13 Pages
English

Evidence for early Devensian (Weichselian) fluvial sedimentation: geochronological and palaeoenvironmental data from the upper pleistocene deposits at deeping St James, Lincolnshire, England [ Mise en évidence d'une sédimentation fluviatile du Devensien (Wechselien) ancien : données géochronologiques et paléoenvironnementales des dépôts du Pléistocène supérieur de deeping St James, Lincolnshire, Angleterre.] - article ; n°1 ; vol.15, pg 5-15

-

Gain access to the library to view online
Learn more

Description

Quaternaire - Année 2004 - Volume 15 - Numéro 1 - Pages 5-15
La détermination du fonctionnement du système fluvial au Devensien (Weichselien) nécessite l'utilisation de plusieurs méthodes de datation, les recherches antérieures ayant été gênées par les limites de la technique de datation au carbone 14. Cet article présente les résultats d'une étude dans laquelle les datations au carbone 14 et par luminescence stimulée optiquement (OSL) ont été réalisées simultanément sur des paires d'échantillons de sédiments de remplissage de chenaux contemporains et sur des échantillons plus dispersés. Cette recherche géochronologique a été entreprise en relation avec une description sédimentologique et paléontologique détaillée des sédiments pour déterminer la réponse du système fluvial à différentes échelles de changements climatiques pendant le Devensien. La succession sédimentaire sous la Première Terrasse de la rivière Welland à Deeping St James, Lincolnshire, se divise en trois associations de faciès sur la base de leur sédimentologie et des caractéristiques de leurs surfaces de contact. Les deux associations de faciès inférieures suggèrent une mise en place du dépôt dans un contexte climatique periglaciaire peu sévère, avec une abondance significative de végétation herbacée. Au contraire, la troisième association est attnbuée à une modification significative du climat périglaciaire. Des datations par OSL sur la partie inférieure de la séquence la datent du Devensien ancien (60-110 000 ans), et l'interprétation de l'environnement de dépôt est en accord avec les témoignages climatiques extérieurs pour cette période. Au contraire, les datations au carbone 14 pour la sequence sont toutes différentes de celles obtenues par OSL et sont regroupées entre -35 et -40 000 ans BP. Cela reflète une distorsion systématique de la technique radiocarbone à ce pas de temps, ce que l'on peut reconnaître seulement quand plusieurs méthodes de datation sont utilisées.
Determining fluvial system behaviour in the Devensian (Weichselian) Stage requires the application of multiple dating methods, because previous research has been hindered by the limitations of the radiocarbon dating technique. This paper presents results of a study where radiocarbon and optically-stimulated luminescence (OSL) dating was undertaken both on paired 'tie-point' samples from contemporaneous channel fills and on a number of control 'framework' samples. This geochronological investigation was undertaken in conjunction with detailed sedimentological and palaeontological description of the sediments to determine fluvial system response to different scales of climate change dunng the Devensian Stage. The sedimentary succession beneath the 'First Terrace' of the River Welland at Deeping St James, Lincolnshire is divisible into three faciès associations on the basis of their sedimentology and bounding surface characteristics. The two lower faciès associations suggest deposition in a penglacial climate of limited seventy, with abundant herb-rich vegetation In contrast, the third faciès association is attributed to significant penglacial modification of the sequence. OSL dates from the lower part of the sequence date from the Early Devensian Substage (~60 - 1 10 ka) and the interpreted environment of deposition is consistent with external climate evidence from this penod In contrast, radiocarbon dates from the sequence are consistently different from OSL age estimates, clustenng between ~35 and 40,000 BP. It is argued that this reflects a systematic bias in the radiocarbon technique at this time range, which can only be recognised when multiple dating techniques are applied
11 pages
Source : Persée ; Ministère de la jeunesse, de l’éducation nationale et de la recherche, Direction de l’enseignement supérieur, Sous-direction des bibliothèques et de la documentation.

Subjects

Informations

Published by
Published 01 January 2004
Reads 39
Language English
Document size 2 MB

Rebecca M.Briant
G. Russell Coope
Richard C.Preece
philip Gibbard
Evidence for early Devensian (Weichselian) fluvial
sedimentation: geochronological and palaeoenvironmental data
from the upper pleistocene deposits at deeping St James,
Lincolnshire, England [ Mise en évidence d'une sédimentation
fluviatile du Devensien (Wechselien) ancien : données
géochronologiques et paléoenvironnementales des dépôts du
Pléistocène supérieur de deeping St James, Lincolnshire,
Angleterre.]
In: Quaternaire - Volume 15 - Numéro 1-2 - 2004. pp. 5-15.
Citer ce document / Cite this document :
M.Briant Rebecca, Coope G. Russell, C.Preece Richard, Gibbard philip. Evidence for early Devensian (Weichselian) fluvial
sedimentation: geochronological and palaeoenvironmental data from the upper pleistocene deposits at deeping St James,
Lincolnshire, England [ Mise en évidence d'une sédimentation fluviatile du Devensien (Wechselien) ancien : données
géochronologiques et paléoenvironnementales des dépôts du Pléistocène supérieur de deeping St James, Lincolnshire,
Angleterre.]. In: Quaternaire - Volume 15 - Numéro 1-2 - 2004. pp. 5-15.
doi : 10.3406/quate.2004.1750
http://www.persee.fr/web/revues/home/prescript/article/quate_1142-2904_2004_num_15_1_1750Résumé
La détermination du fonctionnement du système fluvial au Devensien (Weichselien) nécessite
l'utilisation de plusieurs méthodes de datation, les recherches antérieures ayant été gênées par les
limites de la technique de datation au carbone 14. Cet article présente les résultats d'une étude dans
laquelle les datations au carbone 14 et par luminescence stimulée optiquement (OSL) ont été réalisées
simultanément sur des paires d'échantillons de sédiments de remplissage de chenaux contemporains et
sur des échantillons plus dispersés. Cette recherche géochronologique a été entreprise en relation avec
une description sédimentologique et paléontologique détaillée des sédiments pour déterminer la
réponse du système fluvial à différentes échelles de changements climatiques pendant le Devensien.
