Grain size distribution analysis of sediments containing Younger Toba Tephra from Ghoghara, Middle Son valley, India more

Accepted Manuscript Grain size distribution analysis of sediments containing Younger Toba Tephra from Ghoghara, Middle Son valley, India Laura Lewis, Peter Ditchfield, J.N. Pal, Michael Petraglia PII: DOI: Reference: To appear in: S1040-6182(11)00675-6 10.1016/j.quaint.2011.12.002 JQI 3092 Quaternary International Received Date: 22 May 2011 Revised Date: 30 November 2011 Accepted Date: 1 December 2011 Please cite this article as: Lewis, L., Ditchfield, P., Pal, J.N., Petraglia, M., Grain size distribution analysis of sediments containing Younger Toba Tephra from Ghoghara, Middle Son valley, India, Quaternary International (2011), doi: 10.1016/j.quaint.2011.12.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Manuscript Click here to view linked References ACCEPTED MANUSCRIPT Grain size distribution analysis of sediments containing Younger Toba Tephra from Ghoghara, Middle Son valley, India Laura Lewis a,*, Peter Ditchfield a, J. N. Pal b and Michael Petraglia a a Keywords: Toba, paleoenvironments, sedimentology, India 1. Introduction The eruption of the Toba super-volcano in northern Sumatra ~74 ka was the largest volcanic eruption of the Quaternary period (Ninkovich et al., 1978a, b). It is classified as an 8.8 magnitude eruption (Mason et al., 2004) and has been estimated to have produced at least 2800 km3 DRE (Dense Rock Equivalent) of pyroclastic material (Rose and Chesner, 1987), with pyroclastic flows covering an area of 20-30,000 km2 (Ninkovich et al., 1978b). Tephra from the eruption is found across the Indian Ocean, the Indian peninsula, Malaysia and the Arabian and South China Seas (Westgate et al., 1998; Jones, 2007a). The Younger Toba Tuff (YTT) has been geochemically characterised at locations across the whole of peninsular India (e.g. Acharyya and Basu, 1993; Shane et al., 1995; Westgate et al., 1998). Rose and Chesner (1987) estimate an ash blanket covering at least 4,000,000 km2, or ~0.8% of the world‟s surface. Chesner et al. (1991) give a best age estimate of 74 ± 2 ka for the Toba eruption, based on 40A/39A ages and Ninkovich et al.‟s (1978b) K-Ar dates. This also lies within the age range of 71,100 ± 5000 years ago given by Zielinski et al. (1996) based on the chronology of the GISP2 ice-core. Researchers have hypothesized that the Toba eruption had a severe impact on human populations. Ambrose (1998) contends that the Toba eruption was the cause of a severe human genetic bottleneck, with humans reduced to a breeding population of a few thousand. Additionally, some have argued on the basis of climatic data that the Toba super-eruption induced a „volcanic winter‟, reducing surface temperatures on a global scale (Rampino and Self, 1992, 1993). Williams et al. (2009) have more recently conducted paleoenvironmental AC C EP TE D M AN U 1 Abstract The Toba super-eruption in northern Sumatra ~74 ka was the largest eruption of the Quaternary period. Terrestrial deposits of distal Toba tephra have been found across the Indian subcontinent, although few localities have been adequately described with respect to stratigraphy and sedimentary history. This study provides the first detailed description of the Ghoghara 1A ash section, Middle Son Valley, India. The sediment analyses provide insights into the depositional processes in the valley before and after the initial period of ash deposition. Grain size distribution analysis of sediments from Ghoghara 1A indicates that 4-5 cm of primary air-fall ash was laid down at the time of the Toba super-eruption. This study has implications concerning previous isotopic studies in the Middle Son Valley, concluding that reconstructions demonstrative of dramatic ecological changes are questionable. SC RI PT School of Archaeology, University of Oxford, Oxford, OX1 2PG, UK Department of Ancient History, Culture and Archaeology, University of Allahabad, Allahabad, Uttar Pradesh, India *Corresponding author. Email address: laura.lewis@arch.ox.ac.uk b ACCEPTED MANUSCRIPT research in India which indicates that the Toba eruption resulted in a period of prolonged climatic cooling, increased aridity and deforestation in terrestrial environments in India. Based on archaeological evidence, it has been argued that human populations inhabiting the Indian subcontinent prior to the Toba eruption survived through the event (Petraglia et al., 2007, this volume), though certain populations may have been affected (Jones, 2007a, 2007b). Middle Paleolithic stone tools found above and below the YTT at Jwalapuram Locality 3 in the Jurreru river valley, northern India, exhibit a strong degree of technological continuity, suggesting that the same populations inhabited the area before and after the supereruption (Haslam et al., 2010; Clarkson et al., this volume). The Middle Son valley, India, is one of the most important terrestrial sites for the investigation of the Toba super-eruption, with a long history of archaeological and geological research (e.g. Williams and Royce, 1982; Sharma and Clark, 1983). The area was the place of the first discovery of Toba tephra in India (Williams and Royce, 1982), later characterised as YTT (Rose and Chesner, 1987; Petraglia et al., 2007; Smith et al. 2011). Cultural and paleoenvironmental reconstructions often depend upon an assumption that the Toba tephra layer corresponds very closely in time with the 74 ka eruption. For example, Williams et al.‟s (2009) analysis of the carbon isotopes of carbonate nodules immediately above and below the Toba ash at three Indian sites, including two in the Son valley, assumes that the reworked ash can be used as an highly precise temporal marker of environments. However, previous studies – including those by Williams et al. (2006, 2009, 2010) – lack detailed chronometric, stratigraphic and sedimentological analysis, focussing instead on valley-wide stratigraphic reconstructions. In the authors‟ view, it is also necessary to examine the position of the Toba ash through detailed field observations and sedimentological analyses. Jones (2010) addresses this issue specifically in her study of sediments from the Jurreru and Middle Son valleys, which combines sedimentological, granulometric and stratigraphical analyses. This study builds on and extends Jones‟ work in the Middle Son Valley, using a much longer and more intensely sampled section from a new trench in the Rehi-Ghoghara cliff section, providing the first characterisation of such a long sedimentary sequence containing the YTT. This study utilises grain size analysis of sediments from the site of Ghoghara 1A (GG1A) in order to characterise the ash deposit and to determine sedimentary environments before, during and after its deposition. The results have implications concerning the Williams et al. (2009) carbon isotope study, which indicates that the Middle Son Valley witnessed dramatic changes in environments soon after the Toba eruption, changing from forests to grassland habitats. 