® .Geomorphology 33 2000 183­207
Long-term fluvial archives in NW France: response of the Seine
and Somme rivers to tectonic movements, climatic variations and
sea-level changes
Pierre Antoine a,)
, Jean Pierre Lautridou b
, Michel Laurent c
a
UMR 9944 CNRS ``Prehistoire et Quaternaire'' unite: Stratigraphie et Paleoenironments Quaternaires, UFR de Geographie,´ ´ ´ ´
Uniersite des Sciences et Technologies de Lille, 59 655 Villeneue d'Ascq cedex, Francé
b
ER 109 CNRS, Centre de Geomorphologie, 24 rue des Tilleuls 14 000 Caen, Francé
c
ENS Chimie Paris, 11 rue Pierre et Marie Curie, 75231 Paris, France
Accepted 18 July 1999
Abstract
The Seine and the Somme are the two main rivers flowing from northwestern France into the Channel. During the
Pleistocene cold stages both rivers were tributaries of the River Manche which was exporting sediments into the central
deeps of the Channel. The River Seine has a very well developed terrace system recording incision that began at around 1
Ma. The same age is proposed for the beginning of the main incision in the Somme Valley on the basis of morphostratigra-
phy, pedostratigraphy, palaeontology, palaeomagnetism and ESR datings. The uplift rate deduced from analysis of the Seine
and Somme terrace systems is of 55 to 60 mrMa since the end of the Lower Pleistocene. The response of the two rivers to
climatic variations, uplift and sea-level changes is complex and variable in the different parts of the river courses. For
example, the evolution of the lower Seine system is influenced by uplift and climate changes but dominated by sea-level
changes. In the middle Seine the system is beyond the impact of sea-level variations and shows a very detailed response to
climatic variations during the Middle and Upper Pleistocene in a context of uplift. The Somme Valley response appears to be
more homogeneous, especially in the middle valley, where the terrace system shows a regular pattern in which incision
occurs at the beginning of each glacial period against a general background of uplift. Nevertheless, the lower Somme Valley
and the Palaeo-Somme in the Channel area indicate some strong differences compared with the middle valley: influence of
sea-level variations and probably differences in rates of tectonic uplift between the Channel and the present continent. The
differences in the responses of the two river valleys during the Pleistocene are related to differences in the size of the fluvial
basins, to the local tectonic characteristics, to the geometry of the platform connected to the lower parts of the valleys and to
)
Corresponding author. Fax: q33-3-20-33-60-75.
® .E-mail address: pierre.antoine@univ-lille1.fr P. Antoine .
0169-555Xr00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved.
® .PII: S0169-555X 99 00122-1
( )P. Antoine et al.rGeomorphology 33 2000 183­207184
the hydrodynamic characteristics of each river. Finally, it is shown from these examples that the multidisciplinary study of
Pleistocene rivers is a very efficient tool for the investigation of neotectonic activity. q 2000 Elsevier Science B.V. All
rights reserved.
Keywords: River Seine; River Somme; Quaternary; neotectonics
1. Introduction: location and background
Today, the Seine and the Somme are two of the
major rivers flowing from northwestern France into
the Channel. During the Quaternary cold stages they
were the main tributaries of the ``Channel River''
®Auffret et al., 1980, 1982; Gibbard, 1988; Lerico-
.lais, 1997 and accounted for a very important part
of the drainage system of the present-day eastern
® .Channel area Fig. 1A . Both rivers are characterised
by very well developed terrace systems, especially in
their middle parts and are mainly incised into Creta-
® .ceous and Jurassic calcareous bedrock Fig. 1B .
Nevertheless, in the area submerged beneath the
Channel at the present-day, the Seine Valley is in-
cised into Jurassic calcareous bed-rock, while the
®Somme is incised into Tertiary sands and clays Fig.
.1B . Both rivers are located in the same climatic area
and share the same tectonic background, charac-
terised by slow uplift of the northwestern part of the
® .Paris Basin Pommerol, 1978; Founiguet, 1987 . Lo-
cal ``bloc tectonic'' activity during the Pleistocene
has been demonstrated for the northern part of the
® .area Colbeaux et al., 1977, 1978, 1979, 1980 . The
whole area is also characterised by a very clear
adaptation of the river courses to the geological
structure, which is mainly dominated by NW­SW
features.
Nevertheless these two rivers show some strong
differences: the Seine Valley is characterised by very
large meanders while the Somme Valley is straighter,
especially in its central part, and the sizes of the
® 2
fluvial basins are very different Seine: 76,000 km ;
2 .Somme: 5,800 km .
The aim of this paper is to attempt a comparison
between the responses of the Seine and Somme
rivers to tectonic movements, climatic variations and
sea-level changes during the Pleistocene, based on
new data concerning the Pleistocene evolution of
®these river systems Lautridou, 1982, 1985, Lautri-
dou et al., 1984, 1999; Lefebvre et al., 1994; An-
toine, 1990, 1993, 1994a,b, 1997a,b, 1998; Antoine
et al., 1995, 1998, 1999; Laurent, 1993, Laurent et
.al., 1994; Lericolais, 1997 .
2. The fluvial system of the River Seine from the
end of the Pliocene to the Weichselian Lateglacial
2.1. The Pliocene­Lower Pleistocene history of the
Seine­Loire
The first stage of Seine­Loire Pliocene­Lower
Pleistocene history is represented by the ``Sables de
® .Lozere'' Lozere Sands . They are plateau deposits` `
and are characterised by granules and fine sands
derived from the Massif Central. At that time the
River Loire was a tributary of the Seine. Down-
stream of Rouen these sands are fluvio-marine and
® .their age is about 3.5 My Lautridou, 1985 .
The second stage concerns the ``very high ter-
® . ® .races'' q120 m of the River Seine Figs. 2 and 3 .
After deposition of the Sables de Lozere, the Upper`
Normandy plateau was covered by the marine Saint-
® .Eustache sands Fig. 3 and locally, near Rouen, by
the lagoonal Reuverian and Praetiglian La Londe
®Clay Kuntz and Lautridou, 1974; Clet-Pellerin,
.1983; Lautridou, 1985 .
The ``very high terraces'' are the first witnesses
of the incision by the rivers into the Cretaceous
® .chalk plateau 30 to 40 m . The gravels of these very
high terraces contain augite, derived from volcanoes
of the Massif Central, that has been dated at 1 Ma
® .Tourenq and Pomerol, 1995 . It is probable that the
River Loire continued to be a tributary of the Seine
at this time. Based on the record of the long se-
®quences of La Londe, the tectonic events graben
.formation had largely finished by the end of the
® .Lower Pleistocene Lautridou, 1985 .
( )P. Antoine et al.rGeomorphology 33 2000 183­207 185
® . ® .Fig. 1. A Location of the Seine and Somme river valleys and of their extension beneath the present-day English Channel ; B Simplified
®geology and structure of the Eastern Channel according to Lericolais, 1997 modified, and to the Carte Geologique de la France, BRGM,´
.1996 . NPV: Northern Palaeovalley, MPV: Median Palaeovalley, GRD: Greenwich Deep, CD: Cotentin Deep, BF: Bray Fault, FS: Seine
.Fault, SVBF: St. Valery­Boismont faults .
2.2. Middle and Upper Pleistocene in the lower
( )Seine Valley Rouen­Le Hare
®In this area there are relatively few terraces 7
.large bedrock steps in comparison to those of the
® .River Somme 10 bed-rock steps , but in a system of
very wide meanders the record of the fluvial estuar-
® .ine sediments is very well preserved Fig. 3 . For
®example, at the famous site of Tourville Lautridou,
.1982 , near Rouen, the erosion step of the low
()P.Antoineetal.rGeomorphology332000183­207186
Fig. 2. Longitudinal profiles of the Seine valley from the upper course to the Channel deeps.
()P.Antoineetal.rGeomorphology332000183­207187
® . ® . ® . ® . ® .Fig. 3. Synthetic cross-section of the lower Seine terrace system near Rouen. 1 Holocene 2 Periglacial sands and gravels 3 Estuarine beds of Tourville 200 and 300 ky 4
® . ® . ® . ® . ® . ® .Brown leached soils 5 Calcareous tufa Saint­Pierre 6 Lower Pleistocene loess of La Londe 7 Clay with flints 8 Reuverian clay of La Londe 9 Saint­Eustache Sands
® . ® . ® . ® . ® .Pliocene 10 Lozere Sands 11 Oligocene 12 Tertiary sandstone 13 Location of the cross-section.`
( )P. Antoine et al.rGeomorphology 33 2000 183­207188
® .terrace q2 m is covered by three periglacial allu-
vial sediment bodies separated by two interglacial
estuarine silt beds, including a typical Saalian mam-
® . ®mal fauna macrofauna and rodents Lautridou,
.1982, 1985 . The upper one has been dated at 200
Ka by ESR­OSL, and the lower at 300 Ka by OSL
® .Balescu et al., 1991, 1997 . The other periglacial
gravels in this area are the Oissel Formation lying at
10 m above the bedrock of the Tourville terrace, and
then the Elbeuf Formation, 15 m higher, which is
covered by the famous loessic Saint Pierre-Forma-
® .tion Lautridou, 1985 , which contains four
palaeosols, and a tufa on the top of the Elbeuf IV
® .soil Fig. 3 .
The molluscan fauna from this tufa, as at Vernon
® .between Rouen and Paris and at La Celle-sous-
® .Moret upstream of Paris , indicates a forested inter-
glacial environment warmer than the Holocene. The
®fauna, dated at 350­400 Ka using UrTh from
.Vernon, Lecolle et al., 1990 , has also been de-´
scribed in Britain from the Hoxnian tufa sites of
®Hitchin and Icklingham Rousseau, 1987; Rousseau
.et al., 1992 . Above this Saint-Pierre-les-Elbeuf ter-
race, at 30 m O.D. near Rouen, there is an older,
® .very weathered terrace Martot that can be traced
downstream to Le Havre, but is very discontinuous
® . ®upstream Fig. 2 . Below the Tourville terrace low
.terrace , there are the two Weichselian valley-bottom
® .gravels: Rouen 1 and Rouen 2 Figs. 2 and 3 . These
gravel beds, like the Tourville Formation, continue
® .below the English Channel Fig. 2 , as far as the
® . ®large depression of the Hurd Deep Fig. 1A Alduc
et al., 1979; Auffret et al., 1980, 1982; Lericolais,
.1997 .
