). These changes, influenced by the climate, have probably increased because of upper Neolithic human activities. This period corresponds to the beginning of the so-called “Chalain” phase (Magny, 1992, 2004), which is characterised by the rise of the Jura lakes ca. 3700-3250 BC. At the same time, the palynological sequences show an increase in agricultural practices during the late Neolithic in accordance with the recent discovery of numerous sites of this period (Salanova et al., 2011). Despite the weaker correlation of the sequences with the archaeological sites, the presence of cereals and ruderals is almost universally recorded (Fig. 5). The impact on the forest also appears more important even if it is difficult to see this influence due to the filtering effect by the alder populations in the valley bottoms (Leroyer, 2003). The great sedimentary changes that characterise the turning point of the end of the Neolithic/Chalcolithic takes place ca. 2550 BC. In the main river channels of the large valleys, this change is demonstrated by major silty-clayey units, which indicate a first major disturbance of the pedogeneised loam cover on the slopes. In the Oise valley, the organic sedimentation (peaty silt) of the lower part of the riverbanks, well dated by late Neolithic remains, is replaced by clayey and silty sedimentation that contains archaeological remains from the end of the Neolithic (e.g., site of Lacroix-Saint-Ouen; Pastre et al., 1997). The decrease of the organic carbon (from 5% to less than 2%) is synchronous with an increase in clayey sedimentation, which indicates the settling and flocculation processes on the riverbanks. The causes of this change are probably various and it is difficult to determine the respective role of anthropogenic and climate factors. Although it may have been increased by several aggravating factors, it can be also be explained as a consequence of the progressive openness of the landscape and the development of new and not easily reversible sedimentary conditions, although variable in detail. The hypothesis of climate deterioration acting as a triggering factor, as during the second part of the “Chalain” phase ca. 2900-2850 BC (Magny, 1992, 2004), can be considered, but the decrease of Jura lakes levels and the retreat of the Alpine glaciers, which imply an improvement of climatic conditions after ca. 2650 BC (ibid.), does not support the determining role of climate. An increase of the prevalence and importance of fire can be correlated with the frequence of micro-charcoal and even with more significant overbank units near the archaeological sites, but unfortunately the fluvial context does not allow us to estimate their volume. It cannot be denied that such a sedimentary input suggests relatively large clearings on the slopes and sufficiently persisting tiled arable land surfaces.

17In the small valleys, organic sedimentation continues, indicating the limits of the opening-up of the landscape. This deterioration of environmental conditions reaches its ‘climax’ during the second part of the Subboreal during the lower and middle Bronze Ages with signiicant river-discharge variability. In the large valleys, the incision of secondary river channels adjacent to the main ones can be frequently observed since the beginning of the Holocene. Even if the incision phenomena are short-lived, they imply important variation in intensity and duration of sedimentary phenomena depending upon their geographical location. In Vignely in the Marne valley, the progressive infilling of the Subboreal channel thus begins ca. 2460-2030 BC and ends ca. 1406-1060 BC. In the middle Oise valley in Longueuil-Sainte-Marie, a small channel is marked by a short phase of organic and minerogenic sedimentation ca. 2350 BC before a substantial new phase of silty sedimentation. In many small valleys and in some channels of the large valleys (e.g., Marne), peaty sedimentation can extend beyond 1750 BC, but the period between 1750 and 1250 BC shows, almost everywhere, important new phases of silty overbank deposition that cover the riverbanks more or less widely and extensibly. This evolution is well illustrated by the example of the Biberonne valley in Compans where the last peat deposits covered by more than 1 m of silt are dated to ca. 1750 BC. In the Crould valley (Val d’Oise), the peat deposits are still forming much later on, until ca. 1050 BC. This period of silty sedimentation augments Holocene soil developed on the riverbanks begun during the irst half of the Postglacial. In some channels, signiicant incision and sedimentation episodes, such as in the Esquillons channel in Houdancourt (Oise), reveal substantial hydro-sedimentary activity approximatively between 1750 and 1250 BC (Pastre et al., 1997). Although this second half of the Subboreal is characterised by a significant increase of detrital sedimentation, the silt deposits of the main valleys generally retain a slightly organic composition. This strong hydrodynamic activity is probably explained by an increase of precipitation, which partly coincides with the transgressive lacustrine phase of “Pluvis” (Magny, 1992, 2004). This activity can be compared with that observed in several French rivers (e.g., Lefèvre et al., 1993) or in the NW part of Europe (e.g., Brown, 1997; Macklin, 1999).

Fig. 7. Sedimentary sequence of the Serre valley in Courbes “Les Patures” (Aisne) – (core C1).

