Fluvial system evolution and environmental changes during the
Holocene in the Mue valley (Western France)
Laurent Lespez a,, Martine Clet-Pellerin b
, Nicole Limondin-Lozouet c
,
Jean-François Pastre c
, Michel Fontugne d
, Cyril Marcigny e
a
Université de Caen-Basse-Normandie, GEOPHEN-LETG UMR CNRS 6554, Esplanade de la Paix, BP 5186, 14032 Caen cedex, France
b
M2C-UMR CNRS 6143, Esplanade de la Paix, BP 5186, 14032 Caen cedex, France
c
LGP UMR CNRS 8591, 1 place A. Briand, 92195 Meudon cedex, France
d
LSCE-UMR CEA-CNRS 1572, avenue de la Terrasse, 91198 Gif/Yvette cedex, France
e
INRAP Basse-Normandie, Le Chaos, 14400 Longues-sur-Mer, France
Received 11 January 2006; received in revised form 29 May 2006; accepted 20 February 2007
Available online 13 May 2007
Abstract
Geomorphological and palaeoenvironmental research on Holocene sedimentation in the Mue valley provides evidence for fluvial
system changes related to climate and human activities in Normandy, a poorly studied area of the Paris basin. The 24-km long valley
bottom has been investigated through a systematic survey. It shows an original longitudinal sedimentary pattern in relation with valley
morphology and local geological controls. Minerogenic, tufaceous and peaty deposits provide opportunities for multi-proxy analyses
and radiocarbon dating control. Sedimentation began around 9500 14
C BP with silt deposition in a meandering system. The Boreal
and the Lower Atlantic periods (8500­6000 14
C BP) were mainly characterized by unlithified calcareous tufa. Locally, these deposits
are very thick (7 to 13 m). The tufa formed barrages across the valley bottom, providing an autogenic control on upstream
sedimentation. During the Upper Atlantic period (6000­4700 14
C BP), the valley experienced a decrease in calcareous sedimentation
and the development of organic deposits. At the beginning of the Subboreal (4700­3500 14
C BP), peat deposits expanded, especially
behind the tufa barrages. The valley bottom was characterized by large marshy areas whereas the regional vegetation was
progressively modified by human activities. At the end of the Subboreal (3300­3000 14
C BP) the infilling of the valley by calcareous
silt was caused by an increase of river activity related to climatic and land use changes. From the Iron Age and Gallo-Roman periods
(2800­1700 14
C BP), the valley bottom was filled by silty overbank deposits related to an increase of soil erosion. The slopes and
river system were once again coupled and the fluvial system functioned as a continuum from upstream to downstream. The alluvial
record of the Mue valley reflects a broad regional pattern of environmental changes but presents particular features, which highlight
the need of longitudinal studies to take into account spatial and temporal discontinuities of Holocene hydro-sedimentary systems, even
in small order valleys.
 2007 Elsevier B.V. All rights reserved.
Keywords: Fluvial System; Alluvial records; Longitudinal arrangement; Tufa; Holocene; Normandy
Available online at www.sciencedirect.com
Geomorphology 98 (2008) 55­70
www.elsevier.com/locate/geomorph
 Corresponding author. Tel.: +33 231566427; fax: +33 231566386.
E-mail addresses: laurent.lespez@unicaen.fr (L. Lespez), martine.clet-pellerin@unicaen.fr (M. Clet-Pellerin),
Nicole.Limondin@cnrs-bellevue.fr (N. Limondin-Lozouet), Jean-François-Pastre@cnrs-bellevue.fr (J.-F. Pastre), Michel.Fontugne@lsce.cnrs-gif.fr
(M. Fontugne), cyril.marcigny@inrap.fr (C. Marcigny).
0169-555X/$ - see front matter  2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.geomorph.2007.02.029
1. Introduction
Research on the fluvial responses to climatic and
anthropogenic changes during the Holocene have been
numerous in Europe and give evidence of the role of the
different controls (e.g. Starkel, 1983; Gregory et al., 1995;
Brown, 1997; Benito et al., 1998; Maddy et al., 2001).
Nevertheless, they also underline the complexity of
Holocene fluvial archives and particularly their fragmentary
pattern, which renders precise chronostratigraphic
studies difficult, and the distinction between autogenic
controls and external environmental influences (Brown,
1997; Lewin et al., 2005). To take into account the lateral
variability of the valley-bottom alluvial filling, most
investigations are based on extended cross-sections using
backhoe trenches or closely spaced augers while, in
lowland areas, the longitudinal complexity is more often
neglected even if it appears important to understand the
fluvial system as a whole (Brown, 1990; Macklin, 1999).
Indeed, the discontinuity of sediment routes through a
catchment and along a river is the norm (Brown, 1990;
Harvey, 2002; Hooke, 2003; Kasai et al., 2006). In
addition, in sedimentary basins, specific alluvial features
like transverse tufa barrages can interrupt the continuity of
sediment transfer (Pedley et al., 2000; Pentecost, 2005).
These observations underline the necessity of systematic
Fig. 1. Location of bore holes and core drillings in the Mue river basin.
56 L. Lespez et al. / Geomorphology 98 (2008) 55­70
studies from upstream to downstream to understand the
heterogeneity of valley bottom deposits, particularly in
small-scale river basins in which it appears most feasible
to conduct such analyses (Brown, 1990).
The Mue valley, located in the Plain of Caen (Normandy),
provides an opportunity to analyse the complexity
of a small fluvial system in a poorly studied area
of northern France (Fig. 1). Various studies have documented
the alluvial sedimentary sequences of the Paris
Basin during the Lateglacial and the Holocene, determining
the main changes experienced by the fluvial
system in the northern (Antoine et al., 2003) and central
parts (Pastre et al., 2001, 2003, 2006) of the basin.
Therefore, although the Seine valley has recently been
studied (Sebag, 2002; Lesueur et al., 2003), the western
part of the basin is poorly known, making comparison
with changes of western European rivers difficult. Based
on a systematic hand and power auger survey, the aim of
this geomorphological and palaeoecological research is
to determine the longitudinal pattern of the alluvial
filling in the Mue valley and define the role of climate,
human activities and autogenic controls on Holocene
fluvial dynamics.