2.3. The Middle Pleistocene fluial system of the
(middle Seine Valley in the ``Region Mantaise'' From
)Elbeuf to Paris
Fifteen kilometres upstream of Elbeuf­Les An-
delys, characterised by a straight and narrow channel
® .without terraces Fig. 2 , there is a characteristic
system of meanders, smaller than downstream and
more stable, with many steps eroded into the chalk
bedrock. Dating from 600 to about 20 Ka there are
16 bedrock-steps, many more than downstream, per-
haps because of differences in the longitudinal gradi-
® .ent, which is lower upstream 0.2% , and in the
®pattern of the meanders stable meanders in the
middle valleyrmigrating meanders in the lower val-
.ley .
Each bedrock step is covered by the same kind of
® . ® .alluvial sequence Fig. 4 , Lecolle, 1989 :´
1. Deposition of fluvial loams and clays in a cold
®environment first phase of the cycle beginning
.after an interglacial .
2. Deposition of gravels and coarse sands in a
®periglacial environment fluvial activity with high
seasonal contrasts in water and sediment supply,
.braided river channel .
3. Deposition of sandy to silty fluvial sediments
®reduction of the lateral soliflucted sediment input
.into the valley in a drier environment .
Fig. 4. Model of morpho-sedimentary cycle of the middle Seine
® . ®area, Region Mantaise according to Lecolle, 1989 see text for´ ´
. ® . ® . ® .explanation . 1 Chalk bedrock 2 Fluvial gravels 3 Fluvial
® . ® . ® .sands 4 Fluvial loams 5 Pedogenesis brown soil .
( )P. Antoine et al.rGeomorphology 33 2000 183­207 189
4. End of the cold period, climatic improvement and
vegetation colonisation: major erosion and lateral
®incision creating a new step into the bedrock 1­2
.m maximum. .
®5. Climatic Optimum interstadial, rarely inter-
.glacial : pedogenesis and soil formation at the top
® .of the previous deposits brown soil , deposition
® .of overbank silts final phase of the cycle .
The cycle then returns to stage 1 at the beginning
of a new cold period.
Attempts to correlate the middle Seine sequence
with the lower Seine system, and especially with the
Tourville Formation, have used altimetry of the main
bedrock steps, large mammal remains and the posi-
tion of the interglacial deposits in the various se-
® .quences Lautridou et al., 1984 .
(2.4. The upper Seine Valley from Paris to Mon-
)tereau
In the city of Paris there are again several large
®meanders incised into soft Tertiary sediments mainly
.sands and clays and a stepped terrace system similar
® .to that of the lower Seine Fig. 2 . Upstream, the
system, with only few meanders is characterised by a
few terraces separated by well developed bedrock
® .steps 5 to 10 m , especially near Montereau at the
confluence with the River Yonne. It seems that the
main terraces defined at Rouen can be correlated
with those of the region of Paris and Montereau. For
example, at Paris mammal remains discovered in a
silty bed included in the alluvial formation of the
low terrace have been dated to 162"9 to 206"18
® .ka by UrTh technique Durbet et al., 1997 . These
dates are in agreement with that of the upper part of
®the Tourville Formation near Rouen Balescu et al.,
.1997 .
Finally, the upper Seine valley is also charac-
terised by well preserved Lateglacial silty sediments.
These sediments have been dated using 14
C from
many Magdalenian archaeological settlements
® .11.9­13 ky BP, Roblin-Jouve and Rodriguez, 1997 .
In this area a first incision phase is identified be-
tween the end of the Upper Pleniglacial and the
® .Lateglacial Roblin-Jouve and Rodriguez, 1997 , as
®has been demonstrated in the Somme valley Antoine,
.1997a,b and in other northern European Rivers
® .Vandenberghe et al., 1994 . As regards the Upper
Pleistocene, there is good correlation between the
two younger alluvial formations of the lower Seine
® .Rouen 1 and Rouen 2 and the Weichselian
Pleniglacial and Lateglacial gravel accumulations in
the upper Seine.
Finally, the absence of Lateglacial sediments in
the lower Seine may be related to the occurrence of
strong fluvial erosion during the Holocene, espe-
cially at the beginning of the Preboreal, the impact of
® .which was stronger downstream Antoine, 1997a .
3. The terrace system of the middle Somme Valley
3.1. General context
The Somme Valley is a small, NW­SE orientated
® .valley in northern France Fig. 1B , developed upon
homogeneous Chalk bedrock of Upper Cretaceous
age. In the whole area the main fluvial systems are
®parallel to the NW­SE structural features i.e., the
.Somme syncline, Fig. 1B . The area where the ter-
races are best developed is the middle Somme valley
® .about 70 km long . In the lower valley all the
terraces are preserved on the left bank and the
® .occurrence of faults crossing the valley SW­NE
®makes upstream correlations very difficult Figs. 1B
.and 5 .
The Somme Valley is well known for its impor-
®tant prehistoric archaeology Prestwich, 1859; Com-
mont, 1909, 1910, 1911; Breuil and Koslowsky,
1931; Bordes, 1954; Tuffreau, 1987, Tuffreau and
.Antoine, 1995 and its complex Quaternary terrace
®system Bourdier, 1969, 1984; Bourdier et al.,
1974a,b; Somme and Tuffreau, 1978; Somme et al.,´ ´
1984; Haesaerts and Dupuis, 1986; Antoine, 1989,
.1990, 1994a .
New research, based on field surveys combined
®with environmental studies palynology, palaeontol-
ogy, malacology, micromammal: Munaut, 1988, Mu-
naut and Defgne 1997; Moigne, 1989; Auguste, 1995;´
.Rousseau et al, 1992; Antoine et al., 1995 , have
allowed the proposal of a reference sequence for
Pleistocene river development in northern France
®that is summarised by Figs. 5­7 Antoine, 1990,
.1993, 1994a . The chronostratigraphical interpreta-
tion of this sequence is controlled by magneto-
()P.Antoineetal.rGeomorphology332000183­207190
® .Fig. 5. Longitudinal profile of the river Somme terraces and maximum incision Weichselian , from the beginning of the middle valley to the submerged Palaeovalley then to the
® .Greenwich Deep according to Auffret et al., 1982 for the submerged area .
( )P. Antoine et al.rGeomorphology 33 2000 183­207 191
Fig. 6. Simplified model of terrace formation for one Interglacial­Glacial cycle in the Somme valley.
® .stratigraphy Biquand, 1974; Laurent et al., 1994 ,
® .UrTh series Laurent, 1993 , ESR on fluvial Quartz
®grains Laurent, 1993; Laurent et al., 1994, see
.Appendix and Table 1 , TL and IRSL on loess
® .Balescu, 1988; Engelmann and Frechen, 1998 , large
® .mammal material Auguste, 1995 , and amino-acid
® .racemisation of molluscs Bates, 1993 .
Moreover, recent results concerning the Late-
glacial and Early Holocene evolution of the valley
provide a detailed model for the interpretation of the
®Pleistocene fluvial sequences of the terraces Antoine,
.1997a .
3.2. Alluial sequences and terraces
In the new studies on the Somme Valley, the term
terrace is used for the ``morpho-stratigraphical'' pat-
tern resulting from overlapping of one incision sur-
face in bedrock by one alluvial sequence, then by an
®unconformable loess and palaeosol succession slope
. ® .sequence Figs. 6 and 7 . The longitudinal profiles
of the bedrock surface beneath each of the alluvial
sequences are identified on the basis of their height
above the basal contact of the alluvial sequence of
the modern valley. Within the middle valley it is
now possible to describe 10 stepped alluvial forma-
tions from q5r6 m to 55 m relative height above
® .the modern valley bedrock Antoine, 1994a . The
oldest, named Grace-Autoroute Formation, was dis-^
covered and studied in 1994 during the excavations
for a motorway, and lies at a relative height of q55
® .m in the Montieres area Antoine, 1994b, 1998`
® .Fig. 7 .
The longitudinal profiles of the basal contacts of
all the alluvial formations are sub-parallel to the
modern bedrock floor beneath the modern valley
®gravels and show a general gradient of 0.54 Fig.
.5 . The succession of alluvial formations of the
Somme Valley results from the combination of two
different and asynchronous types of erosion: lateral
® .erosion linked to channel migration 0 to 500 m and
® .vertical erosion or incision 5 to 6 m , which results
® .in the separation of each alluvial formation Fig. 6 .
Studies of the Somme terrace alluvial sequences
never show the superposition of more than one cli-
()P.Antoineetal.rGeomorphology332000183­207192
® .Fig. 7. Synthetic cross-section throughout the middle Somme terrace system geometry, stratigraphy, geochronology and chronostratigraphical interpretation .
()P.Antoineetal.rGeomorphology332000183­207193
Table 1
® .ESR data from fluvial Quartz grains from some alluvial formations of the middle Somme according to Laurent, 1993, modified
Da: Annual dose, DR: Palaeodose.
® . ® . ® . ® . ® . ® . ® .SiterAlluv. Form.rRel. height above Sediment P Gy Error Gy DR Gy Error Gy Da mGyry Error m Gyry Age Ky Error Ky
the modern valley bedrock
® .Boutmy Montieres Formation, q12 m Fluvial loam 3120.0 861.0 290.0 80.029 1450.0 103.248 200 57`
.Cambron Montieres Formation, q12 m Fluvial loam 4333.0 439.0 390.0 39.513 1921.18 303.402 203 38`
® .Cagny­Epinette. Epinette Formation, q21m Fluvial loam 2250.0 250.0 270.0 30.0 912.162 128.076 296 53
® .Cagny­Garenne. Garenne Formation, q27 m Fluvial sand 2666.0 214.0 240.0 19.265 600.0 143.641 400 101
® .Abbeville­Carpentier Renancourt Formation, q40 m Fluvial sand 2500.0 326.0 230.0 29.992 383.333 28.418 600 90
® .Ferme de Grace Grace­Autoroute Formation, q55 m Fluvial sand 2312.0 231.0 370.0 36,968 456,790 64,423 810 140^ ^
( )P. Antoine et al.rGeomorphology 33 2000 183­207194
®matic succession glacial­interglacial or glacial­
.lateglacial . The same type of simple fluvial se-
quence is common in other river valleys of northern
France such as Scarpe at Biache­Saint-Vaast
® .Somme et al., 1988 or in northwestern Europe such´
®as at Maastricht­Belvedere Vandenberghe et al.,´ `
. ®1985 and the Haine Valley Haesaerts, 1984; Hae-
.saerts and Dupuis, 1986 , but it does not occur with
the same regularity in these terrace systems.