18Despite the relative scarcity of archaeological sites of the early and middle Bronze Age (Bourgeois and Talon, 2005; Brunet et al., 2008; Salanova et al., 2011), the pollen profiles indicate the maintenance of agricultural and pastoral activities during these periods. Palynological markers of anthropogenic origin are systematically recorded and even appear slightly more important than in the late/final Neolithic. Consequently, human influence in forest history remains difficult to estimate because of the natural filter of the alders in the valleys. The ambiguity of anthropogenic markers at this time is not suprising considering that the increase in regional human settlements during the late Bronze Age is not clear and the agricultural signs increase only very slightly. Furthermore, this period corresponds to a strong reduction in woodland cover in the valley bottoms replaced by humid meadows (Leroyer, 1997). In the Seine valley, during the Subboreal, the molluscs living in the forests decrease whilst openland molluscs and palustral faunas increase. Localy, the aquatic faunas predominate (Limondin and Rousseau, 1991). The diversity index of terrestrial faunas decreases sharply (Granai et al., 2011) and this indicates that the number of the faunas is unevenly spread between the different vegetational species. This inequality is correlative to the progressive domination by species typical of humid meadows Vallonia pulchella, regularly observed in different malacological successions of the valley. On the whole, the malacological populations suggest relatively unstable environments. In Burgundy (NE France), the decrease of the diversity index is local and correlates with the archaeological sites, the malacological signal demonstrating the increasing impact of humans on the landscape (Rousseau et al., 1993, 1994).

Environmental variability during the Subatlantic

19The Subatlantic is marked by a strong, spatial variability of morpho-sedimentary responses in conjunction with local factors. This variability is related to the different environment management practices utilised by different societies, to variable disturbances by river flows and disturbances to river courses by human actions, as well as finally to an increase of the variability of socio-economical factors.

20The beginning of the Subatlantic ca. 850 BC is characterised by a climatic deterioration linked to an increase in wetness (Van Geel et al., 1999) and groundwater levels (Bravard et al., 1990; Magny and Richard, 1992), which explains the growth of new, small, peat units in the small catchments of the Plaine-de-France. In the Crould catchment, the peat deposits developed in the small valley Aulnaies-du-Pont in Goussainville start developing ca. 550 BC (Pastre et al., 1997) while in the Beuvronne in Claye-Souilly, a new peaty phase ends between 780 and 370 BC (Orth, 2003; Orth et al., 2004). The development of beech trees and the appearance of hornbeam for the first time observed during this period (Leroyer, 1997, 2006b) could have been favoured by this humidity. In the large valleys, the relative uniformity of the silty fills and the lack of precise chronological markers do not allow us to clearly distinguish this period from the surrounding periods. The environment corresponding with the early Iron Age is not well known except in the Bassée in the Seine valley (Peake et al., 2005). Most of the palynological sequences seem to indicate a slight increase of human impact, nevertheless, the analysis of thick sequences, such as in Bazoches-lès-Bray in the Bassée, shows that a same location can reveal more or less marked anthropogenic conditions in conjunction with human settlements and that forested areas still remained during this period (Leroyer and Allenet, 2006). During the middle Iron Age, the important development of farming sites (Malrain, 2000) suggests an increase in the openness of agricultural lands and an increase of agricultural activities. However, this evolution is not “linear” as revealed by long sequences such as that from Fresnessur-Marne. The analysis of more detailed profiles shows that the anthropogenic influences can be less obvious than during the early Iron Age and that phases of agricultural decline can still be observed (Leroyer, 2006b; Leroyer and Allenet, 2006). In the middle Oise valley, silty units with some organic matter imply the progressive infilling of several lateral channels as in the typical sequence of Lacroix-Saint-Ouen (Pastre et al., 1997). This evolution suggests a decrease in fluvial activity and in discharge, and sediment loading, although there is a continuation or even an increase of the silty deposition. This increase in silty deposition is observed in some densely populated small valleys of the middle Iron Age, such as in the Crould valley in Louvres (Orville castle sequence) where silty units follow organic and minerogenic silt units between 350 and 50 BC. The local sedimentary impact of La Tène settlements can be relatively important as demonstrated in the Mauldre valley in Jouars-Ponchartrain (Blin et al., 1999). This period thus appears as a transitional period with a significant openness of the landscape.

21During the Roman period, the settling of clayey silt derived from the brown soils on the slopes continues and from then on it is constant although the spatial variability of sedimentary drifts can also be observed. This is probably related to the density of settlements and to the types of agricultural and pastoral systems within the catchments. The clayey silt shows homogeneous facies, which are less and less organic (total organic carbon <1%). The average water levels in the channels of the large valleys, which are reduced in capacity by this silty sedimentation (Pastre et al., 1997), provide difficult conditions in which to reconstruct hydrodynamic and environmental trends with precision. The similarity of the deposits, their reworking, or their scarcity, makes interpretation very difficult. New phases of undermining riverbanks or channel incision are often followed by phases of sedimentation where sandy sedimentation can be significant. Under these conditions the small valleys are now more informative, nevertheless the variability of their records frequently relates only to local influences. Detrital episodes, as at Claye-Souilly or Fresnes-sur-Marne in the Beuvronne valley or in Goussainville in the Crould valley, seem to indicate more local changes caused by neighbouring settlements rather than a massive and widespread response of the environment to the openness of the agricultural lands. Palynological data from the High Roman Empire period, although very rare, also suggest quite strong variability of the vegetation and environment. A general stability of palynological markers is observed in the downstream part of the Marne catchment stemming from the widespread openness of the environment as shown by palynological analysis from Balloy in the Bassée area (Seine valley) or Dourdan and Jouars-Ponchartrain (Mauldre valley) in the Yvelines (Leroyer, 1997; Blin et al., 1999; Leroyer and Allenet, 2006). These differences can be explained by the proximity of these sequences to ancient settlements in a landscape also well marked by agricultural and pastoral activities as is also recorded in the ponds near settlements at Senart and Dourdan. During the Lower Roman Empire, palynological spectra display a slow decrease of anthropogenic markers without indicating a real decrease of human influence (Leroyer, 1997). The local development of gley and peat observed in some small valleys (e.g., Beuvronne, Esches, Serre) could imply important drainage perturbations caused by human activities.