La succession sédimentaire sous la " Première Terrasse " de la rivière Welland à Deeping St James,
Lincolnshire, se divise en trois associations de faciès sur la base de leur sédimentologie et des
caractéristiques de leurs surfaces de contact. Les deux associations de faciès inférieures suggèrent
une mise en place du dépôt dans un contexte climatique periglaciaire peu sévère, avec une abondance
significative de végétation herbacée. Au contraire, la troisième association est attnbuée à une
modification significative du climat périglaciaire. Des datations par OSL sur la partie inférieure de la
séquence la datent du Devensien ancien (60-110 000 ans), et l'interprétation de l'environnement de
dépôt est en accord avec les témoignages climatiques extérieurs pour cette période. Au contraire, les
datations au carbone 14 pour la sequence sont toutes différentes de celles obtenues par OSL et sont
regroupées entre -35 et -40 000 ans BP. Cela reflète une distorsion systématique de la technique
radiocarbone à ce pas de temps, ce que l'on peut reconnaître seulement quand plusieurs méthodes de
datation sont utilisées.
Abstract
Determining fluvial system behaviour in the Devensian (Weichselian) Stage requires the application of
multiple dating methods, because previous research has been hindered by the limitations of the
radiocarbon dating technique. This paper presents results of a study where radiocarbon and optically-
stimulated luminescence (OSL) dating was undertaken both on paired 'tie-point' samples from
contemporaneous channel fills and on a number of control 'framework' samples. This geochronological
investigation was undertaken in conjunction with detailed sedimentological and palaeontological
description of the sediments to determine fluvial system response to different scales of climate change
dunng the Devensian Stage. The sedimentary succession beneath the 'First Terrace' of the River
Welland at Deeping St James, Lincolnshire is divisible into three faciès associations on the basis of their
sedimentology and bounding surface characteristics. The two lower faciès associations suggest
deposition in a penglacial climate of limited seventy, with abundant herb-rich vegetation In contrast, the
third faciès association is attributed to significant penglacial modification of the sequence. OSL dates
from the lower part of the sequence date from the Early Devensian Substage (~60 - 1 10 ka) and the
interpreted environment of deposition is consistent with external climate evidence from this penod In
contrast, radiocarbon dates from the sequence are consistently different from OSL age estimates,
clustenng between ~35 and 40,000 BP. It is argued that this reflects a systematic bias in the
radiocarbon technique at this time range, which can only be recognised when multiple dating techniques
are appliedQuaternaire, 15, (1-2), 2004, p 5-15
EVIDENCE FOR EARLY DEVENSIAN (WEICHSELIAN)
FLUVIAL SEDIMENTATION: GEOCHRONOLOGICAL
AND PALAEOENVIRONMENTAL DATA
FROM THE UPPER PLEISTOCENE DEPOSITS
AT DEEPING ST JAMES, Lincolnshire, England
Rebecca M. BRIANT1 et4, G. Russell COOPE2, Richard C. PREECE3
and Philip L. GIBBARD1
ABSTRACT
Determining fluvial system behaviour in the Devensian (Weichsehan) Stage requires the application of multiple dating methods, because pre
vious research has been hindered by the limitations of the radiocarbon dating technique This paper presents results of a study where radiocarbon and
optically-stimulated luminescence (OSL) dating was undertaken both on paired 'tie-point' samples from contemporaneous channel fills and on a number
of control 'framework' samples This geochronological investigation was undertaken in conjunction with detailed sedimentological and palaeontological
description of the sediments to determine fluvial system response to different scales of climate change dunng the Devensian Stage The sedimentary
succession beneath the 'First Terrace' of the River Welland at Deeping St James, Lincolnshire is divisible into three faciès associations on the basis
of their sedimentology and bounding surface characteristics. The two lower faciès associations suggest deposition in a penglacial climate of limited
seventy, with abundant herb-rich vegetation In contrast, the third faciès association is attributed to significant penglacial modification of the sequence.
OSL dates from the lower part of the sequence date from the Early Devensian Substage (~60 - 1 10 ka) and the interpreted environment of deposition
is consistent with external climate evidence from this penod In contrast, radiocarbon dates from the sequence are consistently different from OSL age
estimates, clustenng between ~35 and 40,000 BP It is argued that this reflects a systematic bias in the radiocarbon technique at this time range, which
can only be recognised when multiple dating techniques are applied
Key-words: Geochronology, optically-stimulated luminescence (OSL), radiocarbon, fluvial sediments and response, climate change
RÉSUMÉ
MISE EN ÉVIDENCE D'UNE SÉDIMENTATION FLUVIATILE DU DEVENSIEN (WEICHSELIEN) ANCIEN DONNÉES GÉOCHRONOLOGI
QUES ET PALEOENVIRONNEMENTALES DES DÉPÔTS DU PLEISTOCENE SUPERIEUR DE DEEPING ST JAMES, Lincolnshire, Angleterre
La determination du fonctionnement du système fluvial au Devensien (Weichsehen) nécessite l'utilisation de plusieurs méthodes de datation,
les recherches anteneures ayant ete gênées par les limites de la technique de datation au carbone 14 Cet article présente les résultats d'une étude dans
laquelle les datations au carbone 14 et par luminescence stimulée optiquement (OSL) ont été réalisées simultanément sur des paires d'échantillons
de sédiments de remplissage de chenaux contemporains et sur des échantillons plus dispersés Cette recherche géochronologique a été entrepnse en
relation avec une descnption sédimentologique et paleontologique détaillée des sediments pour déterminer la réponse du système fluvial à différentes
échelles de changements climatiques pendant le Devensien La succession sedimentaire sous la " Première Terrasse " de la nvière Welland à Deeping St
James, Lincolnshire, se divise en trois associations de faciès sur la base de leur sédimentologie et des caractéristiques de leurs surfaces de contact Les
deux associations de faciès inféneures suggèrent une mise en place du dépôt dans un contexte climatique penglaciaire peu sévère, avec une abondance
significative de végétation herbacée Au contraire, la troisième association est attnbuee à une modification significative du climat penglaciaire Des
datations par OSL sur la partie inféneure de la séquence la datent du Devensien ancien (60-110 000 ans), et l'interprétation de l'environnement de
dépôt est en accord avec les témoignages climatiques exteneurs pour cette pénode Au contraire, les datations au carbone 14 pour la sequence sont
toutes différentes de celles obtenues par OSL et sont regroupées entre -35 et -40 000 ans BP Cela reflète une distorsion systématique de la technique
radiocarbone à ce pas de temps, ce que l'on peut reconnaître seulement quand plusieurs méthodes de datation sont utilisées.