From stratigraphic, sedimentological and biocli-
matic studies, the alluvial sequences appear to repre-
sent a simplified budget of fluvial sedimentation
® .during a glacial­Interglacial cycle Figs. 6­9 :
1. Slope deposits with interstratified fluvial silts
®Early-glacial, at the bottom and at the valley
.margins near the slope; rarely preserved
2. Coarse fluvial gravels and sands: budget of the
pleniglacial sedimentation, well preserved, and
corresponding to the most important and the
thickest part of the fluvial sequence in all the
® .valley Fig. 8 no.1 .
3. Fine grained fluvial silts, humic soils, tufas:
Late-glacial to Interglacial budget, well preserved
near the slopes, rarely in the central part of the
® .valley Figs. 8 and 9, nos.2, 3, 4
3.2.1. Incision and Early-glacial sedimentation
Locally a sequence of slope and fluvial deposits
®chalk debris and unrolled flints in a calcareous silty
matrix interstratified with thin beds of fluvial cal-
.careous silts is described for the first part of the
® .cycle Garenne, Fig. 7 and provides evidence of the
®earliest sedimentation phase in the terrace Antoine,
.1990 .
Palynological data from the fluvial silts of the
Garenne profile indicate a succession of temperate
continental climatic phases which can be attributed
to an Early-glacial context in comparison to the
®Early Weichselian Munaut, 1989; Antoine, 1994b,
® . ® .Fig. 8. Example of alluvial sequence in the middle Somme valley: Cagny-Cimetiere near Amiens Garenne Formation, Fig. 7 . 1 Coarse`
® .calcareous flint gravels 2 Fine grained calcareous overbank silts.
( )P. Antoine et al.rGeomorphology 33 2000 183­207 195
Fig. 9. Upper part of the alluvial sequence at Longpres-les-Corps-`
® . ® .Saints Argoeuve Formation, Fig. 7 . 1 Coarse calcareous grav-
® . ® .els with sand lenses, 2 Bedded calcareous fluvial silts, 3
® .Grey-green organic clayey soil horizon, 4 Calcareous tufa with
molluscs.
.Antoine et al., 1994, 1999 . This interpretation is
reinforced by the large vertebrate remains obtained
®from these beds Cerus elaphus, Equus caballus
mosbachensis, Bos and Megaceros, Moigne, 1989;
.Antoine and Tuffreau, 1993 .
This observation implies that the incision of the
bedrock occured before the full-glacial climatic
phase. The same interpretation can be derived from
the multi-disciplinary study of the Saint­Sauveur
®sequence in the Etouvie Formation Antoine et al.,
.1995 . All the available data show, therefore, that
incision occured probably very quickly at the begin-
ning of a glacial stage in a transitional context, as
® .proposed by Vandenberghe 1993, 1995 .
This interpretation is also consistent with the re-
sults of the study of northwestern European river
morphology and sedimentological modifications at
the end of the Pleniglacial, which show clearly that
incision phases occur during a short period and in a
®transitional climate and environment Vandenberghe,
.1995; Vandenberghe et al., 1994 . Indeed, in the
Somme basin a strong incision into the Weichselian
® .Pleniglacial deposits gravels and loess has also
been demonstrated at the beginning of the Lateglacial,
between the end of the Pleniglacial and 12,400 BP
14
®C from the oldest peat layers infilling the chan-
.nels; Antoine, 1997a,b . This incision is coupled
with a progressive change in the channel pattern
from a braided river system to a transitional system
® .during the Blling two or three stable channels ,
then to a single large meandering channel during the
Allerd.
3.2.2. Pleniglacial sedimentation
The main unit of the alluvial formation always
comprises coarse fluvial gravels, mainly composed
of rolled flints, reworked Tertiary pebbles and chalk
®blocks, in a calcareous sandy matrix Figs. 6­8,
.no.1 . The gravels are generally crudely stratified
and locally include calcareous sand lenses and some
®large ice-rafted Tertiary sandstone blocks lag-de-
.posits . Several sections show a succession of coarse
®gravel beds, sand and gravely sand lenses 1­5 m
.wide with cross-bedded stratification indicating
abrupt changes in energy characteristics of a braided
river system with very unstable channels.
The palynology of sand and silt lenses contained
in the gravels of the Garenne Formation shows that
during the gravel deposition the environment was an
® .open landscape non-arboreal pollen )70% .The
malacology of the Argoeuves Formation also gives
®similar results T. van Kolfschoten, personal commu-
.nication . According to all the available data these
sediments can be related to a periglacial environment
and to deposition in braided channels separated by
gravely channel bars. The gravel sedimentation took
place during the full-glacial under a severe climate,
where the absence of vegetation and strong seasonal
changes caused significant hillside erosion by
gelifraction and solifluction. Regarding the duration
of the complete Pleniglacial phase, the gravels of the
lower unit represent probably the budget of the
®strongest peaks in sediment discharge typical thick-
.ness: 4 to 5 m .
( )P. Antoine et al.rGeomorphology 33 2000 183­207196
3.2.3. Lateglacial to interglacial sedimentation
The fine fluvial deposits of the upper unit corre-
spond to the final phase of alluvial sedimentation
within the terrace formation and are characterised by
® .low energy facies Figs. 8, 9 no.2 , deposited in an
® .interglacial context maximal thickness 1.5 m . In-
deed, during this phase the hillsides were stabilised
by vegetation. They represent the last phase of allu-
vial sedimentation and are contemporaneous with the
infilling of the valley and its stabilisation. This se-
quence ends with the development of immature hu-
® .mic soils and calcareous tufas Fig. 9, no.3, 4 .
The correspondence between the palynology
® .Munaut, 1988, 1989 , small mammal remains
® . ® .Cordy, 1989 and the sedimentology Antoine, 1990
demonstrates that these fine fluvial deposits accumu-
lated during a continental temperate climatic phase
®forested-steppe landscape, arboreal pollen: 60­
.80% . The same results have been obtained in the
Scarpe Valley sequence, northern France, at Biache-
® .Saint-Vaast Somme et al., 1988; Munaut, 1998 .´
Nevertheless, in the Somme Valley sequences, the
absence of ``typical full interglacial conditions''
comparable to the Holocene or to the Eemian opti-
mum, such as have been found in the peat of the Lys
® .Valley Somme et al., 1996 , is a problem. Previ-´
ously, it was thought that this observation could be
explained by assuming the occurrence of more conti-
nental conditions during interglacials in the chalk
® .valleys of NW France Munaut, 1988 . New investi-
gations on the modern valley sequence show that the
geomorphological and sedimentological character-
istics of the sediments preserved at the top of the
fluvial sequences of the terraces are locally very
similar to those of the Weichselian Lateglacial, and
especially to the Allerd and Younger Dryas over-
®bank silts Antoine, 1997b; Limondin, 1995; Munaut
.and Defgnee, 1997 . Moreover the fine-grained sedi-´
® .Fig. 10. General view of the slope sequence overlying the Grace-Autoroute Formation Fig. 7 , and of its contact with the chalk talus^
® .maximum thickness: 11 m .
( )P. Antoine et al.rGeomorphology 33 2000 183­207 197
ments of the terraces apparently always represent
overbank facies while typical channel sediments have
never been found.
From these observations it is possible to infer that
during a full interglacial, sedimentation only takes
place in a narrow channel, as it has been demon-
® .strated for the Holocene Antoine, 1997a . Then, at
the beginning of the next Early-glacial, erosion dur-
ing migration of the channel followed by incision,
has removed all the full interglacial sediments.
On the other hand, the occurrence of full inter-
glacial conditions is demonstrated from the malaco-
logical study of the tufa sequence at Arrest in a small
® .tributary of the lower Somme Rousseau et al., 1992 .
In this sequence, the malacological assemblage shows
®about 53% of forest and semi-forest species 36% in
.individuals and is comparable to the faunas of the
oxygen isotope stage 11 interglacial tufas of Nor-
®mandy, the Paris Basin and southern England Rous-
.seau et al., 1992 .
Finally, full Interglacial deposits are also known
in the lower Somme Valley at Menchecourt, where
®periglacial fluvial gravels relative height above the
.modern valley bedrock : 15 m; Antoine, 1990 are
covered by fluvio-marine beds, including marine and
® .fluvial interglacial molluscs Commont, 1910 . New
investigations are planned to determine more clearly
the relations between marine and fluvial systems
during interglacial periods in the lower part of the
Somme.
3.3. Slope deposits sequences and age-control of the
terrace system
One important characteristic of the Somme valley
is the presence of well developed loess and palaeosol
sequences, which provide good age-control for the
whole fluvial system. Generally speaking, the com-
parison of the various loess and palaeosols sequences
overlying the succession of fluvial units in the Somme
valley shows that terrace formation is characterised
® .by a cyclic pattern Antoine, 1990, 1994a,b . This
pattern is illustrated by the progressive increase of
the number of fossil soils and associated loess and
colluvial deposits, with the increasing antiquity of
® .alluvial formations Fig. 7 .
For example, the analysis of the slope sequence of
the Garenne Formation shows that this alluvial unit
is overlain by at least five climato-sedimentary cy-
cles whereas the Epinette Formation, at Mautort near
Abbeville, is covered by only four cycles. In addi-
tion, the pedosedimentary analysis of the new Grace^
® .slope sequence Antoine, 1994b; Fig. 10 , demon-
strates a minimum of eight climatic cycles overlying
the oldest alluvial formation of the middle Somme
®area dated at around 1 Ma Grace-Autoroute Forma-^
.tion, Fig. 7 . Moreover, the study of slope deposit
sequences provides correlations between the Somme
® .Valley long loess sequences Garenne and Grace^
and the reference loess sequences at Saint-Pierre-
® .les-Elbeuf in Normandy Lautridou, 1985 and at
®Ariendorf in the middle Rhine valley Brunnacker et
.al., 1975, 1982 .