2. Study area
The Mue river is the last right-side tributary of the
Seulles, a coastal river of Normandy. The river basin is
small (97 km2
) and the valley is entrenched from 15 m to
40 m in Mesozoic limestone (Bathonian) which overlays
the marls of Port-en-Bessin (Fig. 1). Being 24 km long,
the longitudinal profile of the valley bottom is complex. It
is characterized by five reaches with different morphologies
(Fig. 2). The steepest reaches (3.5­7.4) are more
often narrow (50­75 m) while the wider reaches (120­
500 m) have a weaker slope (1­2). This longitudinal
pattern is explained by the intraplate tectonic activity on
the southern side of the English Channel (Lagarde et al.,
2000). Locally, uplifted blocks are the results of dislocation
of the late Cenozoic erosional surface by differential
tectonic activity (Font, 2002). In addition to this
pattern, we observe the role of the N50° faults which
define the orientation of the valley (Fig. 1).
Today, the land use of the river basin is characterized
by intensive agriculture on loamy productive soils of the
Plain of Caen, with meadows or wet fallow land in the
valley bottom and fallow land to forested area on slopes.
The river has an average discharge of 0.35 m3
/s and has
experienced a monthly maximum of 1.5 m3
/s and a
monthly minimum of 0.1 m3
/s during the period 1970­
1998. Groundwater of the Bathonian aquifer is the main
water supply and contributes to sustaining flow in
summer even if pumping for the urban area of Caen has
contributed to low water levels in summer (Cador et al.,
2001). Since the Middle-Ages, the flow has been almost
totally controlled. Indeed, 60% of the last 14 km of the
stream correspond to mill or tail races which supplied
more than 20 water-mills in the 18th century (Lespez
et al., 2005a).
In the Plain of Caen and more generally in Normandy,
geomorphological research on Holocene fluvial
systems remains scarce (Elha, 1963; Lespez et al.,
2004, 2005a,b; Germain-Vallée and Lespez, 2006).
Previous research on the Mue valley has provided
evidence of the importance of Holocene alluvial filling
and the local role of a tufa barrage downstream (CletPellerin
et al., 1990). Nonetheless, the Holocene
chronostratigraphy was never established and the question
of climatic and human impacts on the fluvial system
during the Holocene remained open in an area where
archaeological data are numerous and indicate dense
settlement since the Middle Neolithic (c. 5500 BP)
(Desloges, 1997; San Juan et al., 1999; Ghesquire and
Marcigny, in press).
3. Methods
A systematic geomorphological survey was carried
out in the valley bottom to reconstruct the Holocene
alluvial chronostratigraphy (Fig. 1). Partial or complete
cross-sections through the valley, and auger holes and
cores regularly placed from upstream to downstream,
were investigated as the main objective was to understand
the longitudinal pattern of the filling. First, we
obtained thirty-six hand augers and cores (3­7 m deep)
along the valley and twenty-six bore holes (1­4 m deep)
in the Thaon­Fontaine­Henry area. Then five cores
were drilled (3­12 m deep) in key areas. Each core was
described in the field according to texture, grain size,
colour, vegetal remains and macrofossil content. In the
laboratory, sedimentological analyses were focused on
the cores but also concerned the rest of the Holocene
sedimentation to characterize the general pattern of the
alluvial dynamics along the valley bottom. Seven samples
come from the current sediments and sixty-seven
from sampling in the former deposits. Grain size analyses
were made using a Laser Granulometer (BeckmanCoulter,
LS 200). The sand and gravel fraction was
sieved and examined under a binocular microscope.
Suitable cores were used for palaeoecological analyses.
The organic sediments of the Cairon, FontaineHenry,
Reviers and Rots cores were selected for pollen
analyses (120 samples each 3 to 10 cm). Indeed, despite
the problem of preservation and the waterborne
57L. Lespez et al. / Geomorphology 98 (2008) 55­70
Fig. 2. Holocene sedimentation in the Mue valley bottom indicating the longitudinal arrangement of deposition.
58L.Lespezetal./Geomorphology98(2008)55­70
contamination, recent studies highlight the potential
value of pollen data obtained in floodplain fens to
reconstruct the environment of the valley bottom and
surrounding areas (Leroyer, 1997; Brown, 1999).
Continuous sampling at 4­15 cm intervals was adopted
for malacological study of the Thaon core (30 samples)
following methods developed at Saint­Germain­LeVasson,
which is located some 40 km southward
(Limondin-Lozouet and Preece, 2004).
The chronology is based on archaeological remains
and radiocarbon dating. Twenty-one organic samples
were analysed for radiocarbon dating (shells, plant remains,
charcoal, total organic mater). Twelve have been
obtained by conventional methods at LSCE (Gif/Yvette,
France) and nine by the Accelerated Mass Spectrometry
(AMS) method at the Physikalisches Institut of Erlangen
(Germany). Use of organic sediment from fluvial
deposits and shell carbonate for dating may introduce
complications because of the potential of reworking,
contamination and the hard water effect. To improve the
chronological framework, we used Cepaea shells which
incorporate minimal quantities of ancient carbon from
bedrock sources and which are generally considered
ideal for dating purposes (Limondin-Lozouet and
Preece, 2004). Sediment dating was validated by correlation
with local and regional pollen records (CletPellerin
and Verron, 2004; Lespez et al., 2005b). Data are
quoted in 14
C BP in the text, and calibrated age ranges
according to Calib. Rev. 4.3. (Stuiver and Reimer, 1993)
are given in Table 1.
4. Results
The Holocene sediments are 4 to 12 m thick from
Rots (upstream) to Reviers (downstream). They are
characterized by low lateral variability, which is observed
in most of the cross-sections whereas heterogeneity
of the longitudinal arrangement and the
sedimentary facies is high. A classification based on
mean grain size and the sorting index of Folk and
Ward (1957) provide evidence of seven facies (Fig. 3
and Table 2). The mineral deposits are not well developed
in the valley bottom and the sediment is mainly
characterized by carbonate or organic deposits. The
carbonate deposits are related to the high content of
dissolved load of the river water caused by spring
activity and the Ca-rich ground water supply, while the
organic deposits are explained by the extension of
floodplain fens related to a high water level in the
valley bottom.
4.1. Longitudinal pattern of the alluvial filling at the
valley scale
Along the Mue valley and its tributaries, we observed
seven distinct morpho-sedimentary reaches
(Figs. 2 and 4). The chronozones used in the presentation
of the results correspond to those defined by
Mangerud et al. (1974), widely used in the Paris basin
(Leroyer, 1997; Pastre et al., 2001; Clet-Pellerin and
Verron, 2004).