Mots-clés : Geochronologie, datation OSL, datation par le carbone 14, sediments fluviatiles, paléochmats.
' Godwin Institute of Quaternary Research, Department of Geography, University of Cambridge, Downing Place, CAMBRIDGE, CB2 3EN, England
Email Rebecca bnant@kcl ac uk
2 Centre for Quaternary Research, Department of Geography, Royal Holloway, University of London, EGHAM, Surrey, TW20 0EX
3 Department of Zoology, Downing Street, CAMBRIDGE, CB2 3EJ, England
4of Geography, King's College London, Strand, LONDON, WC2R 2LS, England
Manuscrit reçu le 22/12/2002, accepté le 03/01/2004 INTRODUCTION Fenland to different scales of climate change during the
Devensian Stage (Briant, 2002; Briant et al, submitted).
The palaeoenvironment of the Devensian Stage in Bri
tain is known largely from fluvial deposits, since glacial THE DEPOSITIONAL SUCCESSION
and periglacial deposits are more fragmentary and lack
detailed palaeontological and sedimentological informat The site at Deeping St James occurs within a low gravel
spread mapped as 'First Terrace Gravels' ion. Therefore study of these sediments is essential to by the British
understand the nature of landscape response to climate Geological Survey (Horton, 1989), and spans grid refe
fluctuations during this time period. It has previously rences TF 182084 to 184082 (fig. 1). A number of expo
been suggested that the main period of aggradation in sures were examined between February and July 1999.
most river systems in lowland Britain occurred during The sections were recorded using a combination of ver
the Middle Devensian Substage (c. 50-26 ka), with limi tical sediment logs, drawings of exposures and photo
ted deposits preserved from the Early Devensian Subs- graphs, the latter of which were important to record the
tage (c. 110-50 ka, van Huissteden et al, 2001). Howev significant lateral variability within the sequence. Field
er, the chronological control on much of the research description of sediments concentrated on those features
synthesised in this review is based almost entirely on the which enabled elucidation of the relationship between
radiocarbon technique, with isolated dates from single fluvial system behaviour and climate, as outlined in
unconnected sites. Therefore the pattern of sedimentat Lewis et al (2001). Key bounding surfaces were iden
ion observed may be an artefact of the limitations of this tified and classified using the scheme of Miall (1996),
allowing subdivision of the sediments into 'faciès assodating technique.
In addition to the classic Chelford Interstadial sequence ciations'. The faciès within these were described using
Miall 's (1996) faciès codes, with some modifications (Simpson and West, 1958; Coope, 1959), Early Devens
ian deposits of both stadial and interstadial affinities (Briant, 2002).
have been identified on the basis of various criteria The Pleistocene succession consists of up to 4 m of fine
at Wretton (West et al, 1974; Coope, 1974); Stanton to coarse limestone-rich gravel overlying Jurassic Oxford
Harcourt (Gibbard, 1985; Seddon and Holyoak, 1985); Clay bedrock (Horton, 1989), with a lower level of-1 m
O.D. and upper surface between 2 and 3 m O.D. Finer- Brimpton (Bryant et al, 1983, but see Bridgland, 1994;
Worsley and Collins, 1995); Cassington (Maddy et al., grained material (sand, silt, organic material) is present
1998) and Ashton Keynes (Lewis et al, 2001). Howev both as matrix material and within specific beds in the
er, few of these sites have good chronological control, predominantly gravel-rich succession. Samples of this
because age attributions are often limited in number and finer-grained material were collected for radiocarbon
reliant on either radiocarbon or luminescence dating in dating at NERC facilities in East Kilbride and OSL
isolation. Both techniques were applied at Stanton Harat the Godwin Laboratories, Cambridge, from both
'tie-point' and 'framework' locations as defined above. court, where sediments interpreted as forming part of the
same aggradation gave a radiocarbon age of 34,730+440 Palaeontological analysis was also undertaken on those
BP (SRR-2302) and a thermoluminescence (TL) age of samples collected for radiocarbon dating, as shown on
~91 ka (Seddon and Holyoak, 1985). The discrepancy figure 2. The deposits described below overlie organic
between these two age determinations is difficult to inter sediments within the gravel succession attributed to the
pret because the samples were taken from different expo Ipswichian temperate Stage (Keen et al, 1999). These
sures at some distance from each other, and thus the exact formed a large arcuate channel feature over 70 m wide,
relationship between them remains in some doubt. which ran east-west for approximately 1 km within the
In contrast, this paper reports the results of investiga area of the quarry marked on figure lb, overlying hor
tions into Devensian-age sediments at Deeping St James, izontally-bedded gravel. To the north of the channel fea
Lincolnshire, which, for the first time, apply radiocarbon ture, the lower bounding surface of the overlying gra
and optically-stimulated luminescence (OSL) dating to vel-rich sediments declines in altitude, removing all trace
samples interpreted as being directly contemporaneous. of the previous two aggradations, and it is these overlying
This approach involves the use of 'tie-point' locations gravels which are laterally equivalent to the deposits des
where paired radiocarbon and OSL samples were taken cribed below. The succession can be broadly divided
from the same channel fill in an exposed face, together into three faciès associations, often found in stratigraphie
with a larger number of 'framework' dates (both radio superposition, recognisable in the field on the basis of
carbon and OSL) from further channel fills within each significant bounding surfaces within the sequence and
faciès association identified. It could be argued that radio similar faciès types. The sedimentological properties of
carbon dating should not be undertaken on sediments these are summarised in table 1 and described below.