22At the end of Roman period and during the early Middle Ages, detrital sedimentation seems to decrease in a significant way. The absence of silty-clayey deposits between Gallo-Roman and Merovingian and Carolingian sites observed in some small valleys (Esches, Serre) forces us to consider the importance of the decrease of agricultural habits and the increase in pastoralism. In some valleys, peaty areas can increase like at Courbes in the Serre valley between AD 254-414 and 900-1146 (Fig. 7) or at Claye-Souilly in the Beuvronne valley between AD 174 and 533 (Orth, 2003; Orth et al., 2004). Local perturbations of the drainage at the end of the Roman period or at the beginning of the Middle Ages could constitute their origin. Palynological data indicate various situations. From the end of the 5th to the 7th c., a clear decrease in human impact accompanied by a large increase of forest is recorded on the plateaus, in Dourdan as in Sénart. However, the agricultural and pastoral activities have not totally disappeared. In the Vanne catchment, the anthropogenic markers are reminiscent of the predominance of pastoralism over cereal cultivation while small woods remain (Leroyer, 1997). During these same periods in Paris “Quai Branly”, the occupancy of the Seine riverbanks implies important anthropogenic influence (Chaussé et al., 2008). Then, from the 7th to 11th c., the Vanne catchment reveals a stronger human influence than the plateaus (Dourdan and Sénart forests). The progressive increase of detrital sedimentation demonstrated by several sequences with organics during the lower Middle Ages (e.g., Goussainville, Les Aulnaies-du-Pont, Plaine-de-France, Hirson, upper Oise valley) seems related to an increase of cultivation as shown by the palynological data between the 12th and 14th c.: cereals increase and seem to be accompanied by fruit cultivation (chestnut, walnut, vineyards), possibly with textile plants (hemp) in the valley bottoms (Vanne catchment, Seine in Paris “Bercy”), and in the plateaus of the Dourdan and Sénart forests (Leroyer, 1997).

23In the Beuvronne valley, the increase of microquartz related to silty sedimentation is clearly highlighted from AD 895-1031. However, this seems to increase from the 15th c., particularly where there is a close correlation with new, strong erosion. In the Crould catchment (Plaine-de-France), these silty units start covering the ruins of the Orville castle in Louvres or bury the last clayey and organic beds of the sequence of Aulnaies-du-Pont in Goussainville dated ca. AD 1300-1432. The silty-clayey sedimentation is particularly marked in the upper part of small catchments (e.g., Crould, Beuvronne) and in the upper catchments of the large valleys (e.g., Aube, Oise, Aisne) where the slopes are steeper. It coincides with the correlative and significant decrease of the vegetal cover shown in some palynological sequences such as the ones of the Vanne catchment. The upper part of these sequences, subsequent to the 14th c., shows a significant decrease in arboreal pollens (Leroyer, 1997; Pastre et al., 2002 a and b). The obvious increase of this trend during the following centuries produces fills of up to 2 m thick in the upper parts of the catchments. The ploughing of slopes previously used for pasture or woodland and the probable increase of the rainfall during the Little Ice Age have probably favourised intense episodes of erosion and rapid sedimentation. In the upper catchment of the Oise River (e.g., Oise in Hirson, Serre in Courbes), an important component of the silt covering the floodplains seems to date from the Modern period. In the Crould valley in Louvres, a basin dated from the 16th c. or of the beginning of the 17th c. has been rapidly infilled by no less than 22 silty sedimentary flood-units (i.e., 22 flood events) and after the beginning of the 18th c., it was then covered by about 2 m of silt. However, the lack of sufficient archaeological data and radiocarbon dates, due to the predominance of mineral-rich sediments, makes it difficult to provide a precise chronology for these sedimentary units. If the 18th c. seems to be a period of intense erosion, the “reading” of very homogeneous silty-clayey sequences does not allow its precise chronostratigraphic characterisation. The main feature of the silty profiles of the Modern period is the almost complete lack of interbedded soils and the systematic presence of surface anthrosols, enriched by organic matter and well developed (30-40 cm on average). These elements coincide with a progressive and rapid silty sedimentation, which seems to start before the 19th c. It is followed by the important processes of organic improvements and pastures during the 19th and 20th c. If the resulting stratigraphic picture appears to indicate a recent decrease in silty sedimentation due to an improved agricultural practices and climate, this interpretation must be accompanied by strong reservations. The change of the superficial drainage by agricultural improvements and the expansion of clearings and canalisation since the 19th c. have largely modified the sediment transport in the rivers. They have thus limited the alluvial deposition in the floodplains with more sediment being associated with natural in-channel discharge (suspended or saltation transportation) or artificial channel retention, which is removed through clearings and dredging. The homogenisation of the superficial soil horizons of cultivation (Ap horizons) by ploughing permitted the mixing of the recent colluvial and alluvial beds in these anthrosols. It is thus not surprising that we do not find their signature in pedological profiles. To the contrary, a few sequences from the upper part of small valleys show the importance of the recent runoffand damage to the silty soils on the slopes. The frequency of silty floodplain deposits and the importance of the silty fraction reworked from the in-situ loess both imply the importance of recent runoffdue to openfields expansion and to the marked destruction of the loess cover. On the slopes, horizons A and B have mostly been eroded offand they can be found in a reverse position in the upstream stratigraphic sections of the small valley bottoms (Pastre et al., 2006).

