River response to Quaternary climatic fluctuations: evidence from the Son and Belan valleys, north-central India

ARTICLE IN PRESS

Quaternary Science Reviews 25 (2006) 2619–2631

River response to Quaternary climatic fluctuations: evidence from the Son and Belan valleys, north-central India M.A.J. Williamsa,, J.N. Palb, M. Jaiswalc, A.K. Singhvic a

Geographical and Environmental Studies, University of Adelaide, Adelaide, SA 5005, Australia Department of Ancient History, Culture and Archaeology, University of Allahabad, Allahabad 211002, India c Planetary and Geosciences Division, Physical Research Laboratory, Navrangpura, Ahmedabad 380009, India

b

Received 10 April 2005; accepted 26 July 2005

Abstract The last glacial period was cold and dry in peninsular India. In north-central India, the interval from 3979 to 1673 ka was associated with widespread and prolonged aggradation in the Son and Belan valleys. The aggradation ended with sustained vertical incision after 1673 ka and reflects a return to warmer, wetter conditions. In this region, it would appear that terminal Pleistocene to Holocene river incision was broadly synchronous with a strong summer monsoon regime and higher levels of river discharge and the preceding river aggradation with lower discharge and a weaker or more variable summer monsoon regime. Two older phases of prolonged aggradation followed by vertical incision are evident in the Son and Belan valleys before
39 ka. One of these phases is centred towards 7374 ka when ash from the Toba mega-eruption in Sumatra was deposited across peninsular India. The following phase of aggradation has yielded infrared stimulated luminescence ages of 5876 and 45712 ka. The youngest phase of aggradation began towards
5.5 ka and seems to mark a return to a weaker summer monsoon regime. r 2006 Elsevier Ltd. All rights reserved.

1. Introduction The alluvial sediments of the Ganga, Yamuna, Son and Belan valleys of north-central India (Figs. 1 and 2) have been the focus of archaeological investigations over the past three decades (Sharma, 1973; Sharma et al., 1980; Sharma and Clark, 1983; Misra and Pal, 2002). A near complete prehistoric cultural sequence ranging from early Acheulian through Middle and Upper Palaeolithic to Mesolithic and Neolithic has been identified for this region and stands as an important regional datum for Indian prehistoric archaeology. The chronology of the cultural sequence from the late Upper Palaeolithic onwards is based on a growing number of radiocarbon dates, summarised in a later section. In an attempt to extend the chronology further, we have begun a systematic program of infrared stimulated luminescence (IRSL) dating of the alluvial sediments associated with faunal and other prehistoric cultural remains. Corresponding author. Fax: +61 8 8303 3772.

E-mail address: [email protected] (M.A.J. Williams). 0277-3791/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.quascirev.2005.07.018

The aim of this paper is to present a more detailed chronology for the alluvial deposits and prehistoric cultures of this region and to propose a working hypothesis to account for the repetitive pattern of fluviatile aggradation and incision during the late Quaternary. Since the region is well upstream of any possible sea level influence, it is buffered from the impact of glacio-eustatic fluctuations in base level. The alignment of the Son valley is controlled by the NarmadaSon tectonic lineament that is still prone to seismic activity (Vita-Finzi, 2004). However, within the study area shown in Figs. 2a–d, the surfaces of successive late Quaternary alluvial terraces are broadly parallel over distances of several hundred kilometres, indicating that the terraces were formed during a prolonged interval of tectonic stability. Any changes in river sedimentation and incision must, therefore, reflect local changes in plant cover and in the load to discharge ratios that themselves reflect fluctuations in regional climate. 2. The Quaternary alluvial formations of the Son valley Initial archaeological enquiry in the Son and Belan valleys may have unduly emphasised the association

ARTICLE IN PRESS 2620

M.A.J. Williams et al. / Quaternary Science Reviews 25 (2006) 2619–2631

Fig. 1. World map showing domain of the monsoon (adapted from Griffin, 2002), location of the study area in north-central India (black square) and regions influenced by the Southern Oscillation (adapted from Whetton et al., 1990). Correlations (40.4) of annual surface atmospheric pressure anomalies with those in Jakarta, Indonesia: stippled is positive correlation; hatched is negative correlation.

between various gravel horizons and the stone artefact assemblages, almost invariably reworked, that they contained. The joint investigations co-directed by the late Professors G.R. Sharma and J. Desmond Clark in 1980 and 1982 introduced a new approach to archaeological research in this region, with an emphasis on the recovery of primary context undisturbed cultural remains (Sharma and Clark, 1983; Clark and Williams, 1986, 1990). As a means of achieving this, the late Quaternary sedimentary sequence in the middle Son valley was re-examined and four widespread alluvial formations were identified and mapped (Williams and Royce, 1983; Williams and Clarke, 1984, 1995). Each stratigraphic formation was named after the village nearest to the respective type section. This summary is based on the observations of Williams and Royce (1983) and Williams and Clarke (1995) supplemented by additional field observations by MAJW during January 2004 in which a fifth formation was described and sampled. Starting with the oldest, the formations are: 1. Sihawal Formation; 2. Khunteli Formation; 3. Patpara Formation;

4. Baghor Formation; 5. Khetaunhi Formation. Fig. 2 shows the locations from where samples were collected for dating and the locations of all sections discussed in the text. Fig. 3 shows the stratigraphic relations between these alluvial formations. The Sihawal Formation rests directly on bedrock and consists of alluvial fan and debris-flow gravels in a clay matrix overlain by reworked aeolian clay. The next three formations consist of alluvial sands, clays and gravels and abut against the older formations with marked unconformities. The youngest unit comprises alluvial sands and clays and forms the lowest terrace in the landscape. 2.1. Sihawal Formation The Sihawal Formation consists of two units: a basal gravel unit with Lower Palaeolithic evolved Acheulian biface tools and flakes and an upper unit of very well sorted silty clay entirely devoid of both gravel and artefacts. The artefacts are concentrated in the upper part of the gravelly clay and immediately beneath the silty clay. The Sihawal

ARTICLE IN PRESS M.A.J. Williams et al. / Quaternary Science Reviews 25 (2006) 2619–2631

2621

Fig. 2. (a) Location of the study area, with inset map of India. (b) Locations from where samples were collected for radiocarbon and IRSL dating in the Belan valley and (c) in the Son valley.

Fig. 3. Block diagram showing the five major alluvial formations investigated in the middle Son valley.