Table 1
Information on the Mue Valley samples submitted for radiocarbon dating
Location Lab.
code
Core
drilling
Depth
(m)
Materials Methods Dates 14
C BP
(1 )
 13
C
%O
Calibrated dates BP
(2 )
Thaon-Colonie de vacances Gif-11716 S3 3.17­3.28 Peaty silt Conventional 453560 -28.26 3496­3026 BC
Thaon-Colonie de vacances Gif-11714 S3 1.90­1.98 Peaty silt Conventional 330040 -28.05 1686­1462 BC
Thaon-Old Church Erl-6604 Thaon 7.20­7.30 Shell Cepea AMS 921468 -12.4 8607­8287 BC
Thaon-Old Church Erl-6605 Thaon 1.79­1.89 Shell Cepea AMS 522069 -12.5 4243­3806 BC
Fontaine­Henry Gif-11175 FH11 2.47­2.65 Organic silt Conventional 774080 -23.1 6976­6423 BC
Fontaine­Henry Gif-11711 FH14 2.40­2.50 Organic silt Conventional 324540 -27.78 1675­1429 BC
Fontaine­Henry Gif-11713 FH25 2.90­3.00 Organic silt Conventional 503065 -25.91 3962­3669 BC
Fontaine­Henry Gif-11864 FH 7.71­7.75 Peat Conventional 447540 -28.68 3351­2942 BC
Fontaine­Henry Gif-11868 FH 5.05­5.10 Peat Conventional 315045 -29.65 1518­1316 BC
Fontaine­Henry Gif-11867 FH 3.96­4.01 Peat Conventional 242040 -28.6 762­398 BC
Fontaine­Henry Gif-11865 FH 1.45­1.52 Peat Conventional 150045 -28.3 436­643 AD
Reviers Erl-6608 Reviers 11.24­11.30 Organic silt AMS 771566 -28.1 6651­6438 BC
Reviers Erl-6607 Reviers 8.45­8.53 Organic silt AMS 598950 -28.9 4996­4726 BC
Reviers Erl-6606 Reviers 6.75­6.80 Charcoal AMS 531545 -29.7 4317­3999 BC
Cairon Erl-6790 Cairon 1.98­2.02 Organic silt AMS 810258 -28.3 7321­6829 BC
Cairon Erl-6789 Cairon 1.78­1.81 Peat AMS 669549 -29.2 5713­5487 BC
Cairon Erl-6788 Cairon 1.56­1.59 Peat AMS 624249 -29.0 5316­5059 BC
Cairon Erl-6787 Cairon 0.78­0.82 Peat AMS 510847 -29.4 4033­3791 BC
Calibrated age according to CALIB rev. 4.3. (Stuiver and Reimer, 1993).
59L. Lespez et al. / Geomorphology 98 (2008) 55­70
4.1.1. Reach 1
Along the upper Mue and the Chironne, the Holocene
deposits are less than 2 m thick and correspond mainly
to colluvial silts. Along the Vey, the deposits are also 2
to 2.5 m thick but they are more organic sediments.
In the core drillings at Cairon, we have identified four
stratigraphic units with radiocarbon dating control. The
bottom unit corresponds to grey-green sandy silts which
indicate a fluvial dynamic. Between the end of the Boreal
(810258 14
C BP) and the beginning of the Atlantic
periods (post 669549 14
C BP), it is covered by peat
layers, organic silts and organic silts with oncoids. During
the end of the Atlantic (624249 14
C BP) and the
beginning of the Subboreal (510847 14
C BP), the valley
bottom is characterized by peatdeposits in a floodplain fen.
After, colluvial silts have totally covered the valley bottom.
4.1.2. Reach 2
Along this reach, sedimentation is mainly developed in
the valley axis where it reaches 4 m. In the absence of
radiocarbon dating, the dating control is not accurate but a
general chronological pattern is provided by the palynostratigraphy
of two core drillings and particularly the appearance
of walnut and the recent development of pine
(Rots 1&3, Lespez et al., 2005b). At the bottom, a coarse
alluvial deposit (sand and gravel, U1) has been covered by
grey silts (U2). Then, the reach is alternately characterized
by organic and tufaceous sediment. Peat and organic silts
have been deposited during the Atlantic (U3a) and the
Subatlantic periods (U5). They are often interrupted by a
calcareous silt layer (U4). Tufaceous sedimentation is
mainly comprised of poorly sorted, multimodel silts, rich
in oncoids (U3b). Finally, the valley bottom was covered
by colluvial and overbank silts (U6).
4.1.3. Reach 3a
Upstream, this reach begins with a section where
Holocene sedimentation is very low (S11) but, downstream,
it becomes more complex and deeper up to 5 m.
From Cairon to Thaon, we observe a succession of six
sedimentary units in the filling. At the base, on S24 a sandy
silt layer (U1) is covered by grey silts (U2). But the bulk of
the sediment corresponds to whitish calcareous silts with
numerous oncoids and fine layers of organic silt or peat
(U3). A date obtained (453560 BP 14
C BP) in the middle
of this unit (S3) indicates the Suboreal period for its
deposition. It is often bounded by an organic layer (U4)
dated atthe end of theSubboreal onS3 (330040 14
C BP).
Overlaying this we often observed a whitish calcareous silt
layer (U5) before sedimentation returned to more organic
deposits (U6). The Holocene stratigraphy is closed by a
silty sediment which fills the valley bottom (U7).
4.1.4. Reach 3b
From the confluence with the Chironne to the Old
Church of Thaon, the sedimentation is mainly organic.
Fig. 3. Sedimentary facies in relation with grain size analyses.
Table 2
The different sedimentary facies and alluvial environments in the Mue
Valley
Facies Alluvial environment
Group 1 Tufaceous sands
(oncoids)
Unattached travertine Spring,
channel margin
Group 2 Alluvial sands Channel
Group 3 Overbank silts Floodplain
Group 4 Fine silts Marsh, waterholes
Group 5 Fine and organic silts Marsh
Group 6 Heterometric tufaceous
sediment
Channel margin,
backswamps
Group 7 Peat Marsh
60 L. Lespez et al. / Geomorphology 98 (2008) 55­70
Fig. 4. Simplified cross-sections in the Mue valley bottom from upstream (reach 2) to downstream (reach 4).
61L. Lespez et al. / Geomorphology 98 (2008) 55­70
In detail, we observe 5 sedimentary units. The limestone is
covered, locally (S4), by a sandy silt layer (U1) and by
grey silts (U2). However, the main Holocene sedimentation
corresponds to peat and organic silts of 3.5 to 5 m
thickness (U3). Such sediment has been deposited during
the Subboreal and the Subatlantic periods in accordance
with a radiocarbon age (380540 14
C BP) obtained from
the middle of this sequence. The sedimentation becomes
progressively siltier (U5) and we observed silt layers with
fragments of oncoids intercalated in the organic sedimentation
(U4) as in reach 3a. Finally, a silt sedimentation has
filled in the valley bottom except in some places where the
sedimentation remains organic (U6).