24Concerning the Subatlantic, the malacological sequences, even if they are not well dated, always show a majority of open-land molluscs dominated by Vallonia pulchella (Limondin and Rousseau, 1991; Granai et al., 2011). Among these open-land species very robust species like Pupilla muscorum or species favourised by human activities like Helicella itala. These malacological associations reveal a low and homogeneous vegetation offering only a few niches to the molluscs. The diversity index is stabilised at a low level. In stratigraphic context, the appearance of this malacozone frequently coincides with alluviation related to the frequence of floods, which deposit argilaceous silt. In the malacological assemblages, the Subatlantic is characterised by an increasing predominance of anthropogenic species to the detriment of other species (Rousseau et al., 1992). The Subatlantic thus appears as a complex period where the variability of the sedimentary sequences depends very much on the different ways of cultivating arable slopes and the reaction of the sedimentary environments. The relatively imprecise resolution of these data does not always correspond to the environmental precision of some historical sources. Nevertheless, the repeated comparison of the data between several thick sequences will progressively increase our chronological resolution.

Conclusions

25The sedimentary archives of valley bottoms in the Paris basin document the varied regional environments during the Holocene. Important changes in fluvial environments react to the major phases of landscape evolution. The variable sensitivity of the various morpho-sedimentary units is coupled with a variability of each biological or sedimentary indicator. In a less contrasted geomorphological context and with environments favourable to vegetation expansion, the balance with climatic factors can be both weighed and considered. The observed modifications thus indicate significant changes as outlined below:

  • The first half of the Holocene overall appears as a period of relative environmental stability. However it does not show the hydro-sedimentary stability that numerous peaty sequences of the small valleys would imply. The data concerning the main palaeochannels of the large valleys indicate a more variable hydrosedimentary evolution.
  • During the Preboreal, the contrasts that characterise early organic sedimentation and detrital or recurrent erosive episodes reflect the persistence of some instability of the environments. The resultant geographical variability results from several factors: fragility and heterogeneity of fluvial, biological population, variability of geomorphological context, climate variability.
  • The organic sedimentation, which everywhere characterises most of the Boreal, shows an important stabilisation of the valley environments. The importance of the biogenic control reduces the influence of the climatic deterioration during the second part of the chronozone. The increase in detrital material and the channel incision at the end of the period continue during the beginning of the early Atlantic, indicating a renewed hydrodynamic activity probably due to the climate. During the remainder of the early Atlantic and during the upper Atlantic, the renewal of the organic sedimentation is a parallel development with the improvement of the climatic conditions. The slight increase of the mineral load and local sediment instability probably reflects the influence of early and middle Neolithic societies. Their impact in palynological diagrams is an increase of the ruderals and the presence of cereals in the immediate surroundings of archaeological sites. Nevertheless, neither the sedimentological data nor the palynological data indicate a clear environmental deterioration of these environments, with restoration capacities still being important. However, the environmental deterioration progressively increases during the Subboreal.
  • Around 2550 BC, the renewed silty-clayey sedimentation and an increase in fluvial activity during the first third of chronozone precede a more important hiatus. The deposition of the first units of fluvial silt derived from the pedogeneised loess cover is linked to vegetation opening and slope erosion. The environmental impact of human societies during the later and final parts of the Neolithic are recorded in both the palynological diagrams and the sedimentary sequences. The continuation of this evolution and the incision of numerous lateral channels during the second half of the chronozone (Bronze Age) indicate a strong deterioration of the environmental conditions. The stages of this morphodynamic ‘crisis’ seem to be the consequence both of the deterioration of the ecosystems at the end of the Neolithic and at the beginning of Protohistory and the repeated deterioration of climate. The first silty units are seen in the small valleys and indicate an increase in erosion on the interfluves and woodland clearings.
  • The Subatlantic reveals the increasing extent of human impact on the landscape and a greater heterogeneity of impacts. The second Iron Age (La Tène) is marked by an increasing woodland clearance and the development of cultivation, which continue during the Roman period. The early Middle Ages show a decrease in erosion and very variable anthropogenic influences on the environnment. The renewal of the sedimentary aggradation during the late Middle Ages accompanies an increase in agricultural pressure. During the Renaissance and the Modern period, agricultural activities and the climatic deterioration of the Little Ice Age induce a significant erosion of the loam cover on the slopes. This phenomenon is correlative of a further major opening-up of the landscape.

26This evolution shows the increasing complexity of the interactions between climate change and human activities, the variability of the data, research possibilities, but also sometimes also the limits of the data. The gap observed repeatedly between a phase of climatic deterioration and the response of fluvial environments can result from the lapse of time between ecosystem changes and the time required for fluvial dynamics to increase in magnitude. Nevertheless, in an environment, which promotes a rapid fixation of the soil by the herbaceous cover, the continuation of the silty sedimentation during the different phases implies continuous farming of the soils and the practice of ploughing. This is also true for the increase of the anthropogenic markers in the palynological diagrams in this region. Its seems then plausible that the periods poorly represented by archaeological data, such as the end of the Neolithic or the early and middle Bronze Ages, might have seen greater agricultural activities than the low number of sites seem to show. In a general manner, if the anthropogenic deterioration of the vegetation cover appears as the main cause of the detrital inputs, the climate seems to play a significant role during phases of high hydro-sedimentary activities, and particularly during the second part of the Subboreal or during the Little Ice Age.