2. Regional setting 2.1. Quaternary environments of the Middle Son valley The middle reaches of the Son river flow eastward across the state of Madhya Pradesh, central India, bounded on the north by the Vindhya and Kaimur ranges and on the south by the Baghelkhand plateau (Williams and Royce, 1982; Jones, 2010). The Middle Son valley contains extensive Quaternary alluvial deposits, and archaeological and geological research has been carried out here since the 1960s. The valley has a long history of hominin occupation, containing archaeological sites from the Lower Palaeolithic to the Neolithic periods (Sharma and Clark, 1983). The first description of the geology and sedimentology of the valley, including the first discovery of Toba tephra in India, was published by Williams 2 AC C EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT and Royce in 1982. The tephra occurs in the valley as a discontinuous layer over a distance of ~30 km (Gatti et al., in press). Figure 1 illustrates the location of archaeological sites and YTT exposures in the Middle Son valley. Williams and Royce (1982) divided the geological deposits of the Middle Son valley into four Quaternary formations – the Sihawal, Patpara, Baghor (itself divided into a lower coarse member and an upper fine member) and Khetaunhi, from oldest to youngest. Williams et al. (2006) added a fifth formation to this sequence between the Sihawal and Patpara formations; the Khunteli formation. There is some debate over the location of the YTT tephra within this stratigraphy. On the basis of recent fieldwork, Jones (2007a) advises that the YTT deposit at Ghoghara lies between the Patpara formation and the Baghor coarse member, while others have argued that that the YTT lies within the Baghor coarse member (Williams and Royce, 1982; Acharyya and Basu, 1993). Williams et al. (2006) place the YTT tephra within their newly-described Khunteli formation. However, Jones and Pal (2009, pp. 325) remark that this new formation is „problematic‟, having only been described at two sections within the valley and not being associated stratigraphically with the Sihawal or Patpara formations at either. Jones and Pal‟s (2009) analysis of lithic artifacts from the Middle Son valley includes 10 stone tools from Ghoghara main section, which they tentatively place within the middle of the Patpara formation, having been found eroding out of fluvial gravels ~3-4 m below the Toba ash. Stratigraphic integrity is moderate to poor and the artifacts exhibit varying degrees of abrasion, suggesting significant post-depositional transportation of some or all artifacts. However, new work at the nearby archaeological site of Dhaba is providing information on Late Acheulean, Middle Palaeolithic and Microlithic occupations in the valley (Haslam et al., this volume) indicating that archaeological assemblages predate and postdate the YTT event (Petraglia et al., this volume). Matthews et al. (this volume) document the nature of the primary ash layer in the Ghoghara section based on a set of criteria: the unit may have weakly-developed planar bedding but no cross-bedding features; the unit forms a continuous, uniformly-bedded deposit which 3 AC C EP Jones‟ particle size analysis includes sites in the Jurreru and Middle Son valleys, in southern and north-central India, respectively. In the Ghoghara main section, Jones reports that at least six phases of ash redeposition were evident, having been deposited into a water-filled channel. The ash in each of these redeposition events may have been transported from different source regions, and becomes progressively less pure as it moves up the sampling column. Successive periods of ash redeposition may have choked the valley. Jones concludes that in both areas the Toba ash was reworked and redeposited over a period of several years. TE D 2.2. Previous research at Ghoghara and its vicinity Subsequent to the important geological and archaeological work conducted in the Middle Son valley in the 1980s, a new programme of research was conducted more recently (Petraglia et al., this volume). As part of a wider investigation into paleoenvironmental changes in India as a response to the Toba super-eruption, Jones (2010) analysed sediments from the Ghoghara main section located a few hundred meters along the terrace cliff down-stream of the confluence of the Son and Rehi rivers on the northern bank of the river Son (Figure 1). This was the first study to investigate the impacts of the YTT ash-fall on terrestrial paleoenvironments in India using particle size, magnetic susceptibility and loss on ignition analyses of sediment and tephra samples. M AN U SC RI PT ACCEPTED MANUSCRIPT blankets pre-existing topography with only minor lateral variation in thickness locally; the unit is characterised dominantly by glass shards and an absence of non-volcanic detrital material. They observe and demonstrate the difference between the primary ash layer at Ghoghara and the deposit lying immediately above, the latter of which contains crossbedding and herringbone structures, indicating deposition in a fluvial environment. Interestingly, worm casts, escape structures and pellet structures within the cross-bedded strata suggest to the researchers that life persisted in the aftermath of Toba. In a similar vein, Gatti et al. (in press) assess the textural, sedimentological and stratigraphic characteristics of the Toba ash at several sites in the Middle Son valley, as part of their investigation into changes in the dynamic activity of the river Son. They conclude that the primary ash stratum at GG1A is unaltered. 