ARTICLE IN PRESS 2622

M.A.J. Williams et al. / Quaternary Science Reviews 25 (2006) 2619–2631

Formation has a maximum observed thickness of about 1.5 m, and rests unconformably on the underlying Proterozoic bedrock. The gravels are poorly sorted, ranging in size from coarse sand to cobbles with some boulders up to 50 cm in diameter, set in a matrix of grey and brown silty to fine sandy clay. They comprise a variety of locally derived lithologies (quartzite, sandstone, mudstone) and are aligned parallel to the bedrock surface with the long axis of the clasts oriented downslope. Some are matrix supported and appear to be debris-flow deposits; others, less angular, weakly bedded and better sorted, appear to be alluvial fan deposits. The overlying grey and brown mottled silty clay is massive and structureless, and appears to be a wind-blown dust that has been partially reworked by local runoff. 2.2. Khunteli Formation A previously undescribed formation crops out on the right bank of the Son near the village of Khunteli (Fig. 2c). This formation post-dates the Sihawal Formation and predates the Patpara Formation. The formation comprises a basal unit up to 6 m thick of pale yellow-brown unconsolidated medium sand, a discontinuous bed of pure volcanic ash up to 1.5 m thick, at least 4 m of cross-bedded and planar-bedded medium and coarse sands and fine gravels and an upper unit of carbonate cemented gravels. The 30 m thick section on the left bank of the river Son near its confluence with the river Rehi is a lateral variant of this formation and consists of a locally obscured basal unit up to 8 m thick of brown medium sand capped by a laterally discontinuous unit of pure volcanic ash up to 4 m thick (Williams and Royce, 1983). The ash forms a channel-fill and is locally eroded and partly replaced by several lenses of alluvial gravel. Above the gravels there are up to 20 m of inter-bedded clays, sands and rolled carbonate gravels (Fig. 4). The volcanic ash was the first Quaternary ash discovered in India. The ash was erupted from Toba volcano in Sumatra 7374 ka ago (mean 40Ar/39Ar age, Chesner et al., 1991) and is widespread throughout India (Shane et al., 1995; Westgate et al., 1998). 2.3. Patpara Formation The Patpara Formation is at least 10 m thick and unconformably overlies the Sihawal Formation (Fig. 3). It has a characteristic reddish colour and contains rounded to sub-angular clasts of quartz, sandstone and mudstone similar to the underlying bedrock, as well as abundant agate, chalcedony and other microcrystalline silicic rocks. These far-travelled clasts originated from the Deccan Trap basalts in the headwaters of the Son. Clast size ranges from coarse sand to cobbles but with 1–2 cm pebbles most common, set in a clay-rich matrix. Bedding varies from massive to flat or undulating laminations. The formation is partly cemented by iron, giving the sediments their diagnostic reddish colour. In places, the formation is

Fig. 4. Stratigraphic section RS1 (for location see Fig. 2a). The volcanic ash from the 7374 ka eruption of Toba volcano in Sumatra here occupies a former tributary channel. The entire sequence belongs within the Late Pleistocene Khunteli Formation.

overlain by up to 1 m of hard, dark reddish-brown, mottled silty clay, unless truncated by the overlying Baghor Formation. Fresh and abraded Middle Palaeolithic core and flake tools are ubiquitous throughout this formation. 2.4. Baghor Formation The Baghor Formation is up to 20 m thick and is separated from the underlying Patpara Formation by an erosional unconformity (Fig. 3). It comprises two distinct members, each about 10 m thick: a lower coarse member and an upper fine member (Figs. 5 and 6). The coarse member is made up of cosets of unconsolidated crossbedded medium to coarse sands with individual sets 5–85 cm thick and usually bounded by planar horizontal surfaces. Foreset dips are uniformally eastwards, i.e., down-valley. The sands are dominantly quartz and are poorly sorted. Lenses of granule to pebble-sized sandstone, mudstone, quartzite, chalcedony, agate and chert commonly occur as pebble trains on scour surfaces between cross-beds. Discontinuous sheets of massive carbonate cemented sands are common throughout the coarse member and contain well-preserved fossils, including buffalo, hippo, crocodile, antelope, elephant and tortoise, as well as rolled and abraded Middle Palaeolithic artefacts.

ARTICLE IN PRESS M.A.J. Williams et al. / Quaternary Science Reviews 25 (2006) 2619–2631

2623

the modern Son may overtop the terrace, so that sedimentation is still active. Along the right bank opposite Sihawal village, the beds dip 10–151 towards the present river and have very well-preserved sedimentary structures indicating recent, relatively rapid deposition of the beds. This was very evident during January 2004, although future major floods may well erode these beds. 3. The Quaternary alluvial formations of the Belan valley The alluvial sequence identified for the Belan valley is based on logging and sampling eight major geological step trenches supplemented by additional sampling of five other sections for radiocarbon (14C) and IRSL dating (Fig. 2d). The provisional lithostratigraphic sequence for the middle Belan valley is that described by Williams and Clarke (1995) and is summarised below.

Fig. 5. Stratigraphic section S1 showing Middle Pleistocene Sihawal Formation (S) resting on Lower Proterozoic shales (for location see Fig. 2c). The clays and loams of the Late Pleistocene Baghor Formation fine member (BF) lie unconformably above the eroded surface of the Sihawal Formation. Inverted black triangles denote stone artefacts.

The fine member forms the highest aggradational surface in the Son valley, with an elevation of at least 30 m above low water level, and rests conformably on the coarse member with no evidence of an erosional break. It consists of inter-bedded silts, clays and occasional fine sands. Each bed is 1–4 m thick and may be traced laterally for 2–3 km. Any primary depositional structures have been modified by soil-forming processes, and irregular carbonate nodules, tubules and plates occur in well-defined layers. The concretions become more concentrated towards the top of the fine member, indicating progressively longer intervals with minimal flood-plain sedimentation and more prolonged intervals of soil formation. Fresh Upper Palaeolithic artefacts are found in situ within the silts and clays in the upper few metres of the fine member as well as on the surface. 2.5. Khetaunhi Formation The Khetaunhi Formation is up to 10 m thick and forms a well-defined alluvial terrace made up of inter-bedded silts, clays and fine sands (Fig. 3). The silts and clays display laterally uniform parallel lamination and the fine sands display ripples and ripple cross lamination. During flood