which OSL age determinations suggest are beyond the
limit of the radiocarbon technique. However, the use of FACIES ASSOCIATION DSJ-1
both techniques is valuable because such sediments may Faciès association DSJ-1 is present throughout the
not always yield the infinite radiocarbon age estimates quarry and consists of up to 2 m of clast-supported fine
which would show them to be beyond this limit (e.g. to coarse limestone-rich gravel. Beds and eroded clasts of
Roberts et al, 1994; Chappell et al, 1996). The paper organic and inorganic silt and sand are also seen. DSJ-1
forms part of a larger study investigating the response is dominated by horizontal bedding with sand and silt
of the Welland and Nene river systems in the northern drapes (Gh). Also present are lenses of planar cross-bed- gravels (Gp), and shallow channel forms with width- ded frequently infilled with cross-bedding aligned with the
depth ratios greater than 10. These latter are infilled with scour form (Gs, Ss, cf. Bnant, 2002) and occasionally
sand and silt (Sh, Sr, Sm, Fl), and frequently also organic contain beds of finer-grained material or compressed
debris. Wedge forms and involutions are observed rarely, plant debris (Fl). DSJ-2 is underlain by an erosional and
both truncated within the succession and extending its discontinuous fourth-order bounding surface.
full depth. The faciès association DSJ-1 is underlain by a
FACIES ASSOCIATION DSJ-3 sixth-order bounding surface, cutting erosionally into the
underlying Jurassic bedrock. Faciès association DSJ-3 occurs throughout the quarry
beneath the Holocene soil, varying in thickness between
1 and 2 m. It consists of fine to coarse matrix-rich clast- FACIES ASSOCIATION DSJ-2
This faciès association varies in thickness between supported limestone-rich pebbles with a coarse sand
40 cm and 1-2 m and is highly variable in texture, inclu matrix. The faciès association is mostly massive (Gem),
ding clast-supported fine to coarse pebbles with sand named faciès DSJ-3b, with indistinct invo
matrix; sand with fine pebbles; and silts, sometimes con lutions and occasional sandy lenses (Sm). Sedimentary
taining compressed plant debris. The faciès association structures are only observed rarely (near logs I, H, B and
D), named faciès association DSJ-3a. These consist preconsists predominantly of deep intercutting scour forms,
with width-depth ratios mostly less than 10. These are dominantly of horizontal bedding (Gh), with occasional
a) b)
»-|«- Main mads km 0 10 \b Logs /Quarry feces <*W Involutions terminating Drainage ditches Major towns and at»« mi)el - — <- Railway lines _ terminating Tracks Roads W Involutions terminating D«vsnsian ice km» y Wedge form terminating within DSJ-3 Location of -• — *- Electricity lines withm DSJ-1 8 DSJ 2 within DSJ-1 & DSJ-2 main figure
d)
GROUND SURFACE
3
OVERBURDEN STRIP A 0\ \ A «\ A \ \ 0) A A 2.5
Q 2
O 1.5
E0.5
0
-0.5
BEDROCK
TOPSOIL MATERIAL ORGANIC CD SAND GRAVEL
Fig. 1: a) Location of the study site, b) Location of exposures described by Keen et al. (1999) and from February to July 1999, c) Position
of recorded exposures within the gravel quarry, d) Schematic cross-section through the Pleistocene deposits at Deeping St James, based on
exposures seen between February and July 1999, showing the nature of and relationships between the three faciès associations identified at the
site.
Fig / • a) Localisation du site étudié, b) localisation des affleurements décrits par Keen et al (1999) et de février à juillet 1999 par les auteurs, c)
localisation des affleurements relevés dans la graviere, d) coupe transversale des dépôts pleistocenes à Deeping St James, d 'après les affleurements
observés entre février et juillet 1999, montrant la nature et les relations entre les trois associations de faciès identifiées sur le site B c D H
DSJ-3b [DSJ-3a 1 I DSJ-1 DSJ-3a ▲ DSJ50 (CL) FT DSJ-2 IlDSJ48(L)
DSJ-1 ciofto poflo „„ DSJ-1 T j DSJ-1 [A DSJ49 i #DSJ52 XJ I#DSJ46(14C) (CL) (14C)
N
I T DSJ-3b r DSJ-3b DSJ-3a DSJ-3b
1 IDSJ-2 ÎDSJ-2 DSJ-1 DSJ-1 J#DSJ53 Ht1- DSJ-1 (14C)
(14C)
Q R
SiSaP aSaP SiSaP
Fig. 2: Summary sedimentary logs showing spatial variability within sediments at Deeping St James. Locations of samples are shown. Key
as in figure ld. Vertical scale in metres O.D., with values quoted where log heights were obtained by surveying, and blank where they were
extrapolated. Si = silt, sa = sand, p = pebbles. Filled circles = radiocarbon samples (14C), filled squares = luminescence samples (L), filled
triangles = other samples (clast lithology (CL), palaeontology (P) and vertebrate material (V)).
Fig 2 Logs sedimentaires synthétiques montrant la variabilité spatiale des sediments a Deeping St James avec localisation des lieux de prélèvement
des échantillons Légende identique à celle de la figure ld Echelle verticale en metres O D Les espaces blancs correspondent aux zones ou elles ont
ete extrapolées Si = sût, sa = sables, p = cailloux Cercles pleins = prélèvements pour carbone 14, carres pleins = prélèvements pour OSL, triangles
pleins = autres prélèvements (pétrographie des cailloux (CL), malacologie, insectes (P), fragments de vertèbres (V)
lenses of planar cross-bedded sand (Sp) and scour-fills then deposited in abandoned channel forms. The obser
(Gs). With the exception of a small crack near log G, ved dominance of bars and bedforms (Miall's (1996)
only one wedge form was recorded, near log H. The architectural elements GB, SB, tab. 1) suggests that water
other wedge forms at this stratigraphie level are less clear depths were probably shallow, and a suitable analogy for
because faciès association DSJ-3 is predominantly mass deposition is Miall's (1996) shallow gravel-bed braided
ive. The lower bounding surface of faciès association model. The higher amounts of finer-grained material pre
DSJ-3 is gradational, except near log D where it is planar sent in faciès association DSJ-1 than in the model may
and erosional and may be fifth-order. suggest that smaller flood events were also important in
shaping the landscape (Bryant, 1983). These would have
INTERPRETATION OF CHANGING DEPOSITIONAL reworked deposits from high-stage flows and deposited
ENVIRONMENTS finer particle sizes, as in the subarctic nival hydrological
Faciès association DSJ-1 indicates fluvial deposition regime defined by Church (1974). Permafrost may have
by a river transporting gravel at high flows and signifi been present during deposition of the faciès association,
cant quantities of finer material at low flows, which was although evidence for this is limited.