Finally, the chronostratigraphical interpretation of
the system is based on the correlation between the
climatic signal deduced from the multidisciplinary
study of the Somme terrace system, and the global
® .Marine Isotope Stratigraphy Martinson et al., 1987 .
This interpretation is reinforced by the study of the
® .remains of large mammals Auguste, 1995 , the neg-
ative polarity determined for the Grace Formation^
® .Biquand, 1974; Laurent et al., 1994 , aminochronol-
® . ®ogy Bates, 1993 and ESR and UrTh dates Laurent,
.1993; Laurent et al., 1994; Table 1 .
4. Comparison between the Somme and the Seine
River responses to climatic variations, tectonic
movements and sea-level changes since about 1
My ago.
Even if internal factors can produce terraces
® .Schumm, 1977 , it is generally admitted that cli-
matic variations, tectonic movements and sea-level
changes have worked together during terrace forma-
®tion Lowe and Walker, 1984; Vandenberghe, 1995;
Veldkamp and Van Den Berg, 1993; Van Den Berg,
.1996; Maddy, 1997 . Nevertheless, these factors have
operated at different time scales and their relative
impact is variable in time and is linked to the
location of the area within the whole fluvial system.
This variability is well illustrated by the comparison
between the pattern of the different river systems, or
between the different parts of one river system, such
® .as the Seine Fig. 11 .
Finally, it is also possible that, in northwestern
France tectonic uplift has been enhanced during
glacial stages by glacio-isostatic crustal rebound,
( )P. Antoine et al.rGeomorphology 33 2000 183­207198
Fig. 11. General pattern of the bedrock steps and of the alluvial formations in the middle Somme and in the lower and middle Seine.
®with the development of a forebulge Dawson, 1992;
.Devoy, 1996 . This phenomenon was discussed by
® .Boulton 1990 , but its importance with regard to
tectonic uplift in northern France is difficult to eval-
uate, and probably not very significant compared to
® .uplift of tectonic origin Lericolais, 1997 . In the
same way, it is very difficult to differentiate the real
impact of the ``isostatic adjustment'' in response to
the effects of sediment transfer from the areas of
erosion to areas of accumulation, as proposed by
® .Bridgland 1994 , as an explanation for the incision
process.
4.1. Climatic ariations
As elsewhere in NW Europe, the climatic factor
seems to be the major control on the formation of the
®northern French river terrace systems Haesaerts,
1984; Lautridou, 1985; Lautridou et al., 1984; Sommé
.et al., 1984; Antoine, 1989, 1994a; Lecolle, 1989 .´
Within the valleys, climate is the principal control in
the middle and upper courses in which sea-level
changes do not influence the dynamics of the fluvial
® .systems Vandenberghe, 1995; Bonnet, 1998 . For
example, in the middle Somme area the evolution of
the structure and of the sedimentology of the alluvial
sequences, as well as the incision process, are clearly
®controlled by climate variations Antoine, 1990,
.1994a . In the Seine Valley the climatic impact is
® .also very clear in the middle valley Lecolle, 1989´
and in the lower valley where it is superimposed
®upon the effects of sea-level change Lautridou et al.,
.1984; Lefebvre et al., 1994 .
The fact that incision is probably a very short
event or a succession of very short events compared
with the duration of a single glacial­interglacial
cycle is one important problem for the discussion of
the relative impact of climate, tectonic and sea-level
changes on a fluvial system. Moreover, it is very
difficult to evaluate the effects of slow tectonic
movements and sea-level changes and rapid climatic
variations such as those which have been demon-
®strated in ice or deep sea records Bond and Lotti,
.1995; Stuiver et al, 1995 upon this type of record.
Indeed, investigations on the Weichselian
PleniglacialrLate-glacial transition, which has been
well dated by 14
C, have demonstrated that strong
® .incision occurs over a very short time span -1ka ,
during transitional climatic and environmental condi-
® .tions Vandenberghe, 1995; Antoine, 1997a,b . These
incisions, in the upper and middle courses, are totally
independent of sea-level variations and could occur
during a period of sea-level rise, as has been ob-
served during the Weichselian Late-glacial in the
® .Somme Basin Antoine, 1997a .
4.2. Tectonic moements
Uplift is generally and increasingly considered to
be a fundamental factor in the development of a
( )P. Antoine et al.rGeomorphology 33 2000 183­207 199
stepped terrace system as in the middle and lower
®Seine and middle Somme Rivers cf. Veldkamp and
.Van Den Berg, 1993; Van Den Berg, 1996 .
Indeed, in the Somme valley, the progressive and
discontinuous downcutting throughout the whole sys-
tem of the middle valley since the end of the Lower
Pleistocene has taken place in the context of the
general uplift of the northwestern part of the Paris
®Basin Pommerol, 1978, Colbeaux et al, 1977, 1980,
.Antone, 1994a,b . The whole middle Somme valley
is located in an area where uplift rates are uniform
® .Founiguet, 1987 . All the substantial asymmetrical
aggradations of alluvial formations, such as in
® .Amiens-Montieres SommerSelle confluence , show`
no particular orientation and their development ap-
®pears to be linked to hydrodynamic processes i.e.,
preferential accumulation of fluvial deposits in con-
.fluence zones .
In addition, despite numerous observations, no
clear Pleistocene tectonic effects have been recorded
®in the terrace sediments of the middle Somme e.g.,
.clear faults continuing into the chalk bedrock . In-
deed, all the fault systems observed until now in the
fluvial and slope sequences of this area clearly result
from dissolution features in the underlying chalk
® .bedrock Fig. 12 , and are related to weathering and
pedogenesis during interglacial and Early-glacial
® .phases Antoine, 1990
On the other hand, in the lower Somme valley the
terraces are preserved only on the left bank of the
valley. This observation could indicate a tilting of
the left bank of the Somme area and a relative lower
rate of uplift of the right bank. Nevertheless, the
profiles of the alluvial formations are very difficult
to establish in this area because of the small number
of outcrops and of the occurrence of faults crossing
® .the system N508 to N708; Fig. 1B , which could
®have been active during the Pleistocene Dupuis et
.al., 1977 .
Moreover, in the Somme valley the strong differ-
ence between the continental area, with a stepped
terrace system, and the submarine extension with
® .Fig. 12. Dissolution features in the chalk bedrock at the base of weathered coarse alluvial deposits Mautort, area of Abbeville .
( )P. Antoine et al.rGeomorphology 33 2000 183­207200
overlapping alluvial formations, probably indicates a
difference in the absolute values of uplift between
the present-day continental chalk plateau and the
® .Channel area Fig. 5 . This difference could also
explain the preservation of thick Tertiary beds in the
Channel area while they are eroded elsewhere. Ac-
® .cording to Auffret et al. 1980, 1982 , the lower part
of the Somme valley, today submerged beneath the
Channel, shows a very different pattern. Indeed, in
this area, the longitudinal profile is characterised by
® .a lower gradient 0.2 mrkm and the occurrence of a
®thick sequence of superimposed alluvial bodies 4 to
. ® .5 in the Palaeo-Somme Fig. 5 , before the diver-
® .sion Auffret et al., 1980, 1982 . In addition, the
longitudinal profile of the valley floor shows a short
®segment with a higher gradient 0.9 to 1.1 mrkm,
.Fig. 5 , between the lower part of the middle valley
® . ® .0.54mrkm , and the submarine area 0.2 mrkm .
According to these observations, it is proposed
that these important changes in the gradient and in
the fluvial pattern result from differences in the
absolute uplift rates between the continental and the
®submarine area relative subsidence in the submarine
.area, Fig. 5 .
®The short segment of the lower Somme 20 km,
.between Abbeville and the current shore , which is
® .characterised by the highest gradient 1.1 mrkm
® .and the occurrence of faults N50 to N70, Fig. 1B ,
appears therefore as a transitional area in which the
® .uplift rate is higher than upstream SE . In the same
® .way the high chalk cliffs height : 80 to 90 m that
characterise the coast at the south­west of the Somme
® .bay Fig. 1B , and which are parallel to the direction
® .of the faults N508 , could also result from a differ-
ence in uplift rates between the Channel and the
continental area, and thus have a tectonic origin
® .Fig. 5 .
Nevertheless, the lithological composition of the
bedrock could also be an important factor. Indeed,
the submarine Seine valley shows the extension of a
stepped system on hard Jurassic calcareous bedrock
® .Figs. 1B and 3 , while the submarine Somme, in-
cised in a sandy bedrock, is characterised by overlap-
® .ping alluvial formations Figs. 1B and 5 . This kind
of relation between stepped terrace systems on a
hard bedrock and stacked alluvial formations on soft
®bedrock is also described from the Oise valley Paris
. ® .Basin, Fig. 1B or in the Lys Valley northern France
®Somme, 1975, Somme et al., 1996; Colbeaux et al.,´ ´
.1979 . In the latter, the modification in the fluvial
pattern from stepped terraces to superimposed allu-
vial bodies, between the chalky ``Haut Pays'' and
the sandy ``Plaine de la Lys'', is clearly linked to
® .tectonics Colbeaux et al., 1977 . Therefore, in NW
France where the bedrock is characterised by sandy
soft Tertiary beds overlying Cretaceous Chalk, the
effect of a relative subsidence is the preservation of
soft bedrock area and the development of fluvial
patterns characterised by superimposed alluvial bod-
ies.
®In the Seine valley, the River Seine fault Faille
.de la Seine, Fig. 1B , which is a very ancient feature,
has apparently not been active during the Middle and
the Upper Pleistocene, according to the record in this
area and downstream. This contrasts with the Lower
Pleistocene, where there are several indications of
® .tectonic activity, e.g., faults at La Londe Fig. 3 .
But at the same time during the Lower Pleistocene,
in the very high terraces, which are contemporaneous
with the sediments of the graben structures, like at
La Londe, no tectonic sensitivity has been identified
because the gravels are too weathered, too discontin-
uous, and unsuited for the preservation of faults. On
the other hand, in the Seine valley, uplift is clearly,
demonstrated by the occurrence of Pliocene estuarine
sediments on the plateau, around 150 m, before the
®incision of the valley Formation of la Londe; Kuntz
.and Lautridou, 1974; Lautridou, 1985 .