Bibliographie

References

Alley R.B., Mayevski P.A., Sowers T., Stuiver M., Taylor K.C. et Clarck P.U., «Holocene climatic instability: a prominent widespread event 8200 yr ago», Geology, no 25, 1997, p. 483-486.

Blin O., Gebhardt A., Allenet G., Bernard V., Boyer F., Dietrich A., Leroyer C., Marguerie D., Matterne V., Seyriessol K., et Zwierzinski E., «Impact anthropique et gestion du milieu durant l’Antiquité: l’approche paléoenvironnementale pluridisciplinaire du site «la Ferme d’Ithe» à Jouars-Ponchartrain (Yvelines)», Les Nouvelles de l’Archéologie, no 78, 1999, p. 45-56.

Bohncke S.J.P. et Hoek W.Z., «Multiple oscillations during the Preboreal as recorded in a calcareous gyttja, Kingbeekdal, The Netherlands», Quaternary Science Reviews, no 26, 2007, p. 1965-1974.

Bohncke S., Vandenberghe J., Coope R., Reilling R., «Geomorphology and palaeoecology of the Mark valley (southern Netherlands); Palaeoecology, palaeohydrology and climate during the Weichselian Late Glacial», Boreas, no 16, 1987, p. 69-85.

Bortenschlager S., «Ursachen und Ausmass postglazialer Waldgrenzschwankungen in den Ostaalpen» in Frenzel B. (éd.), Dendrochronologie und postglaziale Klimaschwankungen in Europa, Steiner Verlag, Wiesbaden, 1977, p. 260-266.

Bourgeois J., Talon M. (éd.), « L’Âge du Bronze du Nord de la France dans son contexte européen », Actes des congrès nationaux des sociétés savantes, 125e, Lille 2000, Paris, CTHS, 2005, 378 p.

Bravard J.-P., Lebot-Helly A., Savay-Guerraz H., « Le site de Vienne (38), Saint-Romain (69), Sainte-Colombe (69). L’évolution de la plaine alluviale du Rhône de l’âge du Fer à la fin de l’Antiquité: proposition d’interprétation », in Archéologie et Espace : Xes Rencontres Internationales d’Archéologie et d’Histoire d’Antibes, octobre 1989, Juan-les-Pins, APDCA, 1990, p. 437-452.

Brown A.G., «Alluvial geoarcheology: Floodplain archaeology and environmental change», Cambridge University Press, Cambridge, 1997.

Brunet P., Cottiaux R., Hamon T., Magne P., Richard G., Salanova L, Samzun A., « Les ensembles céramiques de la fin du IIIe millénaire (2300-1900 avant notre ère) dans le Centre-Nord de la France », Bulletin de la Société préhistorique française, t. 105, no 3, 2008, p. 595-615.

Burga C., «Vegetationgeschichte seit der Späteiszeit», Geographica Helvetica, no 2, 1987, p. 71-80.

Chaussé C., Leroyer C., Girardclos O., Allenet G., Pion P., Raymond P., «Holocene history of the River Seine, Paris, France: bio-chronostratigraphic and geomorphological evidence from the Quai-Branly», The Holocene, no 18, 2008, p. 967-980.

Coutard S., Ducrocq T., Limondin-Lozouet N., Bridault A., Leroyer C., Allenet G., Pastre J.-F., « Contexte géomorphologique, chronostratigraphique et paléoenvironmental des sites mésolithiques et paléolithiques de Warluis dans la vallée du Thérain (Oise, France) », Quaternaire, no 21, 2010, p. 357-384.

Gauthier A., « Résultats palynologiques de séquences holocènes du Bassin parisien: histoire de la végétation et action de l’Homme », Palynosciences, 3, 1995, p. 3-17.

Granai S., Limondin-Lozouet N., Chaussé C., « Évolution paléoenvironnementale de la vallée de la Seine à Paris (France) d’après l’étude des malacofaune », Quaternaire, no 22, 2011, p. 327-344.

Lefèvre D., Heim J., Gilot J., Mouthon J. (1993), « Évolution des environnements sédimentaires et biologiques à l’Holocène dans la plaine alluviale de la Meuse », Quaternaire, no 4, p. 17-30.

Leroyer C., Homme, climat, végétation au Tardi-et Postglaciaire dans le Bassin Parisien: apports de l’étude palynologique des fonds de vallée, Thèse de l’Université de Paris I, 1997, 2 vol.

Leroyer C., «Évolution de la végétation et emprise de l’Homme sur le milieu à Bercy (Paris, France)» in Le Néolithique du Centre-Ouest de la France, actes du XXIe colloque interrégional sur le Néolithique, Poitiers, 1998, p. 407.

Leroyer C., «Environnement végétal des structures funéraires et anthropisation du milieu durant le Néolithique récent/final dans le Bassin parisien», in Sans dessus dessous - La recherche du sens en préhistoire, Recueil d’études offert à J. Leclerc et C. Masset, Revue Archéologique de Picardie, n.s., no 21, 2003, p. 83-92.