(a) Tabular sandstone conglomerate over Upper Vindhyan (Upper Proterozoic) sandstone bedrock. Lower Palaeolithic artefacts. (b) Calcareous brown clay loam. (c) Planar and cross-bedded sandstone gravels. Middle Palaeolithic artefacts. (d) Reddish-brown sandy clays and clay loams. (e) Shell-bearing gravels, mostly rolled carbonate nodules and black ironstone pisolites. Upper Palaeolithic artefacts. (f) Brown and yellow-brown calcareous clay loams and sandy clays. Minor intra-formational gravels, mainly rolled carbonate nodules. Upper Palaeolithic to EpiPalaeolithic and Mesolithic artefacts. (g) Holocene clays, loams and fine alluvial sands. Neolithic and Proto-historic artefacts. Correlation with the Son valley alluvial sequence is based on similarities in lithology and prehistoric artefact assemblages checked against the radiocarbon and IRSL ages discussed later in this paper. The Sihawal Formation is equivalent to Belan units (a) and (b); the Khunteli Formation to Belan unit (c); the Patpara Formation to Belan units (d); the Baghor Formation to Belan units (e) and (f); and the Khetaunhi Formation to Belan unit (g). 4. Quaternary alluvial history of the Son and Belan valleys Before reviewing the new IRSL chronology now available for selected localities in the Son and Belan valleys, a brief overview of the Quaternary alluvial history of each valley will provide a useful context within which to consider the luminescence ages. 4.1. Alluvial history of the Son (a) Bedrock erosion and pedimentation of the Lower Vindyan (Lower Proterozoic) metasediments (sandstones, siltstones,

ARTICLE IN PRESS 2624

M.A.J. Williams et al. / Quaternary Science Reviews 25 (2006) 2619–2631

Fig. 6. Generalised stratigraphic section near Sihawal in the middle Son valley showing details of the Baghor 2 (B2) section (see Fig. 2c for location). The section shows the Middle Pleistocene Sihawal Formation (S) at the base, overlain unconformably by the Late Pleistocene Baghor Formation (B). ESA denotes Early Stone Age or Lower Palaeolithic. The microlithic cultures in this area immediately pre-date the Neolithic.

limestones), probably during the Middle Pleistocene. Deposition of alluvial fan gravels and gravelly clays. Minor reworking of the gravels by the palaeo-Son close to present low river-level. Accumulation of Lower Palaeolithic flakes, cores and bifaces of quartzose sandstone and quartzite on and in these Sihawal Formation gravels. Middle Pleistocene. (b) Deposition of fine sandy clay wind-blown dust during and after accumulation of Sihawal Formation gravels. Middle Pleistocene. (c) Erosion followed by deposition of alluvial sands, gravels, clays and loams of the Khunteli Formation. Tributary channels choked with volcanic ash. Ashfilled channels preserved from subsequent erosion by avulsion and continuing sediment accumulation. Early Upper Pleistocene. (d) Erosion followed by deposition of Patpara Formation gravely clays and fluvial sands. Transported Lower Palaeolithic and relatively undisturbed Middle Palaeolithic artefacts interstratified among both fine and coarse facies of the formation. Syn-depositional or

post-depositional reddening of clays and sands. Upper Pleistocene. (e) Erosion followed by deposition of Baghor Formation channel sands, overbank clays and yellow-brown wind-blown dust. Dust mantle reworked by slopewash. Middle to Upper Palaeolithic artefacts. Upper Pleistocene. (f) Several intervals of episodic downcutting, lateral planation and erosional terrace formation. Final phase of accumulation of alluvial silts, clays and fine sands of the Khetaunhi Formation as an inset aggradational terrace. Epi-Palaeolithic, Mesolithic and Neolithic artefacts. Upper Pleistocene to Holocene.

4.2. Alluvial history of the Belan (a) Erosion of Upper Vindhyan (Upper Proterozoic) sandstone bedrock. Accumulation of locally derived sandstone gravels. Lower Palaeolithic artefacts. Middle Pleistocene.

ARTICLE IN PRESS M.A.J. Williams et al. / Quaternary Science Reviews 25 (2006) 2619–2631

(b) Deposition of calcareous wind-blown dust. Middle Pleistocene. (c) Erosion followed by deposition of alluvial gravels. Accumulation of Middle Palaeolithic artefacts. Late Middle to early Upper Pleistocene. (d) Deposition of sandy clays and clay loams. Weathering and rubefaction of clays and loams. Upper Pleistocene. (e) Erosion followed by accumulation of shell-bearing fluvial gravels. Upper Palaeolithic artefacts. Late Upper Pleistocene. (f) Accumulation of calcareous loams, sandy clays and intra-formational fine gravels. Upper Palaeolithic to Epi-Palaeolithic and Mesolithic artefacts. Late Upper Pleistocene. (g) River incision followed by accumulation of sandy alluvial bench to +11 m above present low water level. Deposition of alluvial clays, loams and fine sands. Mesolithic, Neolithic and Proto-historic artefacts. Holocene. (h) Continued incision. Holocene.

5. IRSL chronology for the Son and Belan valleys 5.1. Luminescence dating method Samples BN-1, BN-2, BN-3 and S-1 were coarse grained (Table 1). These samples were sequentially pretreated with 10% HCl and 30% H2O2 to remove carbonates and organic matter and then sieved to obtain the 90–150 mm grain-size fraction. Density separation using Na-polytungstate (r ¼ 2.58 g/cm3) was carried out to separate quartz and feldspar minerals. The analysis was done on the feldspar extract of the sediment. These grains were mounted on stainless-steel discs with the help of SilkosprayTM silicone oil for analysing. The choice of feldspar grains was dictated by (a) a rather low yield of quartz in the grains of interest, (b) anticipation of a complex luminescence dose growth curve in the samples of higher antiquity (i.e., higher palaeodose) and (c) the ease of comparison with fine grain samples where also the feldspar IRSL was

2625

probed. The samples from silt-dominated units (B-1, B-3, B-5, B-7 and B-8) were processed separately to isolate the 4–11 mm particles. In this case, samples after treatment by HCl and H2O2 were deflocculated in 0.1 N sodium oxalate followed by washing and then suspended in a 6 cm column of acetone for separation of the 4–11 mm fine silt fraction using appropriate Stokes’ settling times. IRSL analysis was carried out on all the samples using infrared diodes (l ¼ 880780 nm) assembly, integral to a Daybreak 1150 reader. The detection optics comprised Corning 7–59 and Schott BG-39 filters. Beta-irradiation was effected separately using a 90Sr/90Y source having a dose rate of 2.44 Gy/min for polymineral fine grains and 3.65 Gy/min for the coarse feldspar grains. Alpha-irradiation was performed using americium-241 to compute the avalue. The Multiple Aliquot Additive dose method with late light subtraction was used to compute the palaeodose. A preheat of 220 1C for 60 s was used to remove any unstable signal. For dose rates, uranium (238U) and thorium (232Th) concentrations were measured by thick source alpha counting using ZnS (Ag) screens on alpha counters 583 from Daybreak. Potassium (40K) concentration was measured by gamma spectrometry using NaI (Tl) crystal. The overall consistency of dose rates through the profiles indicates that the decay series were not in gross disequilibrium. The results are presented in Table 1 and examples of IRSL shine down and growth curves are shown in Fig. 7. There have been concerns about IRSL ages being underestimated due to athermal or long-term fading (Lamothe et al., 1994; Lamothe and Auclair, 1997; Huntley and Lamothe, 2001). In the present study, no explicit testing of long-term fading was carried out; however, if such fading existed the ages would be underestimated. Some suggestion that the fading may not be significant comes from a comparison of calibrated radiocarbon ages and the present IRSL ages. Thus, e.g., the OSL age of 1673 ka accords well with the suite of radiocarbon ages on charcoal averaging
15.2 ka cal BP. While the error margins are large, the fact that the IRSL age corresponds to the higher side of a simple mean of the calibrated radiocarbon