3. Materials and methods Above the basal layer the ash is darker in colour and shows a much wider range of grain sizes probably representing reworked ash. The first 1.8 m of the reworked ash unit consists of cross-bedded layers of white to yellow-white ash and white-red to orange to brown sands, transitioning from very fine sands towards the base of the reworked ash unit to medium and coarse sands further up the profile. The top 2.4 m of the reworked ash unit consists of less pure reworked ash grading into light brown-grey clayey silt. From ~2.5 m below the top of the section, frequent calcium carbonate nodules probably of pedogenic origin occur. The uppermost 2m of the section consists of pale grey-brown silt. At a depth of 2 m from the top, potsherds and charcoal were recovered, indicating that this sediment is Holocene in age. 3.2. AC C The base of the section consists of a pebble-cobble conglomerate, possibly representing the Sihawal formation, which contains Late Acheulean implements (Petraglia et al., this volume). This is overlain by 3.5 m of brown and yellow-brown coarse and gravelly sands, which are cross-bedded and contain sub-horizontal coarse laminations. Overlying these cross-bedded sands are 65 cm of horizontally laminated yellow and brown-yellow to grey medium and coarse sands, with some laminations of orange sand. Above these sands and immediately underlying the first ash-bearing layer (the primary ash) is a 25 cm layer of light brown-yellow silty clay. There is an abrupt and well-defined boundary between this and the overlying ash unit. Figure 2C illustrates the sharp contact between the base of the ash unit and the underlying silty clay sediment. The primary ash layer is 4-5 cm thick and consists of wellsorted white ash, which shows conspicuous bioturbation. This bioturbation has the effect of contaminating bulk samples of the basal ash layer with sandy clay lithology, although this is clearly restricted to burrow and root traces and is due to the effects of bioturbation rather than whole-scale reworking of the basal layer. Grain size analysis: methodology 4 EP TE D M AN U 3.1. Ghoghara 1A stratigraphy The Ghoghara 1A section (UTM 44R, 0603086N 2710187E) was excavated as part of a new series of sections at Ghoghara (Figures 2A and 2B), several metres along the terrace cliff from the section studied by Williams et al. (2009) and Jones (2010). This is the longest and most detailed section excavated so far in the valley, measuring 11.8 m in length. It contains a 4-5 cm layer of pure ash which field observations suggest is primary (i.e. not reworked) ashfall from the Toba super-eruption. Figure 3 is a section drawing of Ghoghara 1A. SC RI PT ACCEPTED MANUSCRIPT Sampling for grain size analysis was conducted in the middle portion of the trench excavation. Sampling was conducted at evenly spaced intervals of ~5 cm below the ash (GG1A-1 to 6), at ~1 cm intervals in the primary ash layers (GG1A-12 to 16), at 3-5 cm intervals in the subsequent ashy layers (GG1A-17 to 28) and at ~10 cm intervals through the rest of the column (GG1A-29 to 61). A total of 56 sedimentology samples from a span of 4.17 m were analysed. Two additional samples from the coarse sands below the ash layers were also analysed (PE1 and PE2). The location of each of these samples can be seen in Figure 6. In order to attain a definitive grain size distribution profile of the basal ash, one additional sub-sample was taken from a micromorphology block removed from the site. This sub-sample was carefully extracted so as to avoid any contamination from possible bioturbation within the ash (examples of which can be seen in Figure 2C). Sediment samples were first lightly crushed with a pestle and mortar to remove aggregates. Particle size distributions were measured with a Malvern Mastersizer 2000 laser diffraction particle size analyser. All samples had ultrasound applied to them for 20 seconds in order to break up any remaining aggregate particles. Each sample was run three times in order to check the reliability of the measurements. The mean of each set of results was taken to calculate the values for each sample. All analyses were conducted in the School of Geography and the Environment at the University of Oxford. Samples PE1 and PE2 contained grains larger than the Malvern Mastersizer is able to measure (>2mm). The samples were instead hand-sieved using a standard sieving column with meshes ranging from 62.5 µm to 8 mm. The contents of each sieve were then weighed using scales accurate to 0.01g. For samples GG1A-1 to 61 the mean grain size, sorting, skewness and kurtosis were calculated as logarithmic graphical measures using Folk and Ward‟s 1957 formulae (Blott and Pye, 2001). Values for PE1 and PE2 were calculated by the geometric moment method due to the lower number of sieve sizes and thus measurements. 4. Results and discussion 4.1. Pre-Toba sediments and the primary ash layer Figure 5 displays the grain size distributions for the two samples taken from the lower coarse sand layers (green lines), with the three uppermost clay layers above (red lines), the primary ash layer above that (blue lines), and the significantly coarse-grained samples above that between GG1A-29 and 35 (as indicated in Figure 6C) (orange lines). The PE1 and PE2 grain size distributions indicate that the depositional conditions at this location in the valley in the period prior to the deposition of Toba ash and the immediately preceding clays was characterised by a relatively high-energy fluvial environment, involving the transport of coarse sands. Figure 5 demonstrates that that the grain size distributions for these lower coarse sand layers are very similar to those of the coarse-grained samples higher in the sampling column, which may also be indicative of similar deposition environments. AC C EP Table 1 reports the location, mean grain sizes, sorting, skewness and kurtosis of all samples, and Figures 4A and 4B plot the changes in these variables over time as sampling moves up the column. Overall the sediments are predominantly poorly sorted, symmetrical to fineskewed and mesokurtic to leptokurtic in nature. TE D M AN U 5 SC RI PT ACCEPTED MANUSCRIPT However, the lack of samples in the ~3.6 m between PE1 and PE2 currently precludes detailed assessment of the changes in depositional processes over this period. From the graph in Figure 5 two clear peaks emerge; a finer-grained peak with a mode