4.1.5. Reach 3c
Downstream, at the bottom of the Old Church of Thaon,
the sedimentation corresponds mainly to tufa deposits
(Figs. 5 and 6). The core drilling of Thaon presents five
sedimentary units. Calcareous gravel and cobbles in a
greyish sandy matrix (U1) are bounded by grey silts (U2).
A date (921468 14
C BP) gives a Preboreal age for this
unit. These are overlain by tufaceous sediments, which
have built a large unattached travertine deposit which
occupies the entire valley bottom. At first, the sedimentation
corresponds to oncolitic facies, mainly constituted by
fragments of oncoids, interbedded with thin silty layers
(U3). Then, we noted a coarser deposit, with spherical and
cylindrical oncoids predominant (U4). This sequence is
dated to the Boreal and the Atlantic periods by a radiocarbon
dating (522069 14
C BP) obtained at the top. The
tufaceous sediment is covered by grey silts, more or less
organic or carbonated (U5).
4.1.6. Reach 4
Along 6 km, the sediment is mainly organic. The core
drilling at Fontaine­Henry enables us to observe five
sedimentary units. At the base, calcareous gravel and
cobbles in a greyish sandy matrix (U1) are bounded by
grey silt (U2). These are overlain by organic sediment
which fills in the entire valley bottom and can reach 7 m
(U3). The lower sediments in this unit are a fibrous peat
radiocarbon dated to the second part of the Subboreal
(447540 and 242040 14
C BP, Fontaine­Henry) in
accordance with the pollen assemblage identified at R4S.
This was then followed by the deposition of organic silts,
often alternating with minerogenic or tufaceous silts (U4).
Finally, a silty sediment has filled in the valley bottom
except in some places where the sedimentation remains
organic and in some places peaty (U5).
4.1.7. Reach 5
This reach corresponds to the tufa deposit of Reviers.
Core Re1 has five sedimentary units and improved results
obtained by previous studies (Clet-Pellerin et al., 1990). At
the base, we observed sandy grey silt covered by more fine
dark grey silt. Tufa sediment then built a huge unattached
travertine deposit which takes up the entire valley bottom
as in reach 3c. The sediment is of fine tufaceous oncolitic
layers (30­50 cm) alternating with organic silt or peat (5­
10 cm). This sequence is Atlantic, as testified by 3 radiocarbon
dates (with ages ranging from 771560 to 5315
60 14
C BP). Overlaying this is a coarser deposit where the
spherical oncoids are predominant with several thin
organic layers. This unit belongs to the Subboreal and
Subatlantic periods, as suggested by the palynostratigraphy.
Finally, the tufaceous sediment is covered by silt.
4.2. Longitudinal pattern at the local scale, from
tufaceous to organic infilling
Several core drillings with palynological data and
radiocarbon dating control allow an understanding of
the succession of tufaceous to organic deposits from the
third to the fourth reaches (Fig. 5). The onset of the
tufaceous sediment at Reviers is dated from the Boreal
period. The vertical accretionary trend is maintained to
Fig. 5. Bore holes and core drilling locations and main Holocene
sedimentation in the valley bottom from the end of the third to the fourth
reaches.
62 L. Lespez et al. / Geomorphology 98 (2008) 55­70
the Atlantic, while organic sediment developed upstream
is mainly dated to the Subboreal. Consequently,
the organic sedimentation is probably in relation to a
higher water level in the valley bottom partly caused
by the tufaceous barrage, which progressively made
drainage more difficult. Downstream, the transition
from tufaceous to organic sedimentation is progressive
(Figs. 5 and 7). Along 0.5 km, the Subboreal sedimentation
is composed by silts and oncoids which come
from the erosion of oncolithic sediments, and by organic
silts which indicate their deposition in a fen floodplain.
Further downstream, sedimentation is mainly organic.
4.3. Valley bottom environmental changes: contribution
of multi-proxy data
4.3.1. The pollen data
The pollen analyses have been undertaken on seven
core-drillings in the Mue valley and the detailed results
are published in Clet-Pellerin et al. (1990) and Lespez
et al. (2005b). The main data are summarized in Fig. 8
and highlight water level changes and human impacts on
the vegetation cover of the valley bottom and slopes.
During the first part of the Holocene, alder-hazel woodland
was probably surrounded by mixed oak woodland.
But, in the valley bottom, the environment was never
totally forested because the wetlands experienced
specific vegetation. Three periods are characterized by
spreading of wetland vegetation cover in relation with a
high water level: 5200­4800 14
C BP at Rots and
Reviers, 2900­2200 14
C BP and 1500­1000 14
C BP at
Fontaine­Henry and Rots.
Human impact on the vegetation cover was deduced
from the occurrence of ruderal, meadow, pasture and
cultivated plants in the pollen diagrams. During the
Boreal and the Lower Atlantic periods the ecological
impact of Mesolithic peoples was very limited. At the
end of the Atlantic, the impact of the Early and Middle
Fig. 6. Chronostratigraphy and sedimentological analyses of the Thaon core.
63L. Lespez et al. / Geomorphology 98 (2008) 55­70
Neolithic (5700­5200 14
C BP) has been detected
locally near the archaeological sites of Rots and Cairon
by increases of ruderal plants and the first occurrence of
cereals (Lespez et al., 2005b; Ghesquire and Marcigny,
in press). However, during the Late Neolithic, the
clearance episodes were probably localised in time and
in space: the pollen data suggest small-scale temporary
clearances with regeneration. From the Middle Bronze
Age, ruderal and meadow plants appear frequently and
the presence of cereal is regularly testified. The Bronze
Age is a period of widespread but localised vegetation
change. The main period of forest clearance and lasting
land use change in the valley bottom is the Iron Age, as
observed in many valleys of Normandy characterized by
systematic clearance (Clet-Pellerin and Verron, 2004;
Lespez et al., 2004). After this period the entire river
basin was cultivated, and fallow land and wooded areas
were scarce.
4.3.2. The malacological data
The biostratigraphy of the tufa deposits of Thaon
follows the Holocene molluscan succession established
at Saint­Germain­Le-Vasson (Limondin-Lozouet and
Preece, 2004). Molluscan assemblages allow us to
define five biozones illustrating evidence of local
changes in the valley bottom environment. The detailed
results of the malacological analyses are published in
Lespez et al. (2005b). During the Boreal, snail assemblages
indicate a patchy environment with woody areas
and grassland. During the main part of the Lower Atlantic,
low numbers of shells and predominance of aquatic
species may indicate an increase in flood frequency.