Table 1 Infrared stimulated luminescence ages of samples collected from the Son and Belan valleys Sample

Grain size (mm)

U (ppm)

Th (ppm)

K (%)

a-Value

Palaeodose (Gy)

Dose rate (Gy/ka)

Age (ka)

BN-1 BN-2 BN-3 S-1 B-1 B-3 B-5 B-7 B-8

90–150 90–150 90–150 90–150 4–11 4–11 4–11 4–11 4–11

2.870.5 571.4 1.170.3 3.370.8 571.3 2.571.3 1.770.8 6.271.2 2.671.3

9.371.9 45.278.6 3.270.9 15.972.9 9.274.5 13.673.1 16.974.6 975 12.373.1

1.970.1 0.670.03 0.570.02 2.270.11 1.070.05 2.270.1 2.170.1 2.170.1 1.870.1

NA NA NA NA 0.10 0.06 0.05 0.06 0.05

7276 14578 59714 270716 373750 213730 178743 9578 5774

3.870.3 670.9 1.570.1 4.670.4 4.170.8 4.170.6 3.970.5 4.770.7 3.670.5

1972 2473 3979 5876 90720 52710 45712 2073 1673

We assume 2075% and 1075% moisture content for fine grain (4–11 mm) and coarse grain (90–150 mm) samples, respectively. Samples BN-1 to BN-3 were from Baghor Nala, S-1 from Sihawal and B-1, B-3, B-5, B-7 and B-8 from the Belan valley main section. B-1 was below the gravel, B-3 at a depth of 14.9 m, B-5 at 10.9 m, B-7 at 5.9 m and B-8 at 0.3 m below the surface. Palaeodose values in decimals have been rounded off.

ARTICLE IN PRESS M.A.J. Williams et al. / Quaternary Science Reviews 25 (2006) 2619–2631

2626

Sample - BN-1

4.5

8

N+460 Gy N+292 Gy

1.5

N=Natural

N

0 0

100

200

300

Sample - BN-1 100

Paleodose (Gy)

Sample-BN-1

3.0

Normalized Luminescence counts (x100)

Luminescence counts (x100)

6.0

6

4

2

De = 7 2 + 6 Gy 50 De

0

115

230

345

460

Laboratory beta dose (Gy)

0 0

200

100 Time (seconds)

Fig. 7. Typical IRSL shine down curves, growth curves and shine plateau for samples analysed.

ages suggests that the IRSL ages may not be seriously affected by long-term fading. This aspect is being examined further based on red thermoluminescence from volcanic ashes and sediments immediately above and beneath these ashes collected from the Son valley. It is suggested that while the ages cited here may possibly be underestimated, the overall inferences are unlikely to change. 5.2. IRSL ages for the Son valley sediments Sample S-1 comes from the Patpara Formation at Sihawal and is IRSL dated to
5876 ka (Table 1). Three samples were collected from the Baghor Formation at Baghor Nala (Fig. 2c). Sample BN-3 from the lower part of the coarse member gave an age of
3979 ka, sample BN-2 from the middle part of the coarse member is dated to
2473 ka and sample BN-1 from the fine member has an age of
1972 ka (Table 1). We have no IRSL ages for the Khetaunhi Formation but the Neolithic site of Kunjhun located on the Khetaunhi alluvial terrace has calibrated radiocarbon ages between
6.3 and
3.3 ka (Table 2).

Fig. 8 summarises the lithostratigraphic sequence in the Son valley and associated IRSL and radiocarbon ages. 5.3. IRSL ages for the Belan valley sediments Sample B-1 was taken from the Belan ‘Main Section’ (Williams and Clarke, 1995) and samples B-3, B-5, B-7 and B-8 from a section 100 m to the north east on the right bank of the river (Fig. 2d). Sample B-1 gave an IRSL age of 90720 ka (Table 1) and correlates with the late Lower Palaeolithic (Sharma and Clark, 1983). Samples B-3 and B-5 taken from Belan unit (d) gave ages of 52710 and 45712 ka, respectively (Table 1). These luminescence ages correlate with the Middle Palaeolithic (Fig. 9). Sample B-7 from unit (e) and sample B-8 from unit (f) had IRSL ages of 2073 and 1673 ka, respectively (Table 1). These IRSL ages correlate with the Upper Palaeolithic. These ages accord well with the four calibrated radiocarbon dates already obtained from unit (e) that range in age from 29 to 20 ka (Sharma and Clark, 1983; Williams and Clarke, 1995) (Table 2). Fig. 10 summarises the lithostratigraphic

ARTICLE IN PRESS M.A.J. Williams et al. / Quaternary Science Reviews 25 (2006) 2619–2631

2627

Table 2 Radiocarbon dates for shells, charcoal and carbonate from the Son, Belan and Yamuna valleys, India Locality

Lab no.

Material

Yamuna

Beta 4788

Shell

Belan

PRL 409 PRL 408 PRL 407 BAI/MGR77-1 PRL 101 PRL 100 PRL 224 SUA 142 PRL 602 PRL 603 Beta 4789 TF 1245 Beta 4877 PRL 86 Beta 4790

Charcoal Charcoal Charcoal Charcoal Charcoal Charcoal Charcoal Shell Charcoal Charcoal Shell Shell Shell Shell Shell

Son

Beta 4879 Beta 6414 Beta 6415 SUA 1422 Beta 4792 SUA 1420 Beta 4791 Beta 4793

Shell Charcoal Shell Carbonate* Shell Carbonate* Shell Shell

Prehistoric culture

14

C date (a BP)

14

C cal. age ka (2s range)

15 5407540

19 850–17 150

Neolithic Neolithic Neolithic Neolithic Neolithic Neolithic Neolithic Microlithic Epi-Palaeolithic Epi-Palaeolithic Upper Palaeolithic Upper Palaeolithic Upper Palaeolithic Upper Palaeolithic Upper Palaeolithic