Faciès association Faciès assemblage Lower bounding surface (and order) Geometry
DSJ-3b Gem, Sm Gradational (non-fluvial m origin) Massive gravels with involutions and occasional massive sand beds
DSJ-3a Gh, Sp, Gs Planar, erosional (fifth-order) Horizontally-bedded gravel Minor sand and gravel-filled scours
DSJ-2 Gs, Ss, Fl Concave-up, discontinuous, erosional (fourth-order) Deep intercutting scour forms
DSJ-1 Gh, Gp, Sh, Sr, Sm, Fl Planar, erosional (sixth-order) Horizontally bedded gravel, sand and silt drapes
Shallow channel forms infilled with sand, silt and organic debris
Architectural element(s) Interpretation Depositional environment Faciès association
None Penglacial modification of previously-deposited material Area of floodplain no longer fluvially active DSJ-3b
Shallow gravel- bed braided river GB Gravel bars and bedforms Bar-top scour-fills DSJ-3a
HO Scour hollows, infilled during falling stage flov /s Scour-dominated braided river DSJ-2
GB, SB Gravel barforms Sand bedforms deposited within shallow channels Shallow gravel-bed braided river DSJ-1
Tab. 1: Sedimentological characteristics and interpreted environment of deposition of faciès associations 1 to 3 at Deeping St James, Lincolnshire,
England.
Tab 1 Caractéristiques sédimentologiques et interprétation de l'environnement de dépôt des faciès I à 3 In contrast, greater water depths are indicated during Most of the samples from DSJ-1 and DSJ-2 are domi
deposition of faciès association DSJ-2. The dominance nated by leaves, fruits and capsules of Salix or Betula
of scour hollows (HO) suggests that there were a limited nana, with varying richness of herb floras (tab. 2), pos
number of active channels in the fluvial system. However, sibly relating to taphonomic variations (cf. West et al.,
there is no evidence of the widespread overbank depos 1993; West, 2000). Also present are small numbers of
ition characteristic of a meandering or wandering plan- Alnus sp. fruits, which may have come from either shrub
form (Reading, 1996; Church, 1983). For this reason, it or tree alder or, more likely, have been reworked. Whilst
is suggested that the sediments characteristic of faciès Alnus has been recorded from the Devensian in the past
association DSJ-2 represent a 'scour-dominated' braided (Godwin, 1975), these records are scattered and incon
style, possibly with deeper water and better developed clusive and it is generally thought that the taxon was not
bedforms than the shallow gravel-bed braided river of present in Britain at this time.
faciès association DSJ-1. The high proportion of finer- Of the herbs, a number indicate grassland such as
grained particle sizes suggest that this faciès association Armeria maritima, Linum perenne and Potentilla anse-
may also have been deposited under a subarctic nival nna. Meadow taxa include Taraxacum sp., Leontodon
hydrological regime (Church, 1974). sp. and Achillea millefolium. Polygonum aviculare is
The sediments of faciès association DSJ-3 represent common, suggesting disturbed ground. The presence of
two stages of landscape modification. The first stage Carex may indicate significant areas of ground marginal
(DSJ-3a) was fluvial, consisting of gravel deposition in to main water bodies, although these taxa are common
a shallow-bed braided river system, because of the domi in all deposits and may be reworked. Most of the aquatic
nance of gravel bar and bedform deposits. It seems likely taxa (e.g. Potamogeton sp., Scirpus sp., Ranunculus and
that the hydrological regime was dominated by a single Zannichellia palustris), although found in many settings,
peak flood in an arctic nival discharge regime (Church, are typical of Devensian floras (Godwin, 1975; West,
1974). The second stage of landscape modification invol 2000). In contrast, Sparganium erectum is a temperate
ved widespread and severe periglacial activity, leading to species almost certainly reworked from interglacial depos
cryoturbation and ice-wedge development, represented its at the site (cf. Keen et al, 1999). Overall, the local
by involutions, stone erection and wedge forms within environment recorded seems to consist of a braidplain
the predominantly massive sediments of faciès associa with grassland and meadow including dwarf birch and
tion DSJ-3b. This suggests a period of fluvial inactivity willow on established bars, and disturbed ground in more
at the site, coupled with the presence of continuous per active areas. The diversity of the floras in some of these
mafrost. Murton et al. (1995) observe that repeated cryo samples is greater than other cold-stage floras, even from
turbation in the modern active layer at Crumbling Point similar sedimentary settings (Briant et al., submitted).
(North- West Territories, Canada) has destroyed texturally This may indicate either less severe climatic conditions
distinct involutions previously developed during a period (West, 2000) or simply greater reworking at this site.
of thawing, which may also have occurred in this case. A
period of thaw followed. MOLLUSCAN ANALYSIS OF ORGANIC MATERIAL
PRESERVED WITHIN FACIES ASSOCIATIONS DS J- 1
AND DSJ-2 PALAEONTOLOGY
Very few samples from Deeping St James yielded Mol-
PLANT MACROFOSSIL ANALYSIS OF ORGANIC lusca (tab. 3), and DSJ16 comprises only a few Pisidium
MATERIAL PRESERVED WITHIN FACIES ASSO sp. hinges. In contrast both DSJ33 and 34 contain shells
CIATIONS DSJ-1 AND DSJ-2 of a wide diversity of species. However, interpretation of
Plant macrofossil material was separated from bulk this assemblage is hindered by the likelihood of rewor
samples from a number of organic beds, primarily for king of the shells, which are often broken. All the species
radiocarbon dating. It was prepared by wet sieving with present are also seen in the interglacial deposits reported
distilled water, followed by separation and identification by Keen et al. (1999), and a number of these (such as
under a low-power stereo microscope. Samples were Carychium tridentatum and Belgrandia marginata) are
dried overnight at 105°C after identification to avoid usually found only in interglacial deposits. Reworking,
fungal growth following the recommendations of Wohl- especially of shell material, is a common occurrence in
farth et al. (1998). Macrofossil material was mostly high-energy river systems, as may be seen in the presence
poorly-preserved, with leaves broken, and seeds fr of temperate fossils within cold-stage gravel deposits at
equently degraded, although DSJ16 and DSJ33/34 were Somersham (West et al., 1994) and Sidgwick Avenue,
Cambridge (Lambert et al., 1963). slightly better-preserved.