Leroyer C., «L’anthropisation du Bassin parisien du VIIe au IVe millénaire d’après les analyses polliniques de fonds de vallées: mise en évidence d’activités agro-pastorales très précoces», in Richard H. (éd.), Néolithisation précoce. Premières traces d’anthropisation du couvert végétal à partir des données polliniques, Collection Annales littéraires, série Environnement, Sociétés et Archéologie, no 7, Presses Universitaires de Franche-Comté, Besançon, 2004, p. 11-27.

Leroyer C., «L’impact des groupes néolithiques du Bassin parisien sur le milieu végétal: évolution et approche territoriale d’après les données polliniques», in Duhamel P. (éd.), Impacts interculturels au Néolithique moyen. Du terroir au territoire: sociétés et espaces, Actes du 25ème colloque interrégional sur le Néolithique, Dijon 2001. Revue archéologique de l’Est., 25e supplément, 2006a, p. 133-150.

Leroyer C., «L’environnement végétal des sites: les données de la palynologie», in Malrain F., Pinard E. (éd.), Les sites laténiens de la moyenne vallée de l’Oise du ve au ier s. avant notre ère, Contribution à l’Histoire de la société gauloise, Revue archéologique de Picardie, no spécial 23, 2006b, p. 34-42.

Leroyer C., Allenet G., «L’anthropisation du paysage végétal d’après les données polliniques: l’exemple des fonds de vallées du Bassin parisien», in Allée P., Lespez L. (éd.), L’érosion entre société, climat et paléoenvironnement, Presses Universitaires Blaise Pascal, collection Nature et Société, no 3, 2006, p. 63-72.

Leroyer C., Coubray S., Allenet G., Perrière J., Pernaud J.-M., «Vegetation dynamics, human impact and exploitation patterns in the Paris basin through the Holocene: palynology vs. anthracology», Saguntun, extra no 11, 2011, p. 82.

Limondin N., «Late-Glacial and Holocene Malacofaunas from Archaeological Sites in the Somme Valley (North France)», Journal of Archaeological Science, no 22, 1995, p. 683-698.

Limondin-Lozouet N., « Successions malacologiques à la charnière Glaciaire/Interglaciaire: du modèle Tardiglaciaire-Holocène aux transitions du Pléistocène », Quaternaire, no 22, 2011, p. 211-220.

Limondin N., Rousseau D.-D., «New Holocene malacological sequence at Verrières, Seine valley, France», Boreas, no 20, 1991, p. 207-209.

Limondin-Lozouet N., Antoine P., «Palaeoenvironmental changes inferred from malacofaunas in the Lateglacial and Early Holocene fluvial sequence at Conty (Northern France)», Boreas, no 30, 2001, p. 148-164.

Limondin-Lozouet N., Preece R.C., «Molluscan successions from the Holocene tufa of St-Germain-le-Vasson in Normandy, France», Journal of Quaternary Science, no 19, 2004, p. 55-71.

Macklin M. G., «Holocene river environments in Prehistoric Britain: Human interaction and impact», Quaternary Proceedings, no 7, 1999, p. 521-530.

Magny M., «Holocene lake-level fluctuations in Jura and the northern subalpine ranges, France: regional pattern and climatic implications», Boreas, vol. 21, 1992, p. 319-334.

Magny M., «Paleohydrological changes in Jura (France) and climatic oscillations around the north Atlantic from Allerød to Preboreal», Géographie physique et Quaternaire, no 49, 1995, p. 401-408.

Magny M., «Holocene climate variability as reflected by mid-European lake-level fluctuations an its probable impact on prehistoric human settlements», Quaternary International, no 113, 2004, p. 65-79.

Magny M., Richard H., «Essai de synthèse vers une courbe de l’évolution du climat entre 500 BC et 500 AD», Les nouvelles de l’Archéologie, no 50, 1992, p. 58-60.

Magny M., Vannière B., de Beaulieu J.-L., Bégeot C., Heiri O., Milleta L., Peyron O., Walter-Simonnet A.-V., «Early Holocene climatic oscillations recorded by lake-level fluctuations in westcentral Europe and in Central Italy», Quaternary Science Reviews, no 26, 2007, p. 1951-1964.

Malrain F., «Fonctionnement et hiérarchie des fermes dans la société gauloise du iiie siècle à la Période romaine: l’apport des sites de la moyenne vallée de l’Oise», Thèse de doctorat, Université Panthéon-Sorbonne (Paris 1), 2000.

Marinval-Vigne M.-C, Mordant D., Auboire G., Augereau A., Bailon S., Dauphin C., Delibrias G., Krier V., Leclerc A.S., Leroyer C., Marinval P., Mordant C., Rodriguez P., Vilette P., Vigne J.-D., «Noyen-sur-Seine, site stratifié en milieu fluviatile: une étude multi-disciplinaire intégrée», Bulletin de la Société préhistorique française, no 10-12, 1989, p. 370-379.

Orth P., « Évolution et variabilité morphosédimentaire d’un bassin versant élémentaire au Tardi-et au Postglaciaire: l’exemple du bassin versant de la Beuvronne (Bassin parisien) », Thèse de doctorat, Université Panthéon-Sorbonne (Paris 1), 2003.