During the Upper Atlantic molluscan populations allow
us to recognize a composite landscape, mixing grassy
areas and wooded patches. No clear indications of human
impact on the local environment are registered by
malacofaunas despite the proximity of several archaeological
sites of the Late Neolithic. The Late Atlantic has
faunas characterized by aquatic and marshy species. They
might be related to a palaeohydrological oscillation
around 5000 year 14
C BP identified in the palynological
data. Following this period molluscan assemblages are
dominated by open-ground snails. This allows us to
recognize a more open environment and the decline of
forested areas, suggesting a stronger impact of human
activity, although onset of this episode remains unclear
due to the lack of accurate dating controls.
5. Discussion: Fluvial system adjustments to climatic,
human and autogenic controls and comparison with
northwest European river systems
5.1. Preboreal and Boreal (10,000­8000 14
C BP)
Along the valley, Late Pleistocene fluvial sands and
gravels are covered by grey silts (Fig. 9). Most of these
sediments come from the erosion of the loess widely
deposited on the plateau and valley bottom. They
Fig. 7. Longitudinal morphostratigraphic cross-section from the third to the fourth reaches.
64 L. Lespez et al. / Geomorphology 98 (2008) 55­70
Fig. 8. Core drillings in the Mue valley and simplified pollen diagrams (% calculated from total number of pollen grains, fern and aquatic plants excluded).
65L.Lespezetal./Geomorphology98(2008)55­70
Fig. 9. Synthetic diagram of palaeoenvironmental changes in the Mue valley during the Holocene.
66L.Lespezetal./Geomorphology98(2008)55­70
correspond mainly to overbank deposits related to a well
defined stream, probably a meandering system, while the
environment of the valley bottom was composed of wood
and grassland patches, as indicated by the malacological
assemblages. The fluvial system functions as a continuum
from upstream to downstream and the valley bottom still
experienced good drainage. The development of floodplain
backswamp areas, which has characterized many
valley bottoms of southern England and northwest Europe
(Brown, 1995; Pastre et al., 2001; Antoine et al., 2003;
Collins et al., 2006), was experienced later in the Mue
valley. Locally, we observe the onset of tufa deposits
which is more widely registered in valleys of the Paris
basin (Antoine et al., 2003), lowland Britain and western
Europe (Brown, 1997; Collins et al., 2006).
5.2. Lower Atlantic (8000­6500 14
C BP)
The travertine deposit of Thaon is the first element of
the calcareous sediment that occurs in several areas of
the valley bottom. Tufa formation indicates suspended
load decreasing, whereas dissolved load becomes the
predominant discharge in the fluvial system. This is
explained by the stabilization of the slopes in a wellforested
environment which minimizes runoff and the
supply of fine-grained material into the valley (Pastre
et al., 2001). In the Beuvronne valley, Orth et al. (2004)
show that the quartz content of the valley bottom filling
is less than 1% during this period. The increase of
dissolution is related to soil formation, spring activities
and higher water tables. The calcareous sediment is
often interbedded with organic layers which are related
to the increase of marshy areas. This period appears as
the local sedimentary signature of the Holocene Climatic
Optimum. Impacts of the Neolithic populations on
the river system remain low. The building of the main
unattached travertine of Thaon and Reviers is responsible
for the major discontinuity of the fluvial system.
This valley bottom architecture underlines the role
of geological setting as observed in many small river
systems of lowland sedimentary areas in northwest Europe
(Pedley et al., 2000). However, organic sedimentation
appears predominant in the Paris basin (Pastre
et al., 2001, 2003; Antoine et al., 2003; Orth et al., 2004),
lowland Britain and more generally in western European
valleys (Brown, 1997; Gibbard and Lewin, 2002).
5.3. Upper Atlantic and Lower Subboreal (6500­4500
14
C BP)
During the Upper Atlantic, the peak of tufa sedimentation
was reached, as in most northwest European
regions (Goudie et al., 1993; Pedley et al., 2000;
Pentecost, 2005). Nevertheless, organic sedimentation
increased in the valley bottom. This is explained by a
higher water table. This sedimentation pattern indicates
an inactive meandering or anastomosing river system.
The river would not have had the power to migrate
freely in the dense alder floodplain woodland of the
Lower Subboreal as in British valleys (Brown, 1995).
Despite the importance of Middle Neolithic settlement in
the river basin, pollen indicates that woodland clearance
seems to have been scarce and temporary on the valley
bottom and on surrounding slopes. Consequences of
human activities appear only locally, in the immediate
vicinity of archaeological sites. Very low sedimentological
changes in relation with human activities are observed
in the Mue valley during the Atlantic and the
Lower Subboreal periods, like in most of British valleys
(Macklin, 1999; Collins et al., 2006). In contrast, they are
more clearly recorded in other areas of the Paris basin
where lateral erosion processes, silt and sand deposits are
observed, even if they have still little impact on channel
dimensions and geometry (Pastre et al., 2001).
5.4. Middle part of the Subboreal (4500­3500 14
C BP)
During the middle part of the Subboreal, we observe
the generalization of sedimentation along all the reaches.
If the unattached travertine barrage of Reviers was still
active, tufaceous sedimentation had ended at Thaon.
More generally, organic sedimentation (organic silts and
peat) became predominant. In the valley bottom, the
abrupt decrease of the alder swamp forest could be
related to the development of open grassland and pasture
(Brown, 1997). In the surrounding areas, forest cover
changes are more difficult to identify while the available
archaeological data indicate numerous settlements on the
plateau during the Late Neolithic (Clet-Pellerin and
Verron, 2004). Therefore, fluvial activity remained low,
probably in relation to low flood discharges and low
level of soil erosion and colluvial transfer. During this
period, autogenic control was the dominant-controlling
factor related to the travertine deposits which created
discontinuities in the valley bottom. The increase of
brown silts, related to impacts of Neolithic and early
Bronze Age populations, and recorded in the other parts
of the Paris basin (Pastre et al., 2001), is not experienced
in the Mue fluvial system.