of ~63 µm (very fine sand), and a coarse-grained peak with a mode of ~500 µm (coarse sand) (graphical estimations based on size classes). The difference between the coarse fluvial sands and the ash layers is far more pronounced than that between the clay layers and the ash layers. The processes that introduced the coarse fluvial sands would have involved higherenergy transport and deposition of larger grain sizes, via stronger flow regime within the Son River. Figure 6E, which displays the grain size distributions for the primary ash and the uppermost pre-ash sample, demonstrates the influx of a small amount of coarse-grained materials, particularly in samples GG1A-14 and 15. In GG1A-16 the grain size distribution returns to levels similar to the first ash sample (GG1A-12), with much less of a coarse tail than the underlying clay layers. Overall, the primary ash layer samples GG1A-12 to 15 exhibit a sudden and substantial shift in sorting, skewness, kurtosis and, to a lesser extent, mean grain size (Figures 4A and 4B). GG1A-12 exhibits a sharp increase in skewness value (which is not surpassed until much later in the sequence), a considerable drop in kurtosis value, and a small decrease in sorting value (i.e. slightly better-sorted than the preceding clay layers). These results might be expected for sediments deposited through aeolian transport, which would have a sharp drop-off point at the limit at which coarser grains can be transported. Additionally, as Leeder (1982) notes, wind-blown deposits are generally uniform, reflected in the reduced kurtosis value and thus reduced tail size when compared to underlying deposits. Sample GG1A-13 exhibits a small increase in kurtosis compared to GG1A-12. It also displays a sharp increase in sorting and a sharp decrease in skewness. This trend is continued in GG1A-14 and 15, which are considerably less well sorted than all other samples and have the lowest kurtosis values of all samples, with the only platykurtic distributions, i.e. flatter broader granulometry profiles with fewer extreme values. This is a result of the bimodal distributions of these two samples, with the addition of a peak of coarser-grained materials which is reflected in the increase in mean grain size (Figure 4B). The bimodality of these distributions can be seen clearly in Figure 6E. Sample GG1A-14 has the lowest skewness value of all samples and is the only coarseskewed distribution, as a result of this influx of coarser-grained materials which distorts sorting, skewness and kurtosis values. Bimodality in grain size distributions can indicate a combination of transport and depositional processes. In bimodal distributions, the sorting reflects the relative magnitude and separation of the different modes, rather than the dispersion around a central tendency (Pettijohn et al., 1987). A sudden decrease in sorting value is exhibited by sample GG1A-16, indicating better-sorted sediments than even the pre-ash samples. The kurtosis value also increases, returning to a leptokurtic distribution similar to the first ash layer GG1A-12. The skewness values return to roughly-symmetrical pre-ash levels. Overall this indicates a sudden shift to conditions more in keeping with the first ash sample (GG1A-12). The rapidity of the shift in sorting, skewness and kurtosis evident in the transitions to and from the primary ash layer, in the space of ~4-5 cm, indicates a sudden and significant alteration in depositional regime. This shift is indicative of the deposition of primary air fall ash, as represented by the fine-grained peak in the granulometric profiles of these samples (Figure 6E). The sharpness of these changes 6 AC C EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT (shown clearly in Figure 2C) is not compatible with arguments invoking transport or considerable reworking of the primary ash layer, although subsequent bioturbation is possible. It is possible that the presence of a small amount of coarser-grained material within the primary ash layer may be the result of fluvial transport. However, field observations do not support this interpretation, indicating instead that the primary ash layer was introduced by aeolian transport, in accordance with Matthews et al.‟s (this volume) criteria for identifying air-fall ash. The difference between this 4-5 cm layer and the overlying ~1.8 m of ashy sediments is striking, the latter containing numerous fluvial lenses and cross-bedded sands. This demonstrates deposition in a fluvial environment, evidence for which is absent in the 4-5 cm primary ash layer. Any coarser-grained materials in the primary ash layer are therefore likely due to bioturbation processes after laying down of the ash, visible as small intrusions which are signs of root action or burrowing invertebrates (Figure 2C). Figure 7 displays the grain size distribution of a sample taken from the primary ash layer so as to avoid any possible intrusions or influences from bioturbation (examples of which can be seen in Figure 2C). In comparison with Figures 6D and 6E it reveals that when the whitest and purest ash is intentionally extracted a very clear peak in grain size distribution is revealed, with a mean grain size of 50µm and a total absence of coarser-grained sediments. The results of this analysis indicate that pockets of coarser-grained materials, as a result of bioturbation, can be easily identified as discrete occurrences, and that coarser grains have not been mixed in with the primary deposition ash. This supports the argument that the primary ash layer is not reworked. A series of rapid and substantial oscillations in sediment composition occurs between GG1A28 and GG1A-41 (Figures 4A, 4B, 6A and 6C). This section of the sample column is characterised by extreme shifts between moderately- and well-sorted medium sands (average mean grain size 295 µm) and poorly-sorted very fine sands (average mean grain size 67 µm). In this river valley, the appearance of significantly coarser-grained sands is indicative of a change in depositional regime to a higher-energy depositional environment with increased fluvial action resulting in the transport of larger grains. Figure 6C demonstrates the distinction between the fine (red) and coarse (blue) grain size peaks in samples GG1A-28 to 40. The distinctiveness of these peaks can be seen in the bimodality of several of these