33507150 32807120 33907100 34307110 64807185 73907240 85207210 10 0307115 11 3007130 14 1407400 18 0557150 19 7157340 25 4307350 25 7907795 26 2707660

4000–3200 3850–3250 3890–3400 4000–3400 7700–6950 8750–7650 10 250–8950 12 000–11 200 13 400–12 930 18 550–15 750 22 050–20 850 24 450–22 550
30 000–28 000
30 000–28 000
31 000–29 000

Neolithic Neolithic Neolithic

3215770 41307110 4740780 5305790 11 8707120 13 1457140 20 1357220 26 2507420

3630–3260 5000–4250 5610–5310 6280–5910 14 000–13 400 16 100–15 100 24 750–23 450
31 000–29 000

Microlithic Upper Palaeolithic Upper Palaeolithic

*Denotes post-depositional, hence minimum age only. The radiocarbon calibration curve used was that of Reimer et al. (2004) using OxCal version 3.10 (Bronk Ramsey, 2001). The error term for PRL 603 was +410/390, giving a mean of 7400, used in the calibration; the error term for PRL 86 was +830/760, giving a mean of 7795. Approximate calibration of ages over 24 000 years were obtained from the calibration curve of Hughen et al. (2004).

sequence in the Belan valley and associated IRSL and radiocarbon ages. Sample B-8 came from a gravel and sand sub-unit within Belan unit (f) and gave an IRSL age of 1673 ka (Table 1). A radiocarbon age of 13.4–12.9 cal ka (PRL 602) was previously obtained from this unit (Williams and Clarke, 1984) (Table 2). The unit contains Upper Palaeolithic artefacts and Epi-Palaeolithic non-geometric microliths (Table 3). 6. Quaternary environments in the Son and Belan valleys The prolonged interval of Late Pleistocene aggradation in the study area that began towards 3979 ka and terminated towards 1673 ka is of particular interest and may be compared and correlated with events outside India. The interval leading up to and encompassing the Last Glacial Maximum was cold, dry and windy in India and indeed elsewhere in the intertropical world (Williams, 1975, 1985; Prell et al., 1980; Cullen, 1981; Sarnthein et al., 1981; Duplessy, 1982; Hoelzmann et al., 2004). Global warming and a return to wetter conditions in the tropical and sub-tropical regions of Asia, Africa and Australia succeeded this long interval of glacial aridity (Williams et al., 1998; Gasse, 2000; Gagan et al., 2004; Fleitmann et al., 2004; Gasse and Roberts, 2004; Gagan

and Thompson, 2004). The rapid fluvial incision that began soon after 1673 ka also marks the return of the summer monsoon throughout the region depicted in Fig. 1. Marine cores from the Arabian Sea show the onset of strong monsoon circulation in that region
14 ka (Sirocko, 1996). Similar incision into Late Pleistocene valley-fill deposits is also evident at about this time in the southern Gangetic plains of northern India (Gibling et al., 2005) and as far away as the Flinders Ranges of South Australia where it also marks the return of the summer monsoon (Williams et al., 2001). A major increase in the discharge of the White Nile began at
15 ka (Williams et al., 2000) and reflects a significant increase in summer rainfall. The entire region presently under the influence of the summer monsoon (Fig. 1) was generally colder and drier for at least 5 ka before
15 ka and was considerably wetter after that. Fig. 1 also shows the region presently influenced by the Southern Oscillation. Within this region, which includes India, the rivers are strongly influenced by ENSO events both in terms of discharge and in terms of their patterns of sedimentation (Adamson et al., 1987; Whetton et al., 1990; Markgraf and Diaz, 2000; Whitaker et al., 2001). The sedimentary record from a lake in southern Ecuador provides a Holocene record of El Nin˜o events and like the Peruvian record (Sandweiss et al., 2001; Andrus et al., 2002) shows a sudden increase in the frequency of El Nin˜o

ARTICLE IN PRESS 2628

M.A.J. Williams et al. / Quaternary Science Reviews 25 (2006) 2619–2631

Fig. 8. Generalised stratigraphic section from the Son valley showing major depositional units and associated IRSL and calibrated radiocarbon ages. Unit HT* is a Holocene terrace which has been dated between 5.6–5.3 and 3.6–3.3 ka.

events towards 4.8 ka (Rodbell et al., 1999; Moy et al., 2002). Although there is no direct evidence from India, it seems very likely that El Nin˜o events affecting India were also rare during the early Holocene and more frequent from mid-Holocene times onwards. The mid-Holocene aggradation that gave rise to the youngest alluvial terrace in the Belan and Son valleys appears to be coeval with the well-documented global increase in the frequency of El Nin˜o events at this time. In the Son and Belan valleys, widespread aggradation between
39 and
15 ka coincided at least in part with a time of greater cold and drier climate with a much weakened summer monsoon regime. A phase of lesser aggradation that began towards 5 ka also marks a return to a somewhat weaker and more variable summer rainfall regime. In this region, it would appear that terminal Pleistocene to Holocene river incision was broadly synchronous with a strong summer monsoon regime and higher levels of river discharge (Williams and Clarke, 1984,

Fig. 9. Generalised stratigraphic section from the Belan valley showing major depositional units and associated IRSL and calibrated 14C ages. Unit HT* is a Holocene terrace which has been dated at 4.0–3.2 ka. Fig. 8.

1995; Gibling et al., 2005; Juyal et al., this issue) and the preceding river aggradation with lower discharge and a weaker or more variable summer monsoon regime. If this interpretation is correct, then the intervals of erosion between each of the major depositional phases evident in the Son and Belan valleys may coincide very broadly with times of wetter climate and greater fluvial discharge. The converse may equally apply, with valley aggradation reflecting a return to progressively drier conditions, reduced fluvial discharge, and an increase in the ratio of load to discharge. Gross correlations of sedimentary records elsewhere in the Indian sub-continent indicate that Marine Isotope Stage 3 (59–24 ka) was generally humid (although drier episodes did occur) and MIS 2 (24–12 ka) was mainly arid. Sedimentary records from calcretes in aeolian contexts

ARTICLE IN PRESS M.A.J. Williams et al. / Quaternary Science Reviews 25 (2006) 2619–2631

(Andrews et al., 1998; Chamyal et al., 2003), from the Gangetic plains (Srivastava et al., 2003), from the Sabarmati basin (Tandon et al., 1997) and from the Mahi basin (Juyal et al., 2000) all indicate a humid, monsoon-like phase akin to the present in India during MIS 3. Although the chronometric control in these studies is not yet tightly secured, it seems that sediment aggradation/pedogenesis occurred during the arid phase of MIS 4 and was followed by a humid phase during MIS 3. Additional evidence accrues from the Bay of Bengal, where Goodbred (2000, 2003) has summarised the record from the river Ganges since the last interstadial and documented changes in river