Orth P., Pastre J.-F., Gauthier A., Limondin-Lozouet N., Kunesch S., « Les enregistrements morphosédimentaires et biostratigraphiques des fonds de vallées du bassin versant de la Beuvronne (Bassin parisien, Seine-et-Marne, France) : perception des changements climato-anthropiques à l’Holocène », Quaternaire, no 15, 2004, p. 285-298.

Pastre J.-F., Fontugne M., Kuzucuoglu C., Leroyer C., Limondin-Lozouet N., Talon M., Tisnerat N., « L’évolution tardi-et postglaciaire des lits fluviaux au nord-est de Paris (France). Relations avec les données paléoenvironnementales et l’impact anthropique sur les versants », Géomorphologie: relief, processus, environnement, no 4, 1997, p. 291-312.

Pastre J.-F., Leroyer C., Limondin-Lozouet N., Fontugne M., Hatté C., Krier V., Kunesch S., Saad M.C., « L’Holocène du Bassin parisien: variations environnementales et réponses géoécologiques des fonds de vallées », Annales littéraires de l’Université de Franche-Comté, Actes du colloque « Equilibres et ruptures dans les écosystèmes depuis 20000 ans en Europe de l’Ouest », série Environnement, Sociétés et Archéologie, no 3, 2002a, p. 61-73.

Pastre J.-F., Leroyer C., Limondin-Lozouet N., Orth P., Chaussé C., Fontugne M., Gauthier A., Kunesch S., Le Jeune Y., Saad M.C., «Variations paléoenvironnementales et paléohydrologiques durant les 15 derniers millénaires: les réponses morphosédimentaires des vallées du Bassin Parisien (France)», in Bravard J.-P., Magny M. (éd.). Variations paléohydrologiques en France depuis 15 000 ans, Paris, Errance, 2002b, p. 29-44.

Pastre J.-F., Orth P., Le Jeune Y., Bensaadoune S., «L’Homme et l’érosion dans le Bassin parisien (France). La réponse morphosédimentaire des fonds de vallées au cours de la seconde partie de l’Holocène», in Allée P., Lespez L. (éd.), L’érosion entre société, climat et paléoenvironnement, Presses Universitaires Blaise Pascal, Clermont-Ferrand, 2006, p. 237-247.

Peake R., Allenet G., Bernard V., Chaussé C., Clavel B., Dietrich A., Leroyer C., Seguier J.-M., « Un exemple de gestion du milieu humide en fond alluvial à l’âge du Fer à Bazoches-lès-Bray (Seine-et-Marne) », L’âge du Fer en Ile-de-France, Actes du XXVIe colloque de l’AFEAF, 26e supplément à la Revue archéologique du Centre de la France, 2005, p. 157-179.

Rousseau D.-D., Limondin N., Puissegur J.-J., « Réponses des assemblages malacologiques holocènes aux impacts climatiques et anthropiques sur l’environnement », Comptes-rendus de l’Académie des Sciences, no 315, II, 1992, p. 1811-1818.

Rousseau D.-D., Limondin N., Puissegur J.-J., «Holocene environmental signals from mollusk assemblages in Burgundy (France)», Quaternary Research, no 40, 1993, p. 237-253.

Rousseau D.-D., Limondin N., Magnin F., Puissegur J.-J., «Temperature oscillations over the last 10 000 years in Western Europe estimated from molluscan assemblages», Boreas, no 23, 1994, p. 66-73.

Salanova L., Brunet P., Cottiau R., Hamon T., Langry-François F., Martineau R., Polloni A., Renard C., Sohn M., « Du Néolithique recent à l’âge du Bronze dans le Centre Nord de la France: les étapes de l’évolution chrono-culturelle », R.A.P. nospecial 28, Le Néolithique du Nord de la France dans son contexte européen, 2011, p. 77-101.

Starkel L., «Palaeohydrology of the Temperate Zone», in Gregory K.J., Starkel L., Baker V. R. (éd.), Global Continental Palaeohydrology, Wiley, Chischester, 1995, p. 234-257.

Starkel L., Kalicki T., Krapiec M., Soja R., Gebica P., Czyzowska E., «Hydrological changes of valley floors in upper Vistula basin during late Vistulian and Holocene», in Starkel L. (éd.), Évolution of the Vistula river valley during the last 15 000 years, Geographical Studies, Special Issue, 9, Wroclaw, 1996, p. 7-128.

Van Geel B., Buurmann J., Waterbolk H.T., «Archaeological and palaeoecological indications of an abrupt climate change in The Netherlands, and evidence for climatological teleconnections around 2650 BP», Journal of Quaternary Science, no 11, 1996, p. 451-460.