samples. This graph indicates that two different sets of transport and depositional processes were at work during this period. The fine-peaked 7 AC C 4.2. Post-Toba depositional processes Figure 6D compares grain size distributions between the ash-bearing samples from GG1A-12 to 28. With the exception of GG1A-25 which appears to contain an influx of some coarsergrained materials, the samples overlying the primary ash samples (GG1A-12 to 15) all have similar poorly-sorted mostly leptokurtic distributions. This is unexpected given field observations that the ash content of the sediments decreases over this period. Figure 4B also indicates that, again with the exception of GG1A-25, mean grain sizes remain stable from GG1A-16 to 28. A possibility is that the valley was choked with reworked ash to such an extent that the fluvial-deposition system was affected and fluvial activity was reduced, resulting in deposition of few coarse-grained sands. A similar conclusion was drawn for Toba ash in the Jurreru river valley (Jones, 2010). EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT distributions are indicative of low-energy depositional environments, and the coarse-peaked distributions of high-energy ones, possibly indicating that the latter distributions represent an influx of coarser-grained materials from the surrounding area as the result of different fluvial regimes. Jones (2010) relates this to monsoonal processes in the Jurreru river valley, although there is no further evidence to support this at present in Ghoghara, and rates of deposition remain unknown. The continued presence of finer-grained material suggests that the ash remained in the landscape throughout this period, as is also apparent from field observations. Williams et al. report the following results from the carbon isotope work at Ghoghara: low δ13C values (-10.2 to -10.39%) below the ash indicate C3 forests immediately before the Toba eruption; high values (-3.97 to -4.99%) within the upper part of the ash indicate C4 grasslands to wooded grasslands in post-Toba environments; extremely high values (-0.7 to 0.5%, within a range of -5.35 to 0.50%) at the top of the reworked ash indicate that pure C4 grassland “persisted for a long time”; and low to intermediate values (-9.77 to -4.30%) above the ash indicate that a mosaic of forest to wooded grassland habitats eventually returned to the area (Williams et al., 2009, pp. 301-302). The implication is that Toba ash was deposited on a landscape covered by C3 plants (forests) and then replaced by mainly C4 grasslands to wooded grasslands. The identification of nearly pure C4 grassland at the top of the ash was considered to be consistent with, and represent, the cold temperatures during the ~1800 years that followed Toba (i.e. D-O stadial 20) (Williams et al., 2009, pp. 302). 8 AC C Williams et al. sampled four proveniences at Ghoghara: 1) below the ash, 2) within the reworked ash, 3) the topmost surface of the reworked ash, and 4) above the reworked ash. The samples below the ash were from a layer they describe as a 25 cm-thick clayey silt. From within the overlying 4 m thick deposit of reworked ash, samples were from a 8 cmthick level (176-184 cm above the base of ash), which they describe as an „ashy silt loam‟. The samples at the top of the reworked ash were collected from a calcretized horizon measuring about 11 cm in thickness (i.e. 382-393 cm above the base of ash). The samples from above the reworked ash were in dark brown well-developed vertic paleosols, extending 82 cm above the reworked ash (412-482 cm above the base of ash) and noted to be conformably overlying the reworked ash (Williams et al., 2009, pp. 300). EP TE D M AN U 4.3 Evaluation of Williams et al.’s Ghoghara isotope study In their carbon isotope and pollen studies of terrestrial and marine settings, Williams and colleagues (2009) concluded that the Toba super-eruption caused climatic cooling and led to a prolonged drought and period of deforestation in South Asia. Two of the three terrestrial localities selected for environmental reconstruction were from locations in the Middle Son Valley; Khuteli and Rehi, the latter of which is referred to as Ghoghara in this study and others (Jones, 2010; Matthews et al., this volume). The stratigraphic observations and sedimentological analyses at Ghoghara are not wholly consistent with the interpretations made by Williams et al. (2009). In this light, it is worthwhile to evaluate the underlying basis of Williams et al.‟s findings for dramatic and large-scale paleoenvironmental changes in the Son Valley. SC RI PT The lack of fine-coarse oscillations in samples in the uppermost portion of the column, from GG1A-41 to 61, may be indicative of reduced fluvial activity at this locality. Figure 6B demonstrates the lack of distinct finer- and coarser-grained peaks in grain size in these samples. A possible explanation here is that this is a consequence of local variation in the depositional facies due to migration of the active channel belt of the fluvial system. ACCEPTED MANUSCRIPT Given the potential importance of these interpretations some assumptions and weaknesses in the Williams et al. model should be highlighted on the basis of the authors‟ observations at Ghoghara. 1. With respect to pre-Toba environments, it is clear that their 25 cm of clayey silt corresponds with the 25 cm silty clay. The finding that carbon isotopes demonstrate forested habitats immediately before the Toba eruption is somewhat problematic. The silty clays are not an in situ soil horizon, but sediments that are deposited by fluvial processes behind the main channel. The carbonate nodules in the deposit may therefore be mixed and perhaps transported, thus not necessarily representative of a local ecological setting immediately before the Toba eruption. Even accepting the presence of forests, this finding is not particularly surprising given that the samples derive from the fringes of the Son River, which would likely contain woodland habitats. 