Finally, the comparison between the evolution of
the Seine and the Somme shows that the absolute
value of uplift is of 55 to 60 m for 1 Ma in both
® .valleys Fig. 1B . In addition this result demonstrates
that uplift is a fundamental parameter in the terrace
formation process and that there are no strong differ-
ences in the tectonic background throughout the area.
These values are also in good accordance with those
® .given by Veldkamp and Van Den Berg 1993 for
the Meuse valley near Maastricht for the same period
® .about 70 mr1 Ma .
4.3. Sea-leel changes
In the Seine it has been suggested that sea-level
changes have influenced incision in the lower valley
® .Lefebvre et al., 1994 . Indeed, in this area it is
( )P. Antoine et al.rGeomorphology 33 2000 183­207 201
possible to describe more than one glacial­intergla-
cial cycle overlying a single bedrock step, such as at
® .Tourville Lautridou, 1985; Lautridou et al., 1984 .
This pattern indicates that strong incision in the
chalk bedrock does not always occur during each
glacial or Early-glacial phase.
The interpretation proposed by Lefebvre et al.
® .1994 is that the main bedrock steps result from
regressive erosion during the strongest sea-level falls.
It has been suggested that the main incision phases
®in this area may occur during the coldest stages OIS
.2, 6 , 12, 16 and 22 .
Nevertheless, the problem with determining the
effect of sea-level change lies with the identification
of the section of the river that has been controlled by
this factor, especially during sea-level fall. More-
over, the change in fluvial pattern in response to
sea-level fall also depends on the gradient of the
platform exposed by the regression.
In the Seine valley, the boundary between the
lower valley, controlled by sea-level changes, and
the middle valley, characterised principally by cli-
matic control is marked in the longitudinal profiles
® .gradients by a knick point Fig. 2 , and a change in
the longitudinal profile from 0.3­0.5 mrkm in the
lower valley to 0.2 mrkm in the middle and upper
® .courses Lautridou et al., 1984 .
In the Somme Valley, a clear difference is also
® .seen between the middle and lower courses Fig. 5 .
In the middle course, where the terraces are best
preserved, the slope of the longitudinal profile is of
0.5­0.54 mrkm, whilst in the lower valley, after a
very pronounced knick-point near Abbeville, the lon-
gitudinal profile of the modern valley bedrock
changes very rapidly and shows a gradient of 1.1
® .mrkm from Abbeville y12.5 m to Le Hourdel
® . ®y30ry33 m in the modern estuary Antoine,
.1990 . Then, below the Channel, the gradient is very
® .low 0.2 mrkm between the present coast and the
Greenwich deep. In this system, where the incision
in the middle valley is related to Early-glacial condi-
® .tions Antoine, 1989, 1990; Antoine et al., 1995 , a
strong lowering of sea-level is not necessary to
change the profile, which is completely emerged
® .when the sea-level is at y50 m Fig. 5 . This
lowering of the sea-level is solely a condition for the
exportation of the sediment produced upstream by
the incision.
According to the model of early and quick inci-
sion described above, the absence of a palaeovalley
®in the western part of the Channel west of the Hurd
.Deep , previously interpreted as the result of marine
® .erosion during sea-level rise Lericolais, 1997 , could
therefore be explained only by the absence of fluvial
incision. Following this idea, and on the basis of the
data from the last climatic cycle, it is suggested that
all the incision process takes place at the beginning
® .of the cycle Early-glacial when the sea-level falls
very quickly to y20r25 m minimum, as after the
® .Eemian Zagwijn, 1989; Somme et al., 1994 , then tó
®y50 to y60m at the end of the Weichselian early
.glacial . When the sea-level was around ­50r­60m,
incision ended because of the huge increase in coarse
sedimentary supply linked to the onset of Pleniglacial
® .conditions solifluction . Thus during the
Pleniglacial, when the sea-level fell up to y130m,
the continental area between the Hurd Deep and the
® .ocean Fig. 1A is an area of transport or of sedimen-
tation but not of incision. Indeed incision is com-
pleted, and during the most important part of the
Pleniglacial the area is characterised by transporta-
tion and deposition of coarse sediments. Then during
sea-level rise, at the beginning of a new interglacial
stage, these deposits are reworked and eroded by
marine erosion.
Finally, even if it is generally thought that sea-
level falls could induce incision in the lower courses
of the fluvial systems, the speed of the phenomena
® .1r5 mmryear is very low compared with the
speed of the incision induced by abrupt climate
changes. Therefore, its impact on large fluvial sys-
tems is not easy to demonstrate. In addition, in the
example of the Somme, the very low gradient of the
eastern part of the Channel could also reduce the
impact of sea-level fall. In the Somme valley, the
changes in the gradient observed in the lower valley
are therefore rather linked to differences in the uplift
rates as it was proposed previously in the discussion
about techtonic movements Section 4.2.
5. Conclusion
According to the evidence for evolution of the
Seine and Somme river valleys during the last 1 Ma,
( )P. Antoine et al.rGeomorphology 33 2000 183­207202
it appears that climatic variations, tectonic move-
ments and sea-level changes have worked together at
different timescales and that their relative impact is
variable in the different areas of the fluvial systems.
Nevertheless, combining the data for a long river
system such as the Seine, it is possible to determine
the principal controls acting on each of the different
parts of the fluvial system. For example, in the Seine
Valley, a low rate of tectonic uplift during the Pleis-
tocene is shown in the whole course by the occur-
rence of Pliocene marine and estuarine sediments on
the plateau before the incision of the valley.
According to the data from the Seine and Somme
valleys, climatic and tectonic factors are strongly
linked in the terrace formation processes but work at
®very different rates. The slow tectonic uplift 0.05 to
.0.06 mmryear is fundamental to create the potential
for incision which is accumulated throughout a cli-
matic cycle. This potential is then quickly released
during the incision, the speed of which is about 50 to
100 times faster than the uplift. This short incision
process takes place at the beginning of the climatic
® .cycle, in a transitional period Early-glacial , charac-
terised by strong seasonal increases in water flow,
®and by a still well developed vegetation limitation
.of the colluvial imput into the valley .
In the lower Seine the main factor seems to be
sea-level variation and especially sea-level falls dur-
ing the coldest glacial periods. On the other hand,
the pattern of the middle and upper Seine valley
fluvial systems is mainly controlled by climatic vari-
ations superimposed on an uplifting background.
In the Somme valley, the middle and upper courses
are located in the same uplifting area as the Seine.
The fluvial pattern of all this area seems to be under
climatic control, as is shown by the study of the
alluvial sequences.
The very strong changes in the gradient of the
bedrock-profiles, observed at the junction between
the middle and lower Somme, then between the
lower Somme and the submarine area, are not linked
to the influence of sea-level variations because of the
very low gradient of the submarine profile which is
not very sensitive to sea-level fall.
On the other hand, according to the strong differ-
ences that appear in the gradient and in the pattern of
the fluvial system between the present day continen-
tal and submarine area, it is proposed that these two
parts of the system are controlled by different rates
of uplift:
-relatively high uplift rates in the middle and
lower valley characterised by a stepped terrace
system and high gradients
® .-lower uplift rate relative subsidence in the sub-
marine area showing stacked alluvial sequences
and a lower gradient.
Finally, in the same general climatic and tectonic
background, the difference between the Seine and
the Somme responses during the Pleistocene seems
to be mainly related to:
-differences in the influence of sea-level changes,
partly linked to the geometry of the submarine
®area gradient of the structural platform, composi-
.tion of the bedrock, tectonic evolution and,
-differences in the general pattern of the valley
linked to the size of the fluvial basins, and to
®hydrodynamic parameters size of the meanders,
.sinuosity
The middle Somme valley terrace system repre-
sents therefore an intermediate pattern stage between
the Middle and the Lower Seine valleys.
It is concluded that the study of the terrace sys-
tems appears to be a very good tool for the measure-
ment of general tectonic movements during the Pleis-
tocene.
Acknowledgements
This work is partly supported by the ``Palaeolithic
occupations and Pleistocene Palaeoenvironments in
the Somme Basin and surrounding areas'' program
®funded by the French National Centre for the Scien-
.tific Research .
We thank B. Morin and J.M. Dolo for their help
in ESR measurements, and Phil Gibbard and David
Bridgland for the review of the English text and their
constructive suggestions.
( )P. Antoine et al.rGeomorphology 33 2000 183­207 203
Appendix A. ESR dating of fluvial quartz grains
from the Somme Valley alluvial formations
A.1. Introduction
® .Electron Spin Resonance ESR dating of sedi-
ments may improve our knowledge of the chronos-
tratigraphy of the Tertiary and Quaternary periods
and especially in the case of non-calcareous deposits.
®Quartz extracted from volcanic materials Shimokawa
.et al., 1984 and sediments baked by lava-flows
® . ®Yokoyama et al., 1986 or burned by fire Monnier
.et al., 1994 have been dated by ESR. Bleaching of
quartz has been studied in Quaternary sediments
®Yokoyama et al., 1985; Buhay et al., 1988; Li et al.,
1993; Brumby and Yoshida, 1994; Laurent et al.,
.1994 , and recently, ESR results have been obtained
on bleached quartz extracted from Tertiary forma-
® .tions Laurent et al., 1998 .
ESR dating of bleached quartz is based on the
® .behaviour of the aluminium centre Al : it is com-
posed of an aluminium atom substituted to a silicium
® .centre Weil, 1984 . The diamagnetic centre
w® q.xAlO rM becomes paramagnetic under ionising4
w® .x ®radiation AlO rh see details and other references4
.in Ikeya, 1993 . Exposure to light releases trapped
electrons which will be collected by an aluminium
hole centre. This process will induce the bleaching
phenomena. However, the energy light scale allows
only a part of trapped electrons, which means that
the bleaching is incomplete.
In order to date quartz, it is important to deter-
mine residual intensity after light bleaching. Modern
coast or overbank sands have a constant ESR inten-
sity after a very long exposure under UV-light; fossil
sands and irradiated modern sands show decreasing
intensity under UV-light. The latter tend to reach
their natural intensity. The Al centre in quartz reaches
a residual level which corresponds to the maximum
of bleaching. Dating of quartz extracted from sedi-
ment requires samples that were exposed to sunlight
for long times in the range of 6 months.