The main problem in interpreting these assemblages is If contemporary, the assemblage from DSJ33 and 34
the possibility of reworking, which is particularly acute suggests a treeless environment with sluggish or stagnant
at this site given the known proximity of interglacial aquatic habitats. It could be argued that this is possible,
since at least 2 1 of the 29 species are known from other deposits (Keen et al., 1999). A number of taxa stand out
as probably reworked and are discussed below, but some Devensian contexts (Holyoak, 1982), excluding the Late-
less distinctive taxa are also affected. For this reason all glacial period. However, these records may also contain
reworked elements. Some species, such as Bithynia ten- the interpretations below, although based on those taxa
taculata, did occur during the thermal optimum of the not previously found at the site, should be treated with
Upton Warren interstadial (Kerney et al., 1982), but the some caution. 10
Faciès association sample DSJ-1 DSJ-1 DSJ-1 DSJ-1 DSJ-1 DSJ-1 DSJ-2 DSJ-2
Sample DSJ16 DSJ22 DSJ45 DSJ46 DSJ52 DSJ53 DSJ33 DSJ34 Sediment type Plants in sand Plant bed in sand Organic silt Organic silt Organic silt Plant bed in sand Plant bed in sand Plant bed in sand
Taxa
Shrubs 24 Alnus sp fr Betulancma L If abundant abundant abundant abundant
Betulanana L fr »200 6 53 Betula nana L esc 1
Dryas octopetala L If 4 Salve herbacea L If abundant abundant abundant Sahxpolans Wahl If Sahx cf repens If abundant abundant sp If 126 Sahx sp cap abundant abundant abundant abundant Sahx/Betula If Herbs 1 2 Achillea millefohum L a Armena mantima (Mill ) Willd c 2 12 77
§ Caryophyllaceae s Cerastium sp s 12 Compositae a 33 72 66
Cruciferae s 1 28 several Draba sp cap Euphorbia sp s Leontodon sp a
Linumperenne agg s 17 Plantago sp s
Polygonum aviculare agg fr 11 37 Potentdla ansenna L a 40 crantzu/tabernaemontam type a
§ Potentdla sp a 68 19
§ Rubus sp s 7 £ Ranunculus (subg ranunculus) a 29
&/e«e sp s Taraxacum sp a 38 Thahctrum sp a 5 § Umbelhferae fr 6 29 2 4 17 § Viola sp s Grasses/Sedges 14 ^ Carex aquatihs -type n 87 19 9 39 $ tngonous-type n 24 13 10
Aquatics Chara o 1 1 1 § Eleochans sp n § Potamogeton sp fst 57 9 12 24 6 12 ^ Ranunculus (subg Batrachium) a $ Scirpus lacustris L n 2 27 5 9 sp n $ Spargamum erectum L fr 6 9 Zannichelha palustns L a 4 1 13
Tab. 2: Plant macrofossil assemblages from samples taken for radiocarbon dating. Fr, fruit; If, leaf fragment; esc, conescales; cap, capsule;
a, achene; c, calyx; s, seed; fr, fruit; n, nutlet; o, oospore; fst, fruitstone. Values given are actual counts. § indicates those taxa recorded from
interglacial deposits at the site (Keen et al, 1999). Sample weights and material processed: DSJ16 sieved ~600 g, picked ~2/3 of >1 mm fraction;
DSJ22 sieved 150 g, picked all >250 fim fractions; DSJ4S sieved 300 g, picked all >250 (am fraction; DSJ46 sieved 1 kg, picked all >500 |im
fractions; DSJ52 sieved 1 kg, picked all >1 mm fraction; DSJ53 sieved 1 kg, half >1 mm DSJ33 1 kg, all >250 \im DSJ34 2 kg, all of >500 (xm fractions.
Tab 2 Assemblages de macrofossûes de plantes des échantillons utilisés pour les datations par le carbone 14 Les taxons utilisés pour les datations
sont indiqués en caractères gras Fr, fruit , If, fragment de feuille , esc, écailles de cône , cap, capsule , a, akène , c, calice , s, graine ; n, noisette , o,
oospore , fst, noyau de fruit Les valeurs données sont des valeurs réellement comptées § indique les taxons enregistrés dans les dépôts interglaciaires
de ce site (Keen et al , 1999)
associated Coleoptera at Deeping St James indicate that teran skeletons degrade rapidly once in contact with
conditions were far colder. It therefore seems likely that air, making them less susceptible to reworking. Theref
Bithynia tentaculata at least is also derived from intergla- ore, despite the evidence for reworking in the mollus-
cial sediments, and possibly also many of the other taxa. can assemblages, this is not thought to have affected the
Coleoptera in these samples.