2. Williams et al. assume that the age of the overlying 4 m of reworked ash represents wooded grassland environments soon after the Toba eruption, asserting that the top-most reworked ash deposit and its pure grassland settings represents D-O stadial 20 (Williams et al. 2009, pp. 302). In the absence of high precision chronometric ages from the section, this temporal assignation remains speculative. Indeed, different temporal interpretations are possible, and Jones (2007b, 2010) argues that the entire 4 m tephra deposit at Ghoghara (including the topmost ash) probably accumulated within weeks to several decades (based on historical eruptions) and possibly over several years (based on her sedimentary studies). Therefore, it is difficult to determine if the reworked ash was deposited over days, weeks, years, decades, or perhaps many hundreds of years or more. In the absence of very high resolution chronometric ages and stratigraphic levels, the 4 m of reworked ash could represent any particular time interval, and that any interpretation invoking D-O stadial 20 is speculative. Though the precise time interval for the reworked ash is difficult to estimate, it is notable that the reworked ash section at Ghoghara is comprised of cross bedded sands within the reworked ash deposit. This indicates the presence of flowing water n the valley, in contrast with Williams et al.‟s interpretation for high aridity and prolonged drought following the volcanic event. 3. Williams et al. found evidence for wooded grasslands and grasslands in the reworked ash and pure grasslands in the top 30 cm of the profile. Notwithstanding the chronological problems as laid out above, Haslam and Petraglia (2010) suggest that the physical presence of the thick ash deposit itself may account for the presence of grasses. That is, the growth of grassy woodlands and grasslands is to be expected from observations of plant community response to deep tephra cover, as grasses and small trees are early colonising taxa. Thus, invoking widespread climatic deterioration from terrestrial settings is not necessary to explain the higher C4 signature within the ash profile. 4. Carbonate samples selected from proveniences ranging up to 80 cm above the reworked ash were found to represent mosaic grassland to woodland conditions (Williams et al., 2009). The age of their dark brown vertic paleosols (the brown-grey clayey silt) suggests the possibility for a young age to this deposit. Potsherds and charcoal were collected 60 cm above the reworked ash, which appears to overlap with the Williams et al. sampling zone. The age of their carbonate samples from this zone is therefore questionable, including the 9 AC C EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT level between the reworked ash and the pottery, which they interpret to represent post-Toba environments following on from D-O stadial 20. The Ghoghara section was one of several terrestrial and marine settings selected by Williams et al. On the basis of study of this key section, the authors suggest the need for more careful stratigraphic observations and sampling strategies in order to reconstruct paleonvironments. Similar assumptions and problems found at Ghoghara underlie Williams et al.‟s findings at Hirapur and Khuteli. More research is obviously needed to understand the complex relationship between the Toba super-eruption, global cooling and the range of terrestrial settings and environments in India. 5. Conclusions Acknowledgements The Archaeological Survey of India granted permission to conduct this research and the American Institute of Indian Studies provided logistical assistance. For fieldwork and discussions we wish to thank Nicole Boivin, Allan Chivas, A.K. Dubey, Adam Durant, Emma Gatti, M. C. Gupta, Michael Haslam, Sacha Jones, Naomi Matthews, Christina Neudorf, H. Prasad, Bert Roberts and Ceri Shipton. This research was funded by major grants from the British Academy and the Leverhulme Trust (to MP) and the Arts and Humanities Research Council (to LL). AC C The landscape and environment of the Middle Son valley would undoubtedly have been altered by the deposition of large quantities of Toba ash, with ash remaining mobile in the landscape for a period of time. However, field observations and granulometric analysis indicates that at some point after the deposition of the primary air-fall YTT, the ash began to be flushed out of the river valley system, with evidence for the transport of coarser river sands and the return of a higher-energy fluvial deposition environment. These observations accord more closely with the hypothesis that the post-Toba environment at Ghoghara would have been potentially viable for hominin populations, in contrast to other studies which hypothesize adaptive impacts (Jones, this volume) and catastrophic ecological and evolutionary changes (Williams et al., 2009). EP TE D M AN U 10 The grain size distribution analysis presented in this paper provides a more detailed characterisation of the depositional processes and environments at Ghoghara over a longer period than was possible from previously excavated sections in the area. It also provides more information about a YTT ash deposit in context. The results of the granulometric analysis, combined with field observations concerning the abrupt and well-defined contacts between the primary ash layer and the underlying and overlying sediments, bolster the argument that the initial 4-5 cm of the ash unit represents primary un-reworked ash. In accordance with Matthews et al.‟s (this volume) criteria for identifying primary ash fall-out units, it is argued that the first 4-5 cm of the YTT deposits are more likely to be primary airfall ash rather than washed in from the local environment. In contrast, the overlying ~4.2 m of the ash unit was reworked, with evidence for structures and cross-bedding as a result of deposition and transport in a fluvial environment. SC RI PT ACCEPTED MANUSCRIPT References Acharyya, S.K. and Basu, P.K. 1993. Toba ash on the Indian subcontinent and its implications for correlation of Late Pleistocene alluvium. Quaternary Research 40, 10-19 Ambrose, S.H. 1998. Late Pleistocene human population bottlenecks, volcanic winter, and differentiation of modern humans. Journal of Human Evolution 34, 623-651 Blott, S.J. and Pye, K. 2001. GRADISTAT: a grain size distribution and statistics package for the analysis of unconsolidated sediments. Earth Surface Processes and Landforms 26, 12371248 Chesner, C.A., Rose, W.I., Deino, A., Drake, R. and Westgate, J.A. 1991. Eruptive history of earth‟s largest Quaternary caldera (Toba, Indonesia) clarified. Geology 19, 200-203 Clarkson, C., Harris, C. and Jones, S. 2011. Continuity