For fossil quartz, the maximal bleaching intensity
is calculated by a least square method fitted by an
®exponential decay added to a constant value Walther
.and Zilles, 1994 :
f x saeyb x
qc® .
The palaeodose, P, is determined by fitting the
dose response curve with a simple saturating func-
tion and extrapolation to the residual intensity.
Appendix B. Sampling and method
Each sample consists of about 400 g of sediment
enclosed in an opaque bag. The water content of
each sample was determined by drying the sediment
at 408C.
Quartz was extracted according to the usual chem-
ical methods already described by Yokoyama et al.
® .1985 . Gamma-ray irradiations are made with a
® .cobalt panoramic source Dolo et al., 1996 ; ten
aliquots were irradiated with the same dose flow.
The range of 10 doses were chosen between 1000 G
and 15,000 Gy. Palaeodose measurements were cal-
culated using the least squares method equation and
® .error measurement Yokoyama et al., 1985 . The
®quartz was exposed to UV-light UV lamp with a
.wavelength in the range 365­400 nm . The U, Ra,
Rn, Th and K activities of the sediments were deter-
mined using gamma-ray spectrometry. A k-value of
0.15 " 0.10 was assumed, alpha and beta attenua-
tions in quartz were estimated from the calculations
® . ® .of Mejdahi 1979 and Bell 1980 . Cosmic doses
were then calculated using the formula of Yokoyama
® .et al. 1981 .
Measurement errors are due to gamma dosimetry,
intensity height determination of the Al centre, ra-
dioelement content counting and moisture content.
The analytical results are given in Table 1.
References
Alduc, D., Auffret, J.P., Carpentier, G., Lautridou, J.P., Lefebvre,
D., Porcher, M., 1979. Nouvelles donnees sur le Pleistocené ´ `
de la basse vallee de la Seine et son prolongement sous-marin´
en Manche orientale. Bull. Ass. Geol. Bass. Par. 16, 27­34.
Antoine, P., 1989. Les terrasses quaternaires du bassin de la
Somme: Etude geologique et geomorphologique, contribution´ ´
a la connaissance du Paleoenvironnement des gisements` ´
paleolithiques, These de Doctorat, Univ. des Sciences et Tech-´ `
nologies de Lille, 437 pp.
Antoine, P., 1990. Chronostratigraphie et Environnement du
Paleolithique du bassin de la Somme. Publ. du Centre d'Etudes´
® .et de Recherches Prehistoriques CERP 2, 231 pp.´
( )P. Antoine et al.rGeomorphology 33 2000 183­207204
Antoine, P., 1993. Le systeme de terrasses du Bassin de la`
Somme: modele d'evolution morphosedimentaire cyclique et` ´ ´
cadre paleoenvironnemental pour le Paleolithique. Quaternairé ´
® .4 1 , 3­16.
®Antoine, P., 1994a. The Somme Valley terrace system Northern
.France ; a model of river response to quaternary climatic
variations since 800,000 BP. Terra-Nova 6, 453­464.
Antoine, P., 1994b. Rapport sur l'operation de suivi de travaux Á
® .16-Montieres Somme . Rapport d'etude CNRSrAFANr` ´
Service Regional de l'Archeologie de Picardie, 44 pp.´ ´
Antoine, P., 1997a. Modifications des systemes fluviatiles a la` `
transition Pleniglaciaire­Tardiglaciaire et a l'Holocene: l'ex-´ ` `
® .emple du bassin de la Somme Nord de la France . Geog. Phy.
® .Quat. 51 1 , 93­106.
Antoine, P., 1997b. Evolution tardiglaciaire et debut Holocene des´ `
vallees de la France septentrionale: nouveaux resultats. Compté ´
Rendus de L'Academie des Sciences, Sciences de la Terre et´
des Planetes Paris 325, 35­42.`
® .Antoine, P. Ed. , 1998 Le Quaternaire de la vallee de la Sommé
et du littoral picard, Livret-guide de l'Excursion de l'Associa-
´tion Franc¸aise pour l'Etude du Quaternaire dans le bassin de la
Somme, 162 pp.
Antoine, P., Munaut, A.V., Somme, J., 1994. Reponse des envi-´ ´
ronnements a l'evolution climatique du Debut Glaciaire we-` ´ ´
ichselien: donnees de la France du Nord Ouest. Quaternaire 5,´ ´
151­156.
Antoine, P., Munaut, A.V., Van Kolfshoten, Th., Limondin, N.,
1995. Une occupation du Paleolithique moyen en milieu fluvi-´
atile dans la sequence de la tres basse terrasse de la Somme á ` `
® . ® .Saint Sauveur Somme . Bull. Soc. Prehist. Fr., T. 92 2 ,´
201­212.
Antoine, P., Lautridou, J.P., Somme, J., Auguste, P., Auffret, J.P.,´
Baize, S., Clet-Pellerin, M., Coutard, J.P., Dewolf, Y., Dugue,´
O., Joly, F., Laignel, B., Laurent, M., Lavolle, M., Lebret, P.,´
Lecolle, F., Lefebvre, D., Limonini Lozouet, N., Munaut,´
A.V., Ozouf, J.C., Quesnel, F., Rousseau, D.D., 1998. Le
quaternaire de la France du Nord-Quest: limites et correla-´
tions. Quaternaire 9, 227­241.
Antoine, P., Rousseau, D.D., Lautridou, J.P., Hatte, C., 1999. In:´
The last interglacial­glacial cycle in loess­palaeosol succes-
sions of northwestern France Boreas, 28 551­563.
Antoine, P., Tuffreau, A., 1993. Contexte stratigraphique, clima-
tique et paleotopographique des occupations acheuleennes dé ´
la moyenne terrasse de la Somme. Bull. Soc. Prehist. Fr., T. 4,´
243­250.
Auffret, J.P., Alduc, D., Larsonneur, C., Smith, A.J., 1980. Car-
tographie du paleoreseau des paleovallees et de l'epaisseur des´ ´ ´ ´ ´
formations superficielles meubles de la Manche orientale.
Ann. Inst. Oceanogr., Paris 56, 21­35.
Auffret, J.P., Horn, R., Larsonneur, C., Curry, D., Smith, A.J.,
1982. La Manche orientale, carte des paleovallees et des bancs´ ´
sableux. BRGM edit.
Auguste, P., 1995. Revision preliminaire des grands mammiferes´ ´ `
des gisements du Paleolithique inferieur et moyen de la valleé ´ ´
® .de la Somme. Bull. Soc. Prehist. Fr. 92 2 , 143­154.
Balescu, S., 1988. Apports de la thermoluminescence a la strati-`
graphie et a la sedimentologie des loess du Nord-Ouest de` ´
l'Europe. These Doct. Scien. Univ. Bruxelles, Vol I: 199 pp.,`
Vol II: annexes, 144 pp.
Balescu, S., Lamothe, M., Lautridou, J.P., 1997. Luminescence
evidence for two Middle Pleistocene interglacial events at
® .Tourville, north-western France. Boreas 26 1 , 61­72.
Balescu, S., Packman, S.C., Wintle, A.G., 1991. Chronological
separation of interglacial raised beaches from northwestern
Europe using thermoluminescence. Quat. Res. 35, 91­102.
Bates, M., 1993. Quaternary aminostratigraphy in Northwestern
France. Quat. Sci. Rev. 12, 793­809.
Bell, W.T., 1980. Alpha dose attenuation in quartz grains for
thermoluminescence dating. Ancient TL 12, 4­8.
Biquand, D., 1974. Position chronologique de la tres haute nappe`
alluviale de Grace vallee de la Somme par rapport a la limite^ ´ `
paleomagnetique Bruhnes-Matuyama. Bull. Ass. Fr. Et. Quat.´ ´
11, 157­159.
Bond, G., Lotti, R., 1995. Iceberg discharge into the North
Atlantic in millenial time scales during the last glaciation.
Science 267, 1005­1010.
Bonnet, S., 1998. Tectonique et dynamique du relief: le socle
armoricain au Pleistocene. Mem. Geosci. Rennes 86, 352 pp.´ `
Bordes, F., 1954. Les limons quaternaires du bassin de la Seine.
Arch. Inst. Paleontol. Hum., Mem. 26, 472 pp.
Boulton, G.S., 1990. Sedimentation and sea-level change during
glacial cycles and their control on glacimarine facies architec-
® .ture. In: J.A., Scourse, J.D. Eds. , Glacimarine Environments:
Processes and Sediments. Dowdeswell, Geo. Soc. London,
Spec. Publ. 53, 15­52.
Bourdier, F., 1969. Excursion dans le bassin de Paris de l'Associa-
tion Internationale pour l'Etude du Quaternaire du 18 au 28
aout 1969: etude comparee des depots quaternaires des bassins^ ´ ´ ^
de la Seine et de la Somme. Bull. Inf. Geol. Bassin Paris 21,
169­220.
Bourdier, F., 1984. Essai d'historique sur la notion de nappes
alluviales periglaciaires. Bull. Ass. Fr. Et. Quat. 21, 5­7.´
Bourdier, F., Chaline, J., Munaut, A.V., Puissegur, J.J., 1974a. La
tres haute nappe alluviale de la Somme. Bull. Ass. Fr. Et.`
Quat. 11, 137­143.
Bourdier, F., Chaline, J., Munaut, A.V., Puissegur, J.J., 1974b. Le
Complexe mindelien: II-La moyenne terrasse de l'Avre Bull.
Ass. Fr. Et. Quat. 11, 168­180.
Breuil, H., Koslovsky, L., 1931. Etudes de stratigraphie
paleolithique dans le Nord de la France, la Belgique et´
l'Angleterre. Anthropologie 41, 249­290.
BRGM, 1996. Carte Geologique de la France a l'echelle dú ` ´
millionieme.. Bureau de Recherches Geologiques et Minieres ,` ´ `
Orleans.´
Bridgland, D.R., 1994. Quaternary of the Thames, Geological
Conservation Review Series. Chapman & Hall, London, 441
pp.
Brumby, S., Yishida, H., 1994. An investigation of the effect of
sunlight on the ESR spectra of quartz centers: implications for
dating. Quat. Sci. Rev. 13, 615­618.