COLEOPTERAN ANALYSIS OF ORGANIC MATE The local riverine environment suggested is still or
RIAL PRESERVED WITHIN FACIES ASSOCIATION slowly flowing, as preferred by Helophorus sp. and
DSJ-2 Colymbetes sp., although this may reflect only the seve
Coleopteran analysis was undertaken on two samples rity of the climate since running water species are not
from the 'tie-point' channel-fill within faciès association common in arctic environments. Parts of the braidplain
DSJ-2 (samples DSJ33/DSJ34), in which fossils were must have been considerably drier, however, because the
fairly well-preserved. The fauna described is consistent Otiorhynchus species present are largely xerophilic. The
abundance of these species also suggests significant vegewith deposition in a single climatic regime, and it may
tation on the braidplain. This vegetation includes mosses be seen from table 4 that the number of species present
(Simplocaria metallica and S. semistriata are obligate which have also been observed in the interglacial depos
its (Keen et al, 1999) is limited and restricted largely to moss-eaters); various herbs (favoured by Otiorhynchus);
marshy vegetation (as indicated by the presence of Nota- taxa which could not be determined to species level. In
ris aethiops) and areas of damp vegetable debris (favouaddition, none of the diagnostic and non-native species
on which interpretation is based have been recorded pre red by Pycnoglypta lurida, Olophrumfuscum and Eucne-
viously. Furthermore, in contrast with mollusca, cosum brachypterum). 11
Faciès association DSJ-1 DSJ-2 DSJ-2 Faciès association DSJ-2 DSJ-2
Sample DSJ 33 DSJ 34 Sample DSJ16 DSJ33 DSJ34
Species Coleoptera in taxonomic order Aquatic Carabidae Valvata cristata Muller 3 Nebna gyllenhah (Schoen ) 1 piscinahs (Muller) 3 66 Notiophilus aquaticus (L ) 1 Belgrandia marginata (Michaud) 7 23 Bembidion bipunctatum (L ) 1 Bithynia tentaculata (L ) 5 16 *Bembidion hyperboraeorum Munst 1
Bithyma opercula 4 Amara quenseh (Schoenh ) 1
Lymnaea truncatula (Muller) 1 11 alpina (Payk ) 1
Dytiscidae palustns agg 2 §Hydroporus sp 1 Lymnaea peregra (Muller) 14 180 Coelambus sp 2 Anisus leucostoma (Millet) 2 §Agabus sp 1 Gyraulus laevis (Alder) 3 Colymbetes sp 1 crista (L ) 1 3 Hydraenidae Hippeutis complanatus (L ) 1 §Hydraena sp 1 Planorbis planorbis (L ) 1 10 Ochthebius sp 2 1
Sphaerium corneum (L ) 6 *Helophorus obscurellus Popp 1 Helophorus grandis Ilhger 1 1 4 Muscuhum lacustre (Muller) 3 ^Helophorus "aquaticus " (L ) = aequahs Thoms 1 Pisidium subtruncatum Malm 11 aspericolhs Angus 6 henslowanum (Sheppard) 5 19 §Helophorus mise small spp 7 Pisidium nitidum Jenyns 3 Staphylinidae sp. 22 240 *Pycnoglypta lurida (Gyll ) 1 Terrestrial Olophrum fuscum (Grav ) Carychium minimum Muller 1 1 Eucnecosum brachypterum (Grav ) or *norvegicum Munst 1 1 tndentatum (Risso) 10 Geodromicus sp
4 Philonthus sp sp Tachyporus sp Pupilla muscorum (L ) 2 21 Tachinus sp Vallonia costata (Muller) 6 §Aleochannae Gen et sp Indet 2 5 Valloma pulchella 6 Elatendae Vallonia pulchella/ excentnca 22 Zorochrus sp 1 Vallonia enniensis (Gredler) 3 Byrrhidae 3 sp Simplocana semistriata (F ) 1 Punctum pygmaeum (Draparnaud) 4 *Simplocana metalhca (Sturm,) 1 Vitrina pellucida Muller 1 Anthicidae
Cepaea / Arianta 2 Anthicus ater (Panz ) 1
Scarabaeidae §Aphodius sp 13 5
Chrysomehdae §Donacia thalassina Germ 1
Chrysomela marginata L 1 Tab. 3: Molluscan assemblages from samples taken for radiocar
Curcuhonidae bon dating and palaeontological analysis. Values given are actual Otiorhynchus arcticus (F ) 2 counts. Pisidium sp. in DSJ16 includes Pisidium amnicum, Pisidium hgneus (01 ) 1 henslowanum', in DSJ34 it includes Pisidium subtruncatum, rugifrons (Gyll ) 2 nitidum, Pisidium casertanum, moitesserianum. Sample Otiorhynchus sp 1 weights and material processed: DSJ16 sieved ~600 g, picked -2/3 Sitona sp 1 of 1 mm fraction; DSJ33 sieved 1 kg, picked all >250 |im fractions; §Notaris acndulus (L ) 1 DSJ34 sieved 2 kg, picked all >500 |xm fractions. Notaris aethiops (F ) 1 Tab 3 Assemblages de mollusques des échantillons utilisés pour les
datations au carbone 14 et les analyses paléontologiques Les valeurs
données sont des valeurs réellement comptées.
Tab. 4: Coleopteran assemblages from samples taken for radiocarClimatically, the assemblages from these two samples bon dating and palaeontological analysis from Deeping St James.