5.5. Late Subboreal (3500­2800 14
C BP)
This period is characterized by important changes in
the fluvial system. Within reaches 2, 3b and 4, organic
67L. Lespez et al. / Geomorphology 98 (2008) 55­70
sediments are often covered by a poorly sorted and multimodal
tufaceous silt sediment. It suggests erosion of
tufaceous sediments of the valley bottom and of loamy
soils developed on the plateau. It implies also well-defined
streams and powerful flows, probably in a meandering
system. The radiocarbon dates obtained at the bottom of
these deposits within the third and fourth reaches indicate
an age post 330040 14
C BP and 324545 14
C BP. This
short fluvial change can be explained by the convergence
of climatic and anthropogenic factors. Indeed, it is contemporaneous
with a wet oscillation around 3500­3000
14
C BP experienced in northwest Europe (Magny, 1992;
Barber et al., 2003) and well documented by the fluvial
systems of the Paris basin (Pastre et al., 2001, 2006) and
Great Britain (Lewin and Macklin, 2003; Lewin et al.,
2005). Furthermore, the higher suspended load supply can
be related to an increase of land use changes, testified by
the onset of cereals and the development of ruderal and
meadow plants in pollen diagrams (Fig. 8). Moreover, the
archaeological data indicate a more dense settlement
pattern during the Middle Bronze Age and the beginning
of the Late Bronze Age (Marcigny et al., in press). Such a
hypothesis is in accordance with research conducted in the
central part of the Paris basin (Pastre et al., 2001, 2006)
and more generally in northwest European valleys, where
accelerated fine alluviation took place and is often
associated with land use controls coupled with erosive
storms or climatic oscillation (Brown, 1997; Pastre et al.,
2006; Collins et al., 2006).
5.6. Subatlantic (post 2800 14
C BP)
The number and the size of floodplain swamps greatly
decreased. The valley bottom experienced silt sedimentation
mainly constituted by overbank and colluvial
deposits. Floodplain vertical accretion was enhanced by
accelerated soil erosion and production of fine sediments
resulting from anthropogenic influences. This change
has also been observed in the Seulles floodplain, where 3
to 4 m of overbank sediment were deposited after 2241
41 BP (Lespez et al., 2006). The farming of the river
basin was probably complete as testified by the pollen
diagrams of the valley and more generally in Normandy
(Clet-Pellerin and Verron, 2004; Lespez et al., 2004).
Furthermore, numerous small settlements are testified
from the Late Bronze age and the beginning of the Iron
Age (Hallstatt) on the plateau. In addition, during the La
Tne period, archaeological data and carpological
remains (cereal, leguminous plants) indicate complete
settlement of the river basin with groups of big farms
which controlled several hundred hectares of cultivated
land (San Juan et al., 1999). Nevertheless, the onset of
this change can also be related to a conjunction with a
wet oscillation, locally witnessed by the available pollen
data and more generally registered in northwest Europe
(Magny, 1992; Van Geel et al., 1996; Barber et al., 2003)
and experienced by numerous fluvial systems (Pastre et
al., 2001; Lewin and Macklin, 2003; Pastre et al., 2006),
including those of the western Paris basin (Larue, 2002;
Sebag, 2002; Germain-Vallée and Lespez, 2006). The
land use change exceeded a threshold which rendered the
fluvial system more sensitive to large scale climatic
changes or individual hydrological events. The fluvial
system once again became a continuum without
significant changes from upstream to downstream. This
was reinforced during the last millennia by the control of
the flow to supply the numerous water-mills built along
the river (Lespez et al., 2005a).
During the first part of the Subatlantic period (Iron Age
and Roman period) increasing alluviation, related to soil
erosion, was widespread throughout the Paris basin
(Kuzucuoglu et al., 1992; Pastre et al., 2001, 2006) and
northwest Europe (e. g. Brown, 1997). Therefore,
important variability in alluvial records reflects catchment
conditions and human intervention within the river
channel. For example, in other parts of the Paris basin,
the development of peat deposition is still registered in
many small tributaries during the first part of the
Subatlantic period (Pastre et al., 2001, 2006). At least,
during the Middle Ages, floodplain management explains
why a small lowland river in northwest Europe has
experienced restricted channel change (Brown, 1997). In
contrast, major and upland river systems are still sensitive
to climate-related variations such as the Little Ice Age
(Macklin, 1999; Pastre et al., 2001; Gregory et al., 2006).
6. Conclusion
Research undertaken in the Mue valley gives for the
first time a precise chronostratigraphical pattern of
Holocene alluvial filling in the Plain of Caen and more
generally in Normandy. It allows comparisons with both
sides of the English Channel. The Mue valley record is
broadly similar to other fluvial systems of the Paris
basin and lowland Britain (Pastre et al., 2001; Brown,
1997; Gibbard and Lewin, 2002) but presents particular
features, especially in relation to its limestone catchment,
important groundwater supply and high solute load.
The longitudinal pattern appears particularly complex.
During the Boreal and Subatlantic, the system was a
continuum with a well-marked stream. In contrast, the
variability of the longitudinal arrangement was high from
around 8500 until 3500 14
C BP. Thus, the impact of
climatic oscillations on the river system is difficult to
68 L. Lespez et al. / Geomorphology 98 (2008) 55­70
demonstrate because of the low stream power and the
limited slope-channel coupling. The role of the travertine
barrage in the triggering of organic sedimentation was
identified, indicating an autogenic control. From 3500 14
C
BP, the fluvial system changed completely, with human
control appearing dominant. Land use effects explain the
increase of suspended load. Flow turbidity is one of the
main factors which explain the "Late Holocene Tufa
decline" registered in the Mue valley, as in other European
areas (Goudie et al., 1993). Nonetheless, as tufaceous
sedimentation reached its peak around 5000 14
C BP,
palaeohydrological influences can also be invoked due to
the loweringof the watertable during thesecond part of the
Holocene. After 2800 14
C BP, the fine mineral sediment
supply has prevented organic and tufaceous sedimentation
in most of the valley bottom. Furthermore, this land use
change increases the potential sensitivity of the catchment
to climatic oscillations and individual hydrological events,
as the river system became once again a continuum due to
the efficiency of slope-channel coupling and the transport
capacity of stream flows. However, since the medieval
period, the coupling between slopes and stream was
progressively broken, especially due to river management
associated with the water mills system.
Small order river systems located in a limestone
catchment, characterized by low fluvial dynamics and
fluvial pattern changes during the last millennium, appear
homogenous andrelatively simple tounderstand.However,
research on the Mue valley has demonstrated spatial and
temporal complexities for such fluvial systems. This
implies the necessity, when undertaking floodplain palaeohydrological
research, to take into account (1) the balance
between surface runoff and groundwater supply and (2)
longitudinal discontinuities resulting from tectonic pattern,
geomorphological heritage and/or autogenic controls.
Acknowledgments
This work was supported by a Research Program
from the Fond National pour la Science (ACI "Jeunes
chercheurs" no. 6079) and a "Prospection thématique"
from the Archaeological services of Calvados and BasseNormandie.