Brunnacker, K., Lohr, H., Boenick, W., Puissegur, J.J., Poplin, F.,
1975. Quartar-Aufschlusse bei Ariendorf am Unteren Mittel-¨ ¨
rhein. Mainzer Naturwiss. Archi. 14, 93­141.
Brunnacker, K., Loscher, M., Tillmans, W., Urban, B., 1982.¨
( )P. Antoine et al.rGeomorphology 33 2000 183­207 205
Correlation of the Quaternary terrace sequence in the Lower
Rhine valley and Northern Alpine foothills of central Europe.
Quat. Res. 18, 152­173.
Buhay, W.M., Schwarczs, H.P., Grun, R., 1988. ESR dating of¨
fault-gouge: the effect of grain size. Quat. Sci. Rev. 7, 515­
522.
Clet-Pellerin, M., 1983. Le Plio-Pleistocene en Normandie. Ap-´ `
ports de la Palynologie. These 3 cycle. Univ. Caen, 135 pp.`
Colbeaux, J.P., Beugnies, A., Dupuis, Ch., Robasynski, F.,
Somme, J., 1977. Tectonique de Blocs dans le Sud de lá
Belgfique et le Nord de la France. Ann. Soc. Geol. Nord. 97,
191­222.
Colbeaux, J.P., Dupuis, Ch., Robasynski, F., Auffret, J.P., Hae-
saerts, P., Somme, J., 1980. Le Detroit du Pas-de-Calais: un´ ´
element dans la tectonique de blocs de l'Europe Nord-Occi-´ ´
dentale. Bull. Inf. Geol. Bassin Paris 17, 41­54.
Colbeaux, J.P., Leplat, J., Paepe, R., Somme, J., 1978. Tectoniqué
recente dans le Nord de la France et le Sud de la Belgique:´
®exemple de la Plaine de la Lys Feuille Hazebrouck, 1r50
.000 . Ann. Soc. Geol. Nord. 98, 179­188.
Colbeaux, J.P., Somme, J., Tuffreau, A., 1979. Tectonique quater-´
naire dans le Nord de la France: l'apport du gissement de
Biache-Saint-Vaast. Bull. Ass. Fr. Et. Quat. 2s 7­8, 183­192.
Commont, V., 1909. Saint-Acheul et Montieres. Notes de Geolo-` ´
gie, de Paleontologie et de Prehistoire. Mem. Soc. Geol. Nord.´ ´
T6, 3, 68 pp.
Commont, V., 1910. Excursion de la Societe geologique du Nord´ ´ ´
et de la Faculte des Sciences de Lille a Abbeville, le 11 juin´ `
1910. Les gisements paleolithiques d'Abbeville. Stratigraphie,´
faune, industrie humaine. Situation par rapport aux terrasses
fluviatiles de la Somme. Ann. Soc. Geol. Nord. 39, 249­293.´
Commont, V., 1911. Les gisements prehistoriques de Saint-Acheul´
et de Montieres. Notes de Prehistoire publiees dans le Bulletin` ´ ´
de la Societe Lineenne du Nord de la France de 1905 a 1910.´ ´ ´ `
Amiens, 189 pp.
Cordy, J.M., 1989. Les micromammiferes de Cagny-l'Epinette`
® .rapport preliminaire . In: Livret-guide de l'excursion dans lá
vallee de la Somme. Colloque L'Acheuleen dans l'Ouest dé ´
l'Europe, Saint-Riquier, juin 89, 80­81.
Dawson, A.G., 1992. Ice Age Earth: Late Quaternary Geology
and Climate. Routledge, London.
Devoy, R., 1996. Deglaciation, Earth crustal behaviour and sea-
level changes in the determination of insularity: a perpective
® .from Ireland. In: Preece, R.C. Ed. , Island Britain: a Quater-
nary Perspective. Geo. Soc. Spe. Publ. 96, 181­208 pp.
Dolo, J.M., Lecerf, N., Mihajlovic, V., Falgueres, C., 1996.
Contribution of ESR dosimetry for irradiation of geological
and archaeological samples with a 60Co panoramic source.
Applied Radiation and Isotopes 47, 1419.
Dupuis, Ch., Haesaerts, P., Heinzelin, J., 1977. Mise en evidencé
de depots a caractere fluvio-marin parmi les nappes alluviales´ ^ ` `
de la Basse Somme. Bull. Ass. Fr. Et. Quat. 14, 86­95.
Durbet, G., Rodriguez, P., Badalian, L., Djillali, H., Gautier, A.,
Laurent, M., Ricard, J.L., Watez, J., 1997. Decouverte d'un´
site paleolithique moyen dans les alluvions saaliennes dú
® .confluent Seine-Marne a Maison-Alfort Val de Marne . C. R.`
® .Acad. Sci., Paris 324, serie IIa, 505­512.´
Engelmann, A., Frechen, M., 1998. Datations TLrIRSL. In:
® .Antoine, P. Ed. , Le Quaternaire de la vallee de la Somme et´
du littoral picard, Livret-guide de l'Excursion de l'Association
´Franc¸aise pour l'Etude du Quaternaire dans le bassin de la
Somme. 41­42.
Founiguet, J., 1987. Geodynamique actuelle dans le Nord et lé
Nord-Est de la France. Apport des comparaisons de nivelle-
ments. Mem. BRGM 127, 173.
Gibbard, P.L., 1988. The History of the great northwest European
rivers during the past three million years. Philo. Trans. R. Soc.
London 318, 559­602.
Haesaerts, P., 1984. Les formations fluviatiles pleistocenes du
Bassin de la Haine Belgique. Bull. Ass. Fr. Et. Quat. 21,
19­26.
Haesaerts, P., Dupuis, Ch., 1986. Contribution a la stratigraphie`
des nappes alluviales de la Somme et de l'Avre dans la region´
d'Amiens. Bull. Ass. Fr. Et. Quat. 26, 171­186.
Ikeya, M., 1993. New Applications of Electron Spin Resonance:
Dating, Dosimetry and Microscopy. World Scientific Publish-
ing, 500.
Kuntz, G., Lautridou, J.P, 1974. Contribution a l'etude du Pliocene` ´ `
et du passage Pliocene­Quaternaire dans les depots de la Foret` ´ ^ ^
de la Londe, pres de Rouen Correlations possibles avec divers` ´
gisements de Haute-Normandie. Bull. Ass. Fr. Et. Quat. 3­4,
117­128.
Laurent, M., 1993. Datation par resonance de spin electroniqué ´
® .ESR de quartz de formations quaternaires: comparaison avec
le paleomagnetisme. These Doct. Museum Nat. Hist. Nat.,´ ´ `
Paris, 103 p., 39 figures, 2 tables.
Laurent, M., Falgueres, C., Bahain, J.J., Yokoyama, Y., 1994.
Geochronologie du systeme de terrasses quaternaires fluvi-´ `
atiles du bassin de la Somme. Comptes Rendus de l'Ac. des
Sc. Paris, Ser. IIa, t. 318, 521­526.´
Laurent, M., Falgueres, C., Bahain, J.J., Rousseau, L., vliet-Lanoe,
B., 1998. ESR dating of quartz extracted from Quaternary and
Neogene sediments: method, potential and actual limits. Quat.
Geochronol. 17, 1057­1062.
® .Lautridou, J.P., Ed. , 1982. The Quaternary of Normandy. Bull.
Centre de Geom. , CNRS, Caen., 26, 88 pp.´
Lautridou, J.P., 1985. Le cycle periglaciaire pleistocene en Europé ´ `
du Nord-Quest et plus particulierement en Normandie. These` `
Lettres, Univ. Caen, 2 vol, 908 pp.
Lautridou, J.P., Lefebvre, D., Lecolle, F., Carpentier, G., De-´
scombes, J.C., Gaquerel, C., Hault, M.F., 1984. Les Terrasses
de la Seine dans le meandere d'Elbeuf, correlatrions avec´ ´
celles de la region de mantes. Bull. Ass. Fr. Et. Quatern. 1­3,´
27­32.
Lautridou, J.P., Auffret, J.P., Lecolle, F., Lefebvre, D., Lericolais,´
G., Roblin-Jouve, A., Balescu, S., Carpentier, G., Cordy, J.M.,
Descombes, J.C., Occhietti, S., Rousseau, D.D., 1999. Le
fleuve Seine Le fleuve Manche. Bull. Soc. Geol. Fr. 170,
545­558.
Lecolle, F., 1989. Le cours moyen de la Seine au Pleistocené ´ `
moyen et superieur. Geologie et Prehistoire. These d'Etat,´ ´ ´ `
® .Univ. Paris VI 1987 . Groupe Seine, Lab. Geol. Univ. Rouen,´
549 pp.
Lecolle, F., Rousseau, D.D., Lautridou, J.P., Puissegur, J.J., Oc-´ ´
( )P. Antoine et al.rGeomorphology 33 2000 183­207206
chietti, S., Quinif, Y., 1990. Le tuf de Vernon: nouvelles
®donnees stratigraphie, paleoclimatologie, datations, correla-´ ´ ´
.tions . Bull. Cent. Geomorphol CNRS Caen 38, 131­150.
Lefebvre, D., Antoine, P., Auffret, J.P., Lautridou, J.P., Lecolle,´
F., 1994. Rythme de reponse des environnements fluviatiles´
aux changements climatiques en France du Nord-Ouest. Qua-
® .ternaire 5 , 3­4, 165­172.
´Lericolais, G., 1997. Evolution Plio-quaternaire du Fleuve Manche:
Stratigraphie et Geomorphologie d'une plateforme continen-´
tale en regime periglaciaire. These Doct. Geol. marine, Univ.´ ´ ` ´
Bordeaux I, 265 pp.
Li, D., Gao, H., Ding, Z., Zhao, X., Yang, B., 1993. On the
estimation of depositional ages of loess by ESR. Appl. Radiat.
Isot. 44, 203­206.
Limondin, N., 1995. Late-Glacial and Holocene malacofaunas
from archaeological sites in the Somme Valley, North France.
J. Archaeol. Sci. 22, 683­698.
Lowe, J.J., Walker, M.J.C., 1984. Reconstructing Quaternary
Environments. Longman, New York, 389 pp.