Values given are actual counts. Species marked with a * are not are similar and suggest that climate was equivalent to native in Britain at the present day. § indicates those species recorthat above the present-day tree-line in Northern Europe. ded from interglacial deposits at the site (Keen et ai, 1999). Sample
weights and material processed: DSJ33 sieved 1 kg, picked all Four obligate cold-adapted species are present which are
>250 um fractions; DSJ34 sieved 2 kg, picked all >250 um fracnot now native to Britain: Bembidion hyperboraeorum, tions. Helophorus obscurellus, Pycnoglypta lurida and Simplo- Tab 4 • Assemblages de coléoptères des échantillons utilisés pour les
datations au carbone 14 et les analyses paléontologiques Les valeurs caria metallica/elongata; all of which have arctic and données sont des valeurs réellement comptées Les espèces notées avec continental distributions today. Amara alpina is a cha une * ne pas présentes en Grande Bretagne actuellement § indi
que les espèces qui sont enregistrées dans les dépôts interglaciaires de racteristic species of the tundra and alpine zones of the
ce site (Keen et al , 1999) mountains (it survives in Britain at very high altitudes in
the Scottish Highlands). Mutual Climatic Range (MCR
-Atkinson et al, 1987; Coope et al, 1998) temperature
'tie-point' sample DSJ34 within faciès estimates from these assemblages give mean July temper sub-samples from
association DSJ-2. Submitted samples consisted of plant atures between 7°C and 9°C, and mean January/February
temperatures between -31°C and -14°C. macrofossils chosen carefully to avoid potential problems
of hard- water error or reworking. They comprised terres
trial species which could have formed part of the contemGEOCHRONOLOGY
porary flora and varied between samples depending on
abundance (tab. 5). RADIOCARBON DATING
Radiocarbon dates for faciès associations DSJ-1 and Samples were submitted for radiocarbon dating from
DSJ-2 range between -35,000-45,000 yr BP with no seven locations within the sequence, with three replicate 12
and considered part of the distribution using Chauvenet's obvious groupings (tab. 5) This similarity may imply
that the transition between faciès associations DSJ-1 and criterion (Taylor, 1997). No preheat plateau was obser
ved for sample DSJ48, and thus this sample must be treaDSJ-2 represents only a small-scale change within the
system, as suggested by the low order bounding surface ted with caution (Murray and Wintle, 2000), especially
at the base of DSJ-2. However, the consistency may also because other characteristics of the sample were also
reflect an 'event horizon' in radiocarbon dating, whereby unsatisfactory (Briant, 2002). Further details of lumines
cence behaviour may be found in Briant (2002). dates cluster around -40,000 BP (Roberts et al, 1994;
The age estimates from DSJ-1 and DSJ-2 suggest that Chappell et al, 1996). This seems likely because of
these two faciès associations were deposited between the observed mismatch between radiocarbon and OSL
age estimates at the 'tie-point' locations. If reliable, the ~60 and 110 ka. They agree well with a previous OSL
radiocarbon dates would suggest that plant material was date of 75.00 ± 7.9 ka from gravels equivalent to DSJ-1
which immediately overlay interglacial deposits (Keen et being preserved within the predominantly sand and gravel
succession in the Middle Devensian Substage between al., 1999). The age reversal between sample DSJ24a from
-37-45,000 BP. However, it should be noted that despite faciès association DSJ-1 and DSJ32 from DSJ-2 appears
yielding MCR temperature estimates similar to those of to contradict the stratigraphy at the site, although it may
Middle Devensian coleopteran faunas, the assemblages reflect the small-scale nature of the change between the
from faciès association DSJ-2 lack many of the exclu two faciès. Alternatively it may be due to the scatter inhe
sively eastern and Asiatic species characteristic of this rent in older OSL samples. These frequently have natur
Substage such as Diacheila polita and Aphodius holde- al OSL values which intersect with the flat section of an
exponential growth curve, causing tight distributions to reri (Coope, 2000).
become both scattered and skewed (Murray, 2001).
A more important issue is that the two 'tie-point' OPTICALLY-STIMULATED LUMINESCENCE (OSL)
OSL dates (DSJ24a and DSJ32) both give ages which DATING
Sand samples for optically-stimulated luminescence are significantly older (-30-40 ka) than their associated
dating were taken in opaque lengths of plastic tubing radiocarbon ages. This mismatch could reflect the fact
and stored in light-tight bags until processed. Prepara the ages require calibration, but approximate
tion to quartz followed the protocol outlined in Bateman calibration only changes the ages of these two faciès
and Catt (1996). Equivalent dose was determined in the associations to -42-44,000 cal. BP (after Beck et al,
2001). Alternatively, some might argue that such an offset Godwin Laboratories, Cambridge by R.M. Briant using
the Single Aliquot Regenerative protocol of Murray and could be caused by partial bleaching of fluvial sedi
Wintle (2000), with modifications to the preheat values ments. However, reported residual values of OSL signal
(Briant, 2002). Dose rates were determined by Neutron in modern temperate river systems do not exceed ~6 ka,
Activation Analysis of representative surrounding sedi significantly less than the offset observed here (Wallinga,
ments. Where sediments within a 30-cm radius of the 2002; Stokes et al, 2001). Furthermore, the rarer events
sample were heterogeneous, dose rates were integrated which could cause less complete bleaching seem unlikely
using Appendix H of Aitken (1985). Cosmic dose rates because of the universal nature of the offset.
were calculated using the expression of Prescott and Therefore, it seems likely that the radiocarbon dates
may be in error. A similar discrepancy has been obserHutton(1994).
The results of the OSL measurements, water content ved before in lake sediments, whereby samples shown
values and dosimetry data are shown in table 6. All OSL to be well beyond the radiocarbon limit on the basis of
samples have been checked for preheat dependence after the Uranium-Series, TL and amino-acid ratio techniques
Bailey et al. (2001) and the equivalent dose calculated have yielded spurious ages of between 37
only from those aliquots located on the 'preheat plateau', and 44,000 BP (e.g. Singh and Geissler, 1985; Magee et
Faciès association Location Sample Species radiocarbon-dated Radiocarbon laboratory code Date
DSJ-3 (within ice-wedge cast) DSJ48 32 57±4 07 ka Log H
DSJ-2 LogT DSJ34 (tie-point) Armeria mantima AA-41160 35,70O±8OO BP
Polygonum aviculare AA-41161 37,240±890 BP
Carex aquatihs -type AA-41162 40,300±1300BP
DSJ-2 LogT 78 56±6 76 ka DSJ32 (tie-point)
DSJ-1 LogO DSJ11 84 86±6 59 ka
DSJ-1 LogX DSJ16 Polygonum aviculare / Taraxacum sp CAMS-73416 40,590±1370BP Salix herbacea / polans CAMS-73420 DSJ22 (tie-pomt) 37,470±940 BP LogQ DSJ-1 37 25±2 55 ka DSJ24a (tie-point) LogU DSJ29 104 60±6 18 ka
DSJ-1 DSJ45 Sahx sp / Betula nana leaves CAMS-73417 37,400±930 BP Log I
DSJ46 Salix sp leaves / capsules DSJ-1 Log H CAMS-73418 37,090±890 BP
LogD DSJ52 Betula nana leaves CAMS-73421 37,270±910BP
DSJ-1 LogK DSJ53 nana/ Sahx herbacea / Salix polans CAMS-73422 44,600±2260 BP
Tab. 5: Radiocarbon and OSL dates for samples from Deeping St James.
Tab 5 Datation OSL et par le carbone 14