Table des illustrations

Légende Fig. 1. General map of the studied area and sequences.
URL http://books.openedition.org/editionscnrs/docannexe/image/22008/img-1.jpg
Fichier image/jpeg, 240k
Légende Fig. 2. Main phases of cut-and-fill and sedimentation in the large valleys of the Paris basin (after Pastre et al., 2002a).
URL http://books.openedition.org/editionscnrs/docannexe/image/22008/img-2.jpg
Fichier image/jpeg, 224k
Légende Fig. 3. Evolution of the sedimentation of the main palaeochannel of the Oise River in Armancourt (Oise). 1: loams and stones (embankment); 2: mixed loams (by coring); 3: bedded organic clay with ligneous fragments; 4: beige silty clay; 5: bedded dark grey organic and minerogenic clay; 6: grey-black organic clay; 7: grey silty clay; 8: dark grey organic and minerogenic clay; 9: gravels and small peebles (silex); 10: dark grey silty clay; 11: bedded organic and minerogenic clay with ligneous fragments; 12: fine grey sand; 13: organic and minerogenic silt; 14: bedded organic clay; 15: sand and gravels.
URL http://books.openedition.org/editionscnrs/docannexe/image/22008/img-3.jpg
Fichier image/jpeg, 207k
Légende Fig. 4. Evolution of sedimentation in the Biberonne valley (Beuvronne catchment) in Compans (after Orth, 2003, modified). 1: present soil; 2: palaeosoil; 3: silt; 4: organic and minerogenic silt; 5: humified black peat; 6: russet coloured peat; 7: brown peat; 8: organic and minerogenic peat; 9: tufa; 10: interbedded silt, peat and tufa; 11: peaty silt; 12: sandy silt; 13: silty sand; 14: sand; 15: clay, sand and gravels; 16: sand of Beauchamp; 17: marly limestone of Saint-Ouen; 18: calibrated radiocarbon date BC; 19: stratigraphic unit.
URL http://books.openedition.org/editionscnrs/docannexe/image/22008/img-4.jpg
Fichier image/jpeg, 249k
Légende Fig. 5. Pollen diagrams of anthropogenic influence (anthropisation) in several Neolithic sequences from the Paris Basin. The criteria for anthropogenic impact or “anthropisation” is indicated by the percentage of ruderals, Plantago lanceolata and Cerealia present in each palynological spectra. 1: local occupation; 2: occupation in the surroundings; 3: calibrated radiocarbon date in BC; 4: cereals; 5: Plantago lanceolata, 6: other ruderals.
URL http://books.openedition.org/editionscnrs/docannexe/image/22008/img-5.jpg
Fichier image/jpeg, 452k
Légende Fig. 6. Biozonation of the Holocene tufa of Saint-Germain-le-Vasson (Calvados) after the malacological faunas (modified after Limondin-Lozouet and Preece, 2004). (a) occurrence of the species in the stratigraphical unit associated with the main and lateral section and the core 2; (b) occurrence of the species in the core 1 that comprises the whole part of the tufa; (c) correlation after the malacofaunas. The species Discus rotundatus, for which the increase corresponds to that of the hazel, characterises the occurrence of the broad-leaved tree forests in the Holocene (Limondin-Lozouet et al., 2005) and offers the basis of biostratigraphical correlation. The appropriateness of this choice is corroborated by the coherence in the spectra of the other key species. The sequence of Saint-Germain-le-Vasson gives the malacological succession of reference for the first half of the Holocene. The six malacological zones are characterised by key taxa remarkable for their constrained ecological demands and/or their limited geographical distribution. The malacofaunas indicate the development of swampy open-lands at the beginning of the Holocene (SGV 1), then the development of a forested environnment since the zona SGV 2, be marked by a decrease of wetness all along the Atlantic (SGV 3 to 5), and which finally becomes clearer at the end of the climatic optimum. The dates give a precise chronological framework to this succession and permit precise comparisons with other indicators (Limondin-Lozouet and Preece, 2004). The dates in bold are in years BP and the dates in normal letters are in years BC.
URL http://books.openedition.org/editionscnrs/docannexe/image/22008/img-6.jpg
Fichier image/jpeg, 395k
Légende Fig. 7. Sedimentary sequence of the Serre valley in Courbes “Les Patures” (Aisne) – (core C1).
URL http://books.openedition.org/editionscnrs/docannexe/image/22008/img-7.jpg
Fichier image/jpeg, 225k

Auteurs

Research Director, 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 (jean-francois.pastre@cnrs-bellevue.fr).

Research Director, 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 (Nicole.Limondin@cnrs-bellevue.fr).

Research Director, 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 (Pierre.Antoine@cnrs-bellevue.fr).

Engineer, 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 (agnes.gauthier@cnrs-bellevue.fr).

Ph. D Student, Mixed Research Unit (UMR 8591) CNRS/Universities of Paris 1 & Paris 12/INRAP (Laboratory of Physical Geography: Present and Quaternary Environments–LGP), Meudon, France, and Mixed Research Unit (UMR 8215) (Trajectories–From Sedentarisation to the State), Nanterre, France (salomegranai@yahoo.fr).

Engineer, Regional Archaeology Service, Regional Directorate of Cultural Affairs (Pays de la Loire), Mixed Research Unit (UMR 8591) CNRS/Universities of Paris 1 & Paris 12/INRAP (Laboratory of Physical Geography: Present and Quaternary Environments–LGP), Meudon, France (lj.yann@gmailcom).

Research Engineer, National Institute for Preventive Archaeological Research (INRAP Centre & Île-de-France), Mixed Research Unit (UMR 8591) CNRS/Universities of Paris 1 & Paris 12/INRAP (Laboratory of Physical Geography: Present and Quaternary Environments – LGP), Meudon, France (Patrice.Wuscher@inrap.fr).

© CNRS Éditions, 2015

Conditions d’utilisation : http://www.openedition.org/6540

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search