2629

discharge—a surrogate for monsoon rainfall, since meltwater flux was a minor fraction of discharge after 58 ka. These inferred changes in discharge accord with the changes in the Son and Belan valleys described above. More recent work by Kudrass et al. (2001) in the Bay of Bengal documents high frequency variation in the monsoon during the past 80 ka and suggest that a more rigorous sediment characterisation, high-resolution chronometry and climate reconstruction may provide a better understanding of monsoon variation in time and space. During the Holocene, full monsoon conditions in the Thar Desert of Rajasthan were achieved by 7 ka and lakes there began drying out towards 5.5–5 ka (Enzel et al., 1999; Thomas et al., 1999; Deotare et al., 2004). Pollen evidence from the central higher Himalaya shows a sharp decrease in summer monsoon strength at 4–3.5 ka (Phadtare, 2000). Archaeologically, it may prove no coincidence that the transition from Lower to Middle Palaeolithic and from Middle to Upper Palaeolithic both appear to be times of transition from a stronger summer monsoon to a weaker, more variable summer rainfall regime (Fig. 10; Table 3). It is also tempting to speculate that the transition from Mesolithic to Neolithic may have been stimulated by the end of the early Holocene wet interval and by the onset of a weaker, more variable summer monsoon regime.

7. Conclusions

Fig. 10. Major phases of aggradation (dark grey) in the Son and Belan valleys of north-central India in relation to global marine isotope stages and sub-stages (light grey) and to palaeoproductivity changes evident in marine core 70 KL from the western Arabian Sea (after Hoelzmann et al., 2004, Fig. 4).

The Son and Belan rivers in north-central India are deeply incised into Late Pleistocene valley-fill deposits for which IRSL and radiocarbon chronometric age controls are now available. An early phase of local alluvial fan and debris-flow activity ended with the accumulation of a thin mantle of aeolian dust dated by IRSL to 90720 ka and marks the end of Lower Palaeolithic settlement in this region. The oldest of three widespread valley fills contains very pure volcanic ash derived from the Toba eruption at
7374 ka. Sediments within the penultimate valley fill have yielded IRSL ages of 5876 and 45712 ka. The most recent and widespread of the valley fills began to accumulate towards 3979 ka and continued to accumulate until about 1673 ka. The aggradation was associated with a progressively weaker summer monsoon and ended with sustained vertical incision soon after 1673 ka

Table 3 Stratigraphic units and ages of associated prehistoric cultures in the Belan and Son valleys (ka is 1000 years) Sihawal Formation Khunteli Formation Patpara Formation Baghor Formation Baghor Formation Khetaunhi Formation

Belan Belan Belan Belan Belan Belan Belan

units (a), (b) unit (c) unit (d) unit (e) unit (f) unit (f) unit (g)

Lower Palaeolithic Middle Palaeolithic Middle Palaeolithic Upper Palaeolithic Epi-Palaeolithic Microlithic Neolithic

X90720
73 ka 5876, 45712 ka 3979, 1972 ka
18.5 to
13 ka
14 to
11 ka X5.5 to
3.5 ka

ARTICLE IN PRESS 2630

M.A.J. Williams et al. / Quaternary Science Reviews 25 (2006) 2619–2631

that reflects a return to warmer and wetter conditions and a stronger summer monsoon. The long interval of valley aggradation between 3979 and 1673 ka was coeval with semi-nomadic Upper Palaeolithic occupation of this region, and the penultimate phase of aggradation synchronous with the presence of Middle Palaeolithic hunter–gather societies within the valleys. The transition from Middle to Upper Palaeolithic cultures in this region appears to coincide with the transition from a strong and reliable summer monsoon regime to a weaker, more erratic summer monsoon regime. The youngest phase of aggradation began towards 5.5 ka and seems to reflect a return to a weaker, more variable summer monsoon regime in the second half of the Holocene. The interval of more erratic summer monsoon that began in the mid-Holocene also marks a change from more intensified trading activities among migratory Mesolithic populations early in the Holocene to the advent of food producing economies based upon plant and animal domestication and a sedentary home base.

Acknowledgements This paper is offered as a salute to Emeritus Professor John Prescott on the occasion of his eightieth birthday. His pioneering work on the physics of luminescence dating resonates to this day. The paper is also offered in revered memory of Professor G.R. Sharma and Professor Desmond Clark, doyens of archaeological research in northcentral India, who inspired a team of outstanding young archaeologists to pursue the route they pioneered so well. We thank the Australian Research Council and Adelaide University for financial support. MW is grateful to the villagers of the Son and Belan valleys for help during fieldwork and especially to his friends and colleagues Umesh Chattopadhyaya, B.B. Misra, V.D. Misra, D. Mandel and Prakash Sinha (University of Allahabad) and Mike Clarke and Keith Royce (Macquarie University) for two memorable seasons of fieldwork: they shared the joys and hardships. Professor Desmond Clark co-directed the excavations with a lively and congenial team from the University of California at Berkeley: Rob Blumenschine, Steve Brandt, Jo-Ann Gutin, Mark Kenoyer, Gurchuran Singh and Carole Sussman.

References Adamson, D.A., Williams, M.A.J., Baxter, J.T., 1987. Complex late Quaternary alluvial history in the Nile, Murray-Darling and Ganges basins: three river systems presently linked to the Southern Oscillation. In: Gardiner, V. (Ed.), Proceedings of the First International Conference on Geomorphology, Manchester, September 1985, Part II. Wiley, Chichester, pp. 875–887. Andrews, J.E., Singhvi, A.K., Kailath, A.J., Kuhn, R., Dennis, P.F., Tandon, S.K., Dhir, R.P., 1998. Do stable isotope data from calcrete record Late Pleistocene monsoonal climate variation in the Thar Desert of India? Quaternary Research 50, 240–251.