Maddy, D., 1997. Uplift-driven valley incision and river terrace
formation in Southern England. J. Quat. Sci. 12, 539­545.
Martinson, D.G., Pisia, N.G., Hays, J.D., Imbrie, J., Moore, T.C.,
Shackleton, N.J., 1987. Age dating and the orbital theory of
the Ice ages: development of a high-resolution 0 to 300,000-
year chronostratigraphy. Quat. Res. 27, 1­29.
Mejdahj, V., 1979. Thermoluminescence dating: beta-dose attenu-
ation in quartz grains. Archaeometry 21, 61­72.
Moigne, A.M., 1989. Cagny-l'Epinette: la grande faune. In:
Livret-guide de l'excursion dans la vallee de la Somme,´
Colloque L'Acheuleen dans l'Ouest de l'Europe. Saint-Riquier,´
82­87.
Monnier, J.l., Hallegouet, B., Hinguant, S., Laurent, M., Auguste,
P., Bahain, J.J., Gebhardt, A., Falgueres, C., Marguerie, D.,
Molines, N., Morzadec, H., Yokoyama, Y., 1994. A new
® .regional group of the Lower Palaeolithic in Brittany France ,
recently dated by Electron Spin Resonance. Compte Rendus
® .Acad. Sci., Paris, S IIa 319 , 155­160.
Munaut, A.V., 1988. L'environnement vegetal de quelques depots´ ´ ´ ^
quaternaires du bassin de la Somme. Cultures et industries
paleolithiques en milieu loessique, Amiens, dec. 86. Rev.´ ´
Archeol. Picardie., 1­2, 45­56.´
Munaut, A.V., 1989. Cagny-l'Epinette: analyses palynologiques.
In: Livret-guide de l'excursion dans la vallee de la Somme.´
Colloque L'Acheuleen dans l'Ouest de l'Europe. Saint-Riquier,´
juin 1989, 75­79.
Munaut, A.V., 1998. Synthese de l'evolution de la vegetation au` ´ ´ ´
® .Tardiglaciaire et a l'Holocene. In: Antoine, P. Ed. , Le Qua-` `
ternaire de la vallee de la Somme et du littoral picard,´
Livret-guide de l'Excursion de l'Association Franc¸aise pour
´l'Etude du Quaternaire dans le bassin de la Somme, 84­87.
Munaut, A.V., Defgnee, A., 1997. Biostratigraphie et environ-´
nement vegetal des industries du Tardiglaciaire et du debut dé ´ ´
l'Holocene dans le bassin de la Somme. CTHS Edit., Le`
Tardiglaciaire en Europe du Nord-Ouest, J.P. Fagnart et A.
Thevenin dir., 27­37.´
Pomerol, C., 1978. Evolution paleogeographique et structurale dú ´
Bassin de Paris, du Precambrien a l'actuel, en relation avec les´ `
® .regions avoisinantes. Geol. Mijnbouw 57 4 , 533­543.´
Prestwich, J., 1859. On the occurence of flint implements associ-
ated with remains of animals of extinct species in beds of a
late geological period in France at Amiens and Abbevillle and
in England at Hoxne. Philos. Trans. R. Soc. London 150,
277­327.
Roblin-Jouve, A. , Rodriguez, P., 1997. Paleogeographie des´ ´
occupatuions humaines du centre du Bassin parisien a partir`
du Tardiglaciaire. In CTHS Edit., Le Tardiglaciaire en Europe
du Nord-Ouest, CTHS Ed., J.P. Fagnart et A. Thevenin dir.,´
141­150.
Rousseau, D.D., 1987. Les associations malacologiques des tufs
holsteiniens de la France septentrionale Une application du
concept de biome. Bull. Cent. Geomor. CNRS Caen 37, 9­18.
Rousseau, D.D., Puissegur, J.J., Lecolle, F., 1992. West-European´ ´
®terrestrial molluscs assemblages of isotopic stage 11 Middle
.Pleistocene : climatic implications. Palaeogeogr., Palaeoclima-
tol., Palaeoecol. 92, 15­19.
Schumm, S.A., 1977. The Fluvial System. Wiley, New York, 338
pp.
Shimokawa, K., Imai, N., Hirota, M., 1984. Dating of a volcanic
rock by ESR. Isot. Geosci. 2, 365­373.
Somme, J., 1975. Les plaines du Nord de la France et leur´
bordure. Etude geomorphologique. These Doctorat Etat, Paris´ `
I, 810 pp, 185 fig.
Somme, J., Tuffreau, A., 1978. Historique des recherches sur lé
Quaternaire de la region du Nord de la France. Bull. Ass. Fr.´
Et. Quat. 15, 5­13.
Somme, J., Fagnart, J.P., Leger, M., Munaut, A.V., Puissegur,´ ´
J.J., Tuffreau, A., 1984. Terrasses fluviatiles du Pleistocené `
moyen en France septentrionale: signification dynamique et
climatique. Bull. Ass. Fr. Et. Quat. 21, 52­58.
Somme, J., Tuffreau, A., Aitken, M. J., Auguste, P., Chaline, J.,´
Colbeaux, J.P., Cunat-Boge, N., Geeraerts, R., Hus, J.,´
Huxtable, J., Juvigne, E., Munaut, A.V., Occhietti, S., Pichet,´
P., Puissegur, J.J., Rousseau, D.D., Van Vliet-Lanee, B., 1988.´ ¨
Chronostratigraphie, climats et environnements. In: Le Gise-
®ment Paleolithique moyen de Biache-Saint-Vaast Pas dé
. ® .Calais vol I. Tuffreau, A., Somme, J. Eds. , Mem. Soc.´
Prehist. Fr. Vol. 21 pp. 115­119.
Somme, J., Munaut, A.V., Etmontspohl, A.F., Limondin, N.,´
Lefebvre, D., Cunat-Boge, N., Mouthon, J., Gilot, E., 1994.´
The Watten boring: an Early Weichselian and Holocene cli-
matic and paleoecological record from the French North-Sea
coastal plain. Boreas 23, 231­243.
Somme, J., Munaut, A.V., Puissegur, J.J., Cunat-Boge, N., Hey-´ ´ ´
vaert, F., Leplat, J., 1996. L'Eemien sous les formations
®weichseliennes et holocenes du sondage d'Erquinghem Nord`
. ® .de la France dans la plaine de la Lys. Quaternaire 7 1 ,
15­28.
Stuiver, M., Grootes, P.M., Brazunias, T.F., 1995. The GISP 2
18
O Climate Record of the Past 16 500 Years and the Role of
the Sun, Ocean, and Volcanoes. Quat. Res. 44, 341­354.
Tourenq, J., Pomerol, C., 1995. Mise en evidence par la presencé ´
d'augite du Massif Central, de l'existence d'une pre Loire -´
pre Seine coulant vers la Manche au Pleistocene. C. R. Acad.´ ´ `
® . ® .Sci., Paris 320 IIa , 1163­1169.
Tuffreau, A., 1987. Le Paleolithique inferieur et moyen du Nord´ ´
® .de la France Nord, Pas-de-Calais, Picardie dans son cadre
( )P. Antoine et al.rGeomorphology 33 2000 183­207 207
stratigraphique. These de doctorat d'Etat, Univ. Sciences et`
Techniques de Lille Flandres-Artois, 609 pp.
Tuffreau, A., Antoine, P., 1995. The Earliest occupation of Eu-
rope: Continental Northwestern Europe. In: Roebroeks, W.,
® .VanKolfshoten, T. Eds. , The Earliest occupation of pleis-
tocene Europe. University of Leiden, Leiden, 147­163.
Van Den Berg, M.W., 1996. Fluvial sequences of the Maas : a
10ma record of neotectonics and climatic changes. Thesis
University of Wageningen, 181 pp.
Vandenberghe, J., 1993. Changing fluvial processes under chang-
ing periglacial conditions. Z. Geomorphol N.F. 88, 17­28.
Vandenberghe, J., 1995. Timescales, climate and river develop-
ment. Quat. Sci. Rev. 14, 631­638.
Vandenberghe, J., Kasse, C., Bohnke, S., Kozarsky, S., 1994.
Climate-related river activity at the Weichselian-Holocene
transition: a comparative study of the Warta and Maas rivers.
Terra Nova 6, 476­485.
Vandenberghe, J., Mucher, H.J., Roebroeks, W., Gemke, D.,
1985. Lithostratigraphy and palaeoenvironment of the Pleis-
tocene deposits at Maastricht-Belvedere, Southern Limburg,´ `
the Netherlands. Analecta Praehist. Leidensia 18, 7­18.
Veldkamp, A., Van Den Berg, M.W., 1993. Three-dimentional
modelling of Quaternary fluvial dynamics in a climo-tectonic
®dependant system. A case study of the Maas record Maastricht,
.The Netherlands . Global Planet. Change 8, 203-218.
Walther, R., Zilles, D., 1994. ESR studies on bleached sedimen-
tary quartz. Quat. Geochronol. 13, 611­614.
Weil, J.A., 1984. A review of electron spin spectroscopy and its
application to the study of paramagnetic defects in crystalline
quartz. Phys. Chem. Miner. 10, 149­165.
Yokoyama, Y., Falgueres, C., Quaegebeur, J.P., 1985. ESR dating
of quartz from Quaternary sediments: first attempt. Nucl.
® .Tracks 10 4­6 , 921­928.
Yokoyama, Y., Falgueres, C., Quaegebeur, J.P., 1986. ESR dating
of sediment baked by lava-flows: comparison of paleodoses
® .for Al and Ti centers. In: Ikeya, Miki Ed. , ESR dating and
dosimetry. IONICS, Tokyo, 197­204.
Yokoyama, Y., Quaegebeur, J.P., Bibron, R., Leger, C., Nguyen,´
® .H.V., Popeau, G., 1981. Electron spin resonance ESR dating
of bones of the Caune de l'Arago at Tautavel. In: Lumley,
® .H.de., Labeyrie, J. Eds. , Absolute Dating and Isotope Analy-
ses in Prehistory-Methods and Limits. CNRS, Paris, 457­492.
Zagwijn, W.H., 1989. Vegetation and climate during warmer
intervals in the Late Pleistocene of Western and Central
Europe. Quat. Int. 3­4, 57­67.