We wish to thank G. San Juan, J.-M. Cador,
J. Douvinet for their help and R. Tipping and R. Fyfe for
useful comments on the manuscript.
References
Antoine, P., Munaut, A.-V., Limondin-Lozouet, N., Ponel, P.,
Dupéron, J., Dupéron, M., 2003. Modification of fluvial system
in relation to climatic modifications during the Lateglacial and
early Holocene in the Somme Basin (Northern France). Quaternary
Science Reviews 22, 2061­2076.
Barber, K.E., Chambers, F.M., Maddy, D., 2003. Holocene palaeoclimates
from peat stratigraphy: macrofossil proxy climate records
from three oceanic raised bogs in England and Ireland. Quaternary
Science Reviews 22, 521­539.
Benito, G., Baker, V.R., Gregory, K.J. (Eds.), 1998. Palaeohydrology
and Environmental Change. Wiley, Chichester.
Brown, A.G., 1990. Holocene floodplain diachrony and inherited
downstream variations in fluvial processes: a study of the River
Perry, Shropshire, England. Journal of Quaternary Science 5,
39­51.
Brown, A.G., 1995. Lateglacial­Holocene sedimentation in lowland
temperate environments: floodplain metamorphosis and multiple
channel systems. In: Frenzel, B., (Ed.), European river activity and
climate change during the Lateglacial and early Holocene, European
Palaoclimate and Man 9, ESF, Gustav Fischer Verlag, Stuttgart.
Brown, A.G., 1997. Alluvial geoarchaeology. Floodplain archaeology
and environmental change. Cambridge manuals in Archaeology.
Cambridge University Press, Cambridge.
Brown, A.G., 1999. Characterising prehistoric lowland environments
using local pollen assemblages. Quaternary Proceedings 7, 585­594.
Cador, J.-M., Cantat, O., Lespez, L., 2001. Un tableau de bord pour les
SAGE, Proposition d'outil de gestion et d'évaluation pour les Schémas
d'Aménagement et de Gestion des Eaux sur le bassin versant de la Mue
(Calvados). Proc. Conf. Hydrosystme, paysage, territoire. CNFG,
Lille, France. CD-Rom.
Clet-Pellerin, M., Verron, G., 2004. Influence de l'Homme sur
l'évolution des paysages normands pendant l'Holocne. In: Richard,
H. (Ed.), Néolithisation précoce. Premires traces d'anthropisation
du couvert végétal  partir de données polliniques. Annales
Littéraires, Série Environnement, sociétés et archéologies, vol. 7.
Presses Universitaires Franc-Comtoises, Besançon, pp. 53­68.
Clet-Pellerin, M., Helluin, M., Pellerin, J., Pareyn, C., 1990. Le Tuf
de Reviers (Calvados). Proc. Conf. Tufs et travertins quaternaires
des bassins de la Seine et de la Somme, et des Régions
limitrophes. Bulletin du Centre de Géomorphologie de Caen 38,
199­213.
Collins, P., Worsley, P., Keith-Lucas, D.M., Fenwick, I.M., 2006.
Floodplain environmental change during the Younger Dryas and
Holocene in Northwest Europe: Insights from the lower Kennet Valley,
south central England. Palaeogeography, Palaeoclimatology, Palaeoecology
233, 113­133.
Desloges, J., 1997. Les premires architectures funéraires de BasseNormandie.
Proc. Int. Conf. La culture de Cerny, nouvelle économie,
nouvelle société au Néolithique, Mémoires du Musée de
Préhistoire d'Ile-de-France, vol. 6, pp. 515­539.
Elha, H., 1963. La Normandie occidentale entre la Seine et le golfe
normano-breton. Etude morphologique. Bire, Bordeaux.
Folk, R.L., Ward, W.C., 1957. Brazos River bar, a study in the significance
of grain size parameters. Journal of Sedimentary Petrology 27, 514.
Font, M., 2002. Signature géomorphologique des déformations en
domaine intraplaque : application  la Normandie. PhD. Thesis,
Université of Caen-Basse Normandie, France.
Germain-Vallée, C., Lespez, L., 2006. Dynamique holocne d'un fond
de vallée normand (Laizon, Calvados), approche géomorphologique
et micromorphologique. In: Allée, P., Lespez, L. (Eds.),
L'érosion entre Société, Climat et Paléoenvironnement. Actes de la
Table Ronde en l'honneur de R. Neboit-Guilhot. Coll. Nature et
Société, vol. 3. Presses Universitaires Blaise Pascal, Clermont
Ferrand, pp. 279­284.
Ghesquire, E., Marcigny, C. (Eds.) sous-presse. La  Pierre
Tourneresse   Cairon (Calvados). Un habitat puis un monument
funéraire du Néolithique moyen, Société Préhistorique Française.
69L. Lespez et al. / Geomorphology 98 (2008) 55­70
Gibbard, P.L., Lewin, J., 2002. Climate and related controls on
interglacial fluvial sedimentation in lowland Britain. Sedimentary
Geology 151, 187­210.
Goudie, A.S., Viles, H.A., Pentecost, A., 1993. The Late Holocene tufa
decline in Europe. The Holocene 3, 181­186.
Gregory, K.J., Starkel, L., Barker, V.R. (Eds.), 1995. Global
continental palaeohydrology. J. Wiley & Sons, Chichester.
Gregory, K.J., Macklin, M.G., Walling, D.E. (Eds.), 2006. Past
hydrological events related to understanding Global ChangeCatena,
vol. 66, pp. 1­2.
Harvey, A.M., 2002. Effective timescale of coupling within fluvial
systems. Geomorphology 44, 175­201.
Hooke, J., 2003. Coarse sediment connectivity in river channel systems: a
conceptual framework and methodology. Geomorphology 56, 79­94.
Kasai, M., Brierley, G.J., Page, M.J., Marutani, T., Trustrum, N.A.,
2006. Impacts of land use change on patterns of sediment flux in
Weraamaia catchment, New Zealand. Catena 64, 27­60.
Kuzucuoglu, C., Lespez, L., Pastre, J.-F., 1992. Holocene colluvial deposits
ontheslopesoftheParisBasin.In:Bell,M.,Boardman,J.(Eds.),Pastand
Present Soil Erosion. Oxbow Monograph, vol. 22. Oxbow Books,
Oxford, pp. 115­123.
Lagarde, J.L., Baize, S., Amorese, D., Delcaillau, B., Font, M., Volant, P.,
2000. Active tectonics, seismicity and geomorphology with special
reference to Normandy (France). Journal of Quaternary Science 15,
745­758.
Larue, J.P., 2002. Small valley bottom deposits in the sandy districts of
the Sarthe basin (France): climatic and/or human origin? Geomorphology
45, 309­323.