and change in the lithic industries of the Jurreru Valley, India, before and after the Toba eruption. Quaternary International (2011) Gatti, E., Durant, A.J., Gibbard, P.L. and Oppenheimer, C. in press. Youngest Toba Tuff in the Son Valley, India: a weak and discontinuous stratigraphic marker. Quaternary Science Reviews (2011), doi: 10.1016/j.quascirev.2011.10.008 Haslam, M. and Petraglia, M. 2010. Comment on „Environmental impact of the 73 ka Toba super-eruption in South Asia‟ by M.A.J. Williams, S.H. Ambrose, S. van der Kaars, C. Ruehlemann, U. Chattopadhyaya, J. Pal and P.R. Chauhan [Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 295-314]. Palaeogeography, Palaeoclimatology, Palaeoecology 296, 199-203 Haslam, M., Clarkson, C., Petraglia, M., Korisettar, R., Jones, S., Shipton, C., Ditchfield, P. and Ambrose, S.H. 2010. The 74 ka Toba super-eruption and southern Indian hominins: archaeology, lithic technology and environments at Jwalapuram Locality 3. Journal of Archaeological Science 37, 3370-3384 Haslam, M., Harris, C., Pal, J.N., Shipton, C., Crowther, A., Koshy, J., Bora, J., Price, K., Dubey, A.K., Clarkson, C. and Petraglia, M. 2011. Dhaba: a new Acheulean, Middle Palaeolithic and microlithic locality in the Middle Son Valley, north-central India. Quaternary International (2011), doi: doi:10.1016/j.quaint.2011.09.007 Jones, S. 2007a. The Toba supervolcanic eruption: tephra-fall deposits in India and paleoanthropological implications. In Petraglia, M. and Allchin, B. Eds. The Evolution and History of Human Populations in South Asia: Inter-disciplinary Studies in Archaeology, Biological Anthropology, Linguistics and Genetics. Springer, Dordrecht, pp. 173-200 Jones, S. 2007b. A human catastrophe? The impact of the ∼74,000 year-old supervolcanic eruption of Toba on hominin populations in India. PhD thesis, University of Cambridge, UK Jones, S. 2010. Palaeoenvironmental response to the ~74 ka Toba ash-fall in the Jurreru and Middle Son valleys in southern and north-central India. Quaternary Research 73(2), 336-350 Jones, S. and Pal, J.N. 2009. The Palaeolithic of the Middle Son valley, north-central India: changes in hominin lithic technology and behaviour during the Upper Pleistocene. Journal of Anthropological Archaeology 28, 323-341 Leeder, M.R. 1982. Sedimentology: Process and Product. Allen & Unwin, London Mason, B.G., Pyle, D.M. and Oppenheimer, C. 2004. The size and frequency of the largest explosive eruptions on Earth. Bulletin of Volcanology 66, 735-748 AC C EP TE D M AN U 11 SC RI PT ACCEPTED MANUSCRIPT Matthews, N.E., Smith, V.C., Costa, A., Durant, A.J., Pyle, D.M. and Pearce, N.J.G. 2011. Ultra-distal tephra deposits from super-eruptions: examples from Toba, Indonesia and Taupo Volcanic Zone, New Zealand. Quaternary International 246, doi:10.1016/j.quaint.2011.07.010 Ninkovich, D., Sparks, R.S.J. and Ledbetter, M.T. 1978a. The exceptional magnitude and intensity of the Toba eruption, Sumatra: an example of the use of deep-sea tephra layers as a geological tool. Bulletin Volcanologique 41, 1-13 Petraglia, M. D., Ditchfield, P., Jones, S., Korisettar, R. and Pal, J.N. 2011. The Toba volcanic super-eruption, environmental change, and hominin occupation history in India over the last 140,000 years. Quaternary International (2011), doi:10.1016/j.quaint.2011.07.042 Rampino, M.R. and Self, S. 1992. Volcanic winter and accelerated glaciation following the Toba super-eruption. Nature 359, 50-52 Rampino, M. R. and Self, S. 1993. Climate-volcanism feedback and the Toba eruption of ~74,000 years ago. Quaternary Research 40, 269-280 Shane P., Westgate J., Williams M. and Korisettar R. 1995. New geochemical evidence for the youngest Toba tuff in India. Quaternary Research 44: 200-204 Sharma, G.R. and Clark, J.D. 1983. Palaeoenvironments and prehistory in the Middle Son Valley. Abinash Prakashan, Allahabad Smith, V., Pearce, N.J.G., Matthews, N., Westgate, J.A., Petraglia, M.D., Haslam, M., Lane, C.S., Korisettar, R. and Pal, J.N. 2011. Geochemical fingerprinting of the widespread Toba tephra using biotite compositions. Quaternary International, 246, doi:10.1016/j.quaint.2011.05.012. Westgate, J.A., Shane, P.A.R., Pearce, N.J.G., Perkins, W.T., Korisettar, R., Chesner, C.A., Williams, M.A.J. and Acharyya, S.K. 1998. All Toba tephra occurrences across peninsular India belong to the 75,000 yr B.P. eruption. Quaternary Research 50, 107-112 Williams, M.A.J. and 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., Pal, J.N., Jaiswal, M. and Singhvi, A.K. 2006. River response to Quaternary climatic fluctuations: evidence from the Son and Belan valleys, north-central India. Quaternary Science Reviews 25, 2619-2631 Williams, M.A.J., Ambrose, S.H., van der Kaars, S., Ruehlemann, C., Chattopadhyaya, U., Pal, J. and Chauhan, P.R. 2009. Environmental impact of the 73 ka Toba super-eruption in South Asia. Palaeogeography, Palaeoclimatology, Palaeoecology 284, 295-314 12 AC C EP TE D Rose, W.I. and Chesner, C.A. 1987. Dispersal of ash in the great Toba eruption, 75 ka. Geology 15, 913-917 M AN U Pettijohn, F.J., Potter, P.E. and Siever, R. 1987. Sand and Sandstone. 2nd ed. Springer-Verlag, New York SC Petraglia, M., Korisettar, R., Boivin, N., Clarkson, C., Ditchfield, P., Jones, S., Koshy, J., Mirazón Lahr, M., Oppenheimer, C., Pyle, D., Roberts, R., Schwenninger, J.-L., Arnold, L. and White, K. 2007. Middle Paleolithic assemblages from the Indian Subcontinent before and after the Toba super-eruption. Science 317, 114-116 RI PT Ninkovich, D., Shackleton, N.J., Abdel-Monem, A.A., Obradovich, J.D. and Izett, G. 1978b. K-Ar age of the late Pleistocene eruption of Toba, north Sumatra. Nature 276, 574-577 ACCEPTED MANUSCRIPT Williams, M.A.J., Ambrose, S.H., van der Kaars, S., Ruehlemann, C., Chattopadhyaya, U., Pal, J. and Chauhan, P.R. 2010. Reply to the comment on „Environmental impact of the 73 ka Toba super-eruption in South Asia‟ by M.A.J. Williams, S.H. Ambrose, S. van der Kaars, C. Ruehlemann, U. Chattopadhyaya, J. Pal and P.R. Chauhan [Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 295-314]. Palaeogeography, Palaeoclimatology, Palaeoecology 296, 204-211 Zielinksi, G.A., Mayewski, P.A., Meeker, L.D., Whitlow, S., Twickler, M.S. and Taylor, K. 1996. Potential atmospheric impact of the Toba mega-eruption ~71,000 years ago. Geophysical Research Letters 23, 837-840 Captions