Andrus, C.F.T., Crowe, D.E., Sandweiss, D.H., Reitz, E.J., Romanek, C.S., 2002. Otolith q18O record of mid-Holocene sea surface temperatures in Peru. Science 295, 1508–1511. Bronk Ramsey, C., 2001. Development of the radiocarbon program OxCal. Radiocarbon 43 (2A), 355–363. Chamyal, L.S., Maurya, D.M., Rachna, R., 2003. Fluvial systems of the drylands of western India: a synthesis of late Quaternary environmental and tectonic changes. Quaternary International 104, 69–86. Chesner, C.A., Rose, W.I., Deino, A., Drake, R., Westgate, J.A., 1991. Eruptive history of Earth’s largest Quaternary caldera (Toba, Indonesia) clarified. Geology 19, 200–203. Clark, J.D., Williams, M.A.J., 1986. Palaeoenvironments and prehistory in north central India: a preliminary report. In: Jacobson, J. (Ed.), Studies in the Archaeology of India and Pakistan. Oxford University Press, New Delhi, pp. 18–41. Clark, J.D., Williams, M.A.J., 1990. Prehistoric Ecology, Resource Strategies and Culture Change in the Son Valley, Northern Madhya Pradesh, Central India. Man and Environment XV (1), 120–133. Cullen, J.L., 1981. Micro-fossil evidence of changing salinity patterns in the Bay of Bengal over the last 20000 years. Palaeogeography, Palaeoclimatology, Palaeoecology 35, 315–356. Deotare, B.C., Kajale, M.D., Rajaguru, S.N., Kusumgar, S., Jull, A.J.T., Donahue, J.D., 2004. Palaeoenvironmental history of Bap-Malar and Kanod playas of western Rajasthan, Thar Desert. Proceedings of the Indian Academy of Sciences: Earth and Planetary Sciences 113 (3), 403–425. Duplessy, J.C., 1982. Glacial to interglacial contrasts in the northern Indian Ocean. Nature 295, 494–498. Enzel, Y., Ely, L.L., Mishra, S., Ramesh, R., Amit, R., Lazar, B., Rajaguru, S.N., Baker, V.R., Sandler, A., 1999. High resolution Holocene environmental changes in the Thar Desert, northwestern India. Science 284, 125–128. Fleitmann, D., Burns, S.J., Neff, U., Mudelsee, M., Mangini, A., Kramers, J., Matter, A., 2004. Holocene records of rainfall variation and associated ITCZ migration from stalagmites from northern and southern Oman. In: Diaz, H.F., Bradley, R.S. (Eds.), The Hadley Circulation: Present, Past and Future. Kluwer Academic Publishers, Dordrecht, pp. 259–287. Gagan, M.K., Thompson, L.G., 2004. Evolution of the Indo-Pacific Warm Pool and Hadley-Walker circulation since the last deglaciation. In: Diaz, H.F., Bradley, R.S. (Eds.), The Hadley Circulation: Present, Past and Future. Kluwer Academic Publishers, Dordrecht, pp. 289–312. Gagan, M.K., Hendy, E.J., Haberle, S.G., Hantoro, W.S., 2004. Postglacial evolution of the Indo-Pacific Warm Pool and El Nin˜o-Southern Oscillation. Quaternary International 118–119, 127–143. Gasse, F., 2000. Hydrological changes in the African tropics since the Last Glacial Maximum. Quaternary Science Reviews 19, 189–211. Gasse, F., Roberts, C.N., 2004. Late Quaternary hydrologic changes in the arid and semiarid belt of northern Africa. In: Diaz, H.F., Bradley, R.S. (Eds.), The Hadley Circulation: Present, Past and Future. Kluwer Academic Publishers, Dordrecht, pp. 313–345. Gibling, M.R., Tandon, S.K., Sinha, R., Jain, M., 2005. Discontinuitybounded alluvial sequences of the southern Gangetic plains, India: aggradation and degradation in response to monsoonal strength. Journal of Sedimentary Research 75 (3), 373–389. Goodbred Jr., S.L., 2000. Enormous Ganges–Brahmaputra sediment discharge during strengthened early Holocene monsoon. Geology 28, 1083–1086. Goodbred Jr., S.L., 2003. Response of Ganges dispersal system to climatic change: a source to sink view since the last interstade. Sedimentary Geology 162, 83–104. Griffin, D.L., 2002. Aridity and humidity: two aspects of the late Miocene climate of North Africa and the Mediterranean. Palaeogeography, Palaeoclimatology, Palaeoecology 182, 65–91. Hoelzmann, P., Gasse, F., Dupont, L.M., Salzmann, U., Staubwasser, M., Leuschner, D., Sirocko, F., 2004. Palaeoenvironmental changes in the arid and subarid belt (Sahara–Sahel–Arabian Peninsula) from 150 kyr

ARTICLE IN PRESS M.A.J. Williams et al. / Quaternary Science Reviews 25 (2006) 2619–2631 to present. In: Battarbee, R.W., Gasse, F., Stickley, C.E. (Eds.), Past Climate Variability through Europe and Africa. Springer, Dordrecht, pp. 219–256. Hughen, K., Lehman, S., Southon, J., Overpeck, J., Marchal, O., Herring, C., Turnbull, J., 2004. 14C activity and global carbon cycle changes over the past 50,000 years. Science 303, 202–207. Huntley, D.J., Lamothe, M., 2001. Ubiquity of anomalous fading in Kfeldspar and the measurement and correction for it in optical dating. Canadian Journal of Earth Sciences 38, 1093–1106. Juyal, N., Raj, R., Maurya, D.M., Chamyal, L.S., Singhvi, A.K., 2000. Chronology of Late Pleistocene environmental changes in the lower Mahi basin, western India. Journal of Quaternary Science 15, 501–508. Juyal, N., Chamyal, L.S., Bhandari, S., Bhandari, S., Singhvi, A.K., 2006. Continental record of SW monsoon during the last 130 ka: evidence from the southern margin of Thar Desert, India. Quaternary Science Reviews 25, doi:10.1016/j.quascirev.2005.07.020. Kudrass, H.R., Hofmann, A., Doose, H., Emeis, K., Erlenkkeuser, H., 2001. Modulation and amplification of climatic changes in the northern hemisphere by the Indian summer monsoon during the past 80 ky. Geology 29, 63–66. Lamothe, M., Auclair, M., 1997. Assessing the datability of young sediments by IRSL using an intrinsic laboratory protocol. Radiation Measurements 27, 107–117. Lamothe, M., Balescu, S., Auclair, M., 1994. Natural IRSL intensities and apparent luminescence ages of single feldspar grains extracted from partially bleached sediments. Radiation Measurements 23, 555–561. Markgraf, V., Diaz, H.F., 2000. The past ENSO record: a synthesis. In: Diaz, H.F., Markgraf, V. (Eds.), El Nin˜o and the Southern Oscillation. Multiscale Variability and Global and Regional Impacts. Cambridge University Press, Cambridge, pp. 465–488. Misra, V.D., Pal, J.N. (Eds.), 2002. Mesolithic India. Pushp Prakashan, Allahabad. Moy, C.M., Seltzer, G.O., Rodbell, D.T., Anderson, D.M., 2002. Variability of El Nin˜o-Southern Oscillation activity at millennial timescales during the Holocene epoch. Nature 420, 162–165. Phadtare, N.R., 2000. Sharp decrease in summer monsoon strength 4000–3500 cal yr BP. in the central higher Himalaya of India based on pollen evidence from alpine peat. Quaternary Research 53, 122–129. Prell, W.L., Hutson, W.H., Williams, D.F., Be, A.W.H., Geitzenauer, K., Molfino, B., 1980. Surface circulation of the Indian Ocean during the Last Glacial Maximum, approximately 18 000 yr BP. Quaternary Research 14, 309–336. Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Bertrand, C., Blackwell, P.G., Buck, C.E., Cutler, K.B., Damon, P.E., Edwards, R.L., Fairbanks, R.G., Friedrich, M., Guilderson, T.P., Hughes, K.A., Kromer, B., McCormac, F.G., Manning, S., Bronk Ramsey, C., Reimer, R.W., Remmele, S., Southon, J.R., Stuiver, M., Talamo, S., Taylor, F.W., van der Plicht, J., Weyhenmeyer, C.E., 2004. IntCal04 terrestrial radiocarbon age calibration, 0–26 cal kyr BP. Radiocarbon 46, 1029–1058. Rodbell, D.T., Seltzer, G.O., Anderson, D.M., Abbott, M.B., Enfield, D.B., Newman, J.H., 1999. An
15,000-year record of El Nin˜o-driven alluviation in southwestern Ecuador. Science 283, 516–520. Sandweiss, D.H., Maasch, K.A., Burger, R.L., Richardson III, J.B., Rollins, H.B., Clement, A., 2001. Variation in Holocene El Nin˜o frequencies: climate records and cultural consequences in ancient Peru. Geological Society of America Bulletin 29 (7), 603–606. Sarnthein, M., Tetzlaff, G., Koopmann, B., Wolter, K., Pflaumann, U., 1981. Glacial and interglacial wind regimes over the eastern subtropical Atlantic and North-West Africa. Nature 293, 193–196. Shane, P., Westgate, J., Williams, M., Korisettar, R., 1995. New geochemical evidence for the Youngest Toba Tuff in India. Quaternary Research 44, 200–204.