Leroyer, C., 1997. Homme, climat, végétation au Tardi-et Postglaciaire
dans le Bassin Parisien, apports des études palynologiques des
fonds de vallées. PhD. Thesis, Université Paris 1, France.
Lespez, L., Clet-Pellerin, M., Davidson, R., Marcigny, C., Levalet, F.,
Hardel, B., 2004. Evolution des paysages et anthropisation depuis le
Néolithique dans la péninsule de La Hague. Revue d'Archéométrie 28,
71­88.
Lespez, L., Garnier, E., Cador, J.-M., Rocard, D., 2005a. Les aménagementshydrauliques
et la dynamiquedespaysagesdes petitscoursd'eau
depuis le XVIIIe
sicle dans le nord-ouest de la France: l'exemple du
bassin versant de la Seulles (Calvados). Aestuaria 7, 89­109.
Lespez, L., Clet-Pellerin, M., Limondin-Lozouet, N., Pastre, J.-F.,
Fontugne, M., 2005b. Discontinuités longitudinales des dynamiques
sédimentaires holocnes dans les petites vallées de l'Ouest du Bassin
Parisien, l'exemple de la Mue. Quaternaire 16 (4), 173­198.
Lespez, L., Cador, J.-M., Germaine, M.-A., 2006. Gestion de l'eau et
trajectoire des petites vallées, exemples normands. Proc. Int. Conf.
Interactions Nature-Société. Analyses et modles. LETG, La Baule.
CD-ROM.
Lesueur, P., Lesourd, S., Lefebvre, D., Garnaud, S., Brun-Cottan, J.-C.,
2003. Holocene and modern sediments in the Seine Estuary
(France): a synthesis. Journal of Quaternary Science 18, 339­349.
Lewin,J., Macklin,M.G., 2003.Preservation potentialfor Late Quaternary
river alluvium. Journal of Quaternary Science 18, 107­120.
Lewin, J., Macklin, M.G., Johnstone, E., 2005. Interpreting alluvial
archives: sedimentological factors in the British Holocene fluvial
records. Quaternary Science Reviews 24, 1873­1889.
Limondin-Lozouet, N., Preece, R.C., 2004. Molluscan successions from
the Holocene tufa of St Germain-le-Vasson, Normandy (France)
and their biogeographical significance. Journal of Quaternary
Science 19 (1), 55­71.
Macklin, M.G., 1999. Holocene river environments in Prehistoric
Britain: Human Interaction and impact. In: Edwards, K.J., Sadler,
J.P. (Eds.), Holocene Environments of Prehistoric BritainQuaternQuaternary
Proceedings, vol. 7, pp. 521­530.
Maddy, D., Macklin, M.G., Woodward, J.C. (Eds.), 2001. River basin
sediment systems: archives of environmental change. Balkema, Lisse.
Magny, M., 1992. Holocene lake-level fluctuations in Jura and the
northern subalpine ranges, France: Regional pattern and climatic
implications. Boreas 21, 319­334.
Mangerud, J., Andersen, S.T., Berglund, B.E., Donner, J.J., 1974.
Quaternary stratigraphy of Norden, a proposal for terminology and
classification. Boreas 3, 109­128.
Marcigny, C., Lespez, L., Ghesquire, E., in press. Emprise et déprise
agricole  l'âge du Bronze moyen sur le littoral de la Manche ? Une
lecture du phénomne grâce aux sites normands. Proc. Conf.
CTHS 2004, Besançon.
Orth, P., Pastre, J.-F., Gauthier, A., Limondin-Lozouet, N., Kunesh, S.,
2004. Les enregistrements morphosédimentaires et biostratigraphiques
des fonds de vallée du bassin versant de la Beuvronne (Bassin
parisien, Seine et Marne, France): perception des changements
climato-anthropiques  l'Holocne. Quaternaire 15, 285­298.
Pastre, J.-F., Limondin-Lozouet, N., Gebhardt, A., Leroyer, C.,
Fontugne, M., Krier, V., 2001. Lateglacial and Holocene fluvial
records from the central part of the Paris Basin (France). In: Maddy,
D., Macklin, M.G., Woodward, J.C. (Eds.), River basin sediment
systems: archives of environmental change. Balkema, pp. 357­373.
Pastre, J.-F., Limondin-Lozouet, N., Leroyer, C., Ponel, P., 2003. River
System evolution and environmentalchanges during the Late Glacialin
the Paris Basin (France). Quaternary Science Reviews 22, 2177­2188.
Pastre, J.-F., Orth, P., Le Jeune, Y., Bensaadoune, S., 2006. 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'Holocne. In: Allée, P., Lespez, L. (Eds.), L'érosion entre Société,
Climat et Paléoenvironnement. Actes de la Table Ronde en l'honneur
de R. Neboit-Guilhot. Coll. Nature et Société, vol. 3. Presses
Universitaires Blaise Pascal, Clermont Ferrand, pp. 237­247.
Pedley, H.M., Hill, I., Denton, P., Brasington, J., 2000. Three-dimensional
modelling of a Holocene tufa system in the Lathkill Valley, north
Derbyshire,usingground-penetratingradar.Sedimentology47,721­737.
Pentecost, A., 2005. Travertine. Springer, Berlin.
San Juan, G., Meniel, P., Matterne-Zeck, V., Savary, X., Jardel, K.,
1999. L'occupation gauloise au Nord-Ouest de Caen. L'évaluation
en sondage du plateau de Thaon (Calvados). Revue Archéologique
de l'Ouest 16, 131­194.
Sebag, D., 2002. Apports de la matire organique pour la reconstitution des
paléoenvironnements holocnes de la basse vallée la Seine, fluctuations
des conditions hydrologiques locales et environnements de dépôt. PhD.
Thesis, Université de Rouen.
Starkel, L., 1983. The reflection of hydrologic changes in the fluvial
environment of the temperate zone during the last 15000 years. In:
Gregory, K.J. (Ed.), Background to the Palaeohydrology, a
Perspective. Wiley, Chichester, pp. 213­235.
Stuiver, M., Reimer, P., 1993. Extended 14C data base and revised Calib.
3.0 14C age calibration programme. Radiocarbon 35, 215­230.
Van Geel, B., Buurman, J., Waterbolk, H.T., 1996. Archaeological and
palaeocological indications about an abrupt climate change in The
Netherlands, and evidence for climatological teleconnections
around 2650 BP. Journal of Quaternary Science 9, 109­118.
70 L. Lespez et al. / Geomorphology 98 (2008) 55­70