2631

Sharma, G.R., 1973. Stone Age in the Vindhyas and The Ganga Valley. In: Agrawal, D.P., Ghosh, A. (Eds.), Radiocarbon and Indian Archaeology. TIFR, Bombay. Sharma, G.R., Clark, J.D., 1983. Palaeoenvironments and Prehistory in the Middle Son Valley, Madhya Pradesh, Central India. Abinash Prakashan, Allahabad, 320pp. Sharma, G.R., Misra, V.D., Mandal, D., Misra, B.B., Pal, J.N., 1980. Beginnings of Agriculture. Abinash Prakashan, Allahabad, 200pp. Sirocko, F., 1996. The evolution of the monsoon climate over the Arabian Sea during the last 24,000 years. Palaeoecology of Africa 24, 53–69. Srivastava, P., Singh, I.B., Sharma, M., Singhvi, A.K., 2003. Luminescence chronometry and Late Quaternary geomorphic history of the Ganga Plains, India. Palaeogeography, Palaeoclimatology, Palaeoecology 197, 15–41. Tandon, S.K., Sareen, B.K., Rao, M.S., Singhvi, A.K., 1997. Aggradation history and luminescence chronology of late Quaternary semi-arid sequences of the Sabarmati basin, Gujarat, western India. Palaeogeography, Palaeoclimatology, Palaeoecology 128, 339–357. Thomas, J.V., Kar, A., Kailath, A.J., Juyal, N., Rajaguru, S.N., Singhvi, A.K., 1999. Late Pleistocene–Holocene history of aeolian accumulation in the Thar Desert, India. Zeitschrift fu¨r Geomorphologie 116, 181–194. Vita-Finzi, C., 2004. Buckle-controlled seismogenic faulting in peninsular India. Quaternary Science Reviews 23, 2405–2412. Westgate, J.A., Shane, P.A.R., Pearce, N.J.G., Perkins, W.T., Korisettar, R., Chesner, C.A., Williams, M.A.J., Acharyya, S.K., 1998. All Toba tephra occurrences across peninsula India belong to 75 ka eruption. Quaternary Research 50, 107–112. Whetton, P., Adamson, D., Williams, M.A.J., 1990. Rainfall and river flow variability in Africa, Australia and East Asia linked to El Nin˜oSouthern Oscillation events. In: Bishop, P. (Ed.), Lessons for Human Survival: Nature’s Record from the Quaternary. Geological Society of Australia Symposium Proceedings, vol. 1, pp. 71–82. Whitaker, D.W., Wasimi, S.A., Islam, S., 2001. The El Nin˜o-Southern Oscillation and long-range forecasting of flows in the Ganges. International Journal of Climatology 21, 77–87. Williams, M.A.J., 1975. Late Pleistocene tropical aridity synchronous in both hemispheres? Nature 253, 616–617. Williams, M.A.J., 1985. Pleistocene aridity in tropical Africa, Australia and India. In: Douglas, I., Spencer, T. (Eds.), Environmental Change and Tropical Geomorphology. George Allen & Unwin, London, pp. 219–233. Williams, M.A.J., Clarke, M.F., 1984. Late Quaternary environments in north-central India. Nature 308, 633–635. Williams, M.A.J., Clarke, M.F., 1995. Quaternary geology and prehistoric environments in the Son and Belan Valleys, north central India. Memoir 32, Geological Society of India, pp. 282–308. Williams, M.A.J., Royce, K., 1982. Quaternary geology of the Middle Son Valley, north central India: implications for prehistoric archaeology. Palaeogeography, Palaeoclimatology, Palaeoecology 38, 139–162. Williams, M.A.J., Royce, K., 1983. Alluvial History of the Middle Son Valley, north central India. In: Sharma, G.R., Clark, J.D. (Eds.), Palaeoenvironments and Prehistory in the Middle Son Valley, Madhya Pradesh, Central India. Abinash Prakashan, Allahabad, pp. 9–23. Williams, M., Dunkerley, D., De Deckker, P., Kershaw, P., Chappell, J., 1998. Quaternary Environments, second ed. Arnold, London, 329pp. Williams, M.A.J., Adamson, D., Cock, B., McEvedy, R., 2000. Late Quaternary environments in the White Nile region, Sudan. Global and Planetary Change 26, 305–316. Williams, M., Prescott, J.R., Chappell, J., Adamson, D., Cock, B., Walker, K., Gell, P., 2001. The enigma of a Late Pleistocene wetland in the Flinders Ranges, South Australia. Quaternary International 83–85, 129–144.