Figure 2A: ~11.8m section at Ghoghara 1A, prior to sampling. Height of scale bar: 1m. Figure 2B: Upper portion of the Ghoghara 1A section, with location of sediment samples and the white ash clearly visible. Figure 2C: Field photograph and drawing showing the basal ash layer at Ghoghara 1A. A is the underlying brown clay layer. B is the fine-grained well-sorted ash layer. C is the coarser grained and more poorly sorted reworked ash layer. D is the sandy clay filled bioturbation after roots. Figure 3: Ghoghara 1A, Middle Son valley, with sedimentary descriptions. Figure 4. A: Sorting, kurtosis and skewness values for Ghoghara 1A sediment samples. A value of 1φ was subtracted from each kurtosis value so that positive values correspond with leptokurtic distributions and negative values correspond with platykurtic distributions. B: Mean grain sizes of Ghoghara 1A sediment samples. Labels are sample numbers. Figure 5: Grain size distributions of sediment samples from Ghoghara 1A. Green lines are from the lower coarse sand layers. Red lines are from the silty clay layer underlying the ash. Blue lines are from the primary ash layer. Orange lines are from the significantly coarsegrained samples above the primary ash layer between GG1A-29 and 35. Figure 6. A: Locations and changes in proportion of sediment types of Ghoghara 1A sediment samples through the sequence, using size classes from the Udden-Wentworth scale. B-F: Grain size distributions of sediment samples from Ghoghara 1A. Figure 7: Grain size distribution of a pure ash sample from the basal (primary) ash layer. Tables Table 1: Descriptions of sediment samples from Ghoghara 1A. Sample location measurements are from base of ash to top of sample. 13 AC C EP TE D M AN U SC Figures Figure 1: Map of the locations of archaeological assemblages and YTT exposures in the Middle Son valley, and its location within India (after Jones and Pal, 2009). RI PT Figure 1 Click here to download high resolution image ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT Figure 2a (colour for web) Click here to download high resolution image ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT Figure 2a (greyscale for print) Click here to download high resolution image ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT Figure 2b (colour for web) Click here to download high resolution image ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT Figure 2b (greyscale for print) Click here to download high resolution image ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT Figure 2c (colour for web) Click here to download high resolution image ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT Figure 2c (greyscale for print) Click here to download high resolution image ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT Figure 3 Click here to download high resolution image ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT Figure 4a Click here to download high resolution image ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT Figure 4b Click here to download high resolution image ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT Figure 5 ACCEPTED MANUSCRIPT 70 60 50 PE1 PE2 GG1a-3 GG1a-2 RI PT SC M AN U 2 3.9 7.8 15.6 31 62.5 125 250 500 1000 2000 4000 8000 GG1a-1 GG1a-12 GG1a-13 GG1a-14 GG1a-15 GG1a-29 GG1a-31 GG1a-33 GG1a-34 GG1a-38 GG1a-35 Volume % 40 30 20 10 0 0.98 Particle size (µm) AC C EP TE D Figure 6 Click here to download high resolution image ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT Figure 7 Volume % 0 0.01 0.02 0.03 0.06 0.10 0.18 0.32 0.55 0.95 1.66 2.88 5.01 8.71 15.14 26.30 Particle size (µm) 1 2 3 4 5 6 AC C EP TE D 45.71 79.43 ACCEPTED MANUSCRIPT M AN U 138.04 239.88 416.87 724.44 1258.93 2000.00 SC RI PT Table 1 ACCEPTED MANUSCRIPT Sample Height above base of ash (cm) GG1A-PE1 -1082 GG1A-PE2 -722 GG1A-6 -27 GG1A-5 -22 GG1A-4 -16.5 GG1A-3 -11.5 GG1A-2 -5.5 GG1A-1 -0.5 GG1A-12 1 GG1A-13 2 a GG1A-14 3 a GG1A-15 4 GG1A-16 5.5 GG1A-17 11 GG1A-18 16 GG1A-19 21 GG1A-20 26 GG1A-21 31 GG1A-22 36 GG1A-23 40 GG1A-24 45 GG1A-25 48 GG1A-26 51 GG1A-27 54 GG1A-28 57 GG1A-29 72 GG1A-30 84 GG1A-31 94 GG1A-32 109 GG1A-33 117 GG1A-34 126 GG1A-36 134 GG1A-37 145 GG1A-38 156 GG1A-39 166 a GG1A-35 175 GG1A-40 176 GG1A-41 186 GG1A-42 198 GG1A-43 208 GG1A-44 218 GG1A-45 228 GG1A-46 238 GG1A-47 248 GG1A-48 258 GG1A-49 268 GG1A-50 278 GG1A-51 288 GG1A-52 299 GG1A-53 309 GG1A-54 319 GG1A-55 329 GG1A-56 339 GG1A-57 349 GG1A-58 359 GG1A-59 371 GG1A-60 381 GG1A-61 390 a Significantly bimodal distributions. Mean grain size (µm) 765 650 71 66 42 51 44 44 47 32 56 91 42 45 44 39 42 47 56 56 63 99 77 60 54 284 95 321 60 241 334 60 60 174 68 121 87 54 84 74 60 56 51 56 74 74 74 68 80 71 74 80 77 68 66 36 58 68 Sorting Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Very poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Very poorly sorted Very poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Moderately sorted Poorly sorted Well sorted Poorly sorted Moderately sorted Moderately sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Very poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Poorly sorted Very poorly sorted Very poorly sorted Poorly sorted Very poorly sorted Very poorly sorted Skewness Symmetrical Symmetrical Symmetrical Symmetrical Fine skewed Symmetrical Symmetrical Symmetrical Very fine skewed Symmetrical Coarse skewed Symmetrical Symmetrical Symmetrical Fine skewed Symmetrical Fine skewed Fine skewed Fine skewed Fine skewed Fine skewed Fine skewed Fine skewed Fine skewed Fine skewed Fine skewed Symmetrical Symmetrical Fine skewed Fine skewed Symmetrical Fine skewed Fine skewed Strongly fine skewed Symmetrical Symmetrical Symmetrical Fine skewed Symmetrical Symmetrical Fine skewed Fine skewed Fine skewed Fine skewed Symmetrical Symmetrical Symmetrical Fine skewed Fine skewed Symmetrical Symmetrical Symmetrical Symmetrical Symmetrical Symmetrical Fine skewed Symmetrical Symmetrical Kurtosis Leptokurtic Mesokurtic Leptokurtic Leptokurtic Very leptokurtic Very leptokurtic Very leptokurtic Very leptokurtic Leptokurtic Leptokurtic Platykurtic Platykurtic Leptokurtic Leptokurtic Leptokurtic Very leptokurtic Leptokurtic Leptokurtic Leptokurtic Mesokurtic Mesokurtic Mesokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Mesokurtic Mesokurtic Leptokurtic Very leptokurtic Leptokurtic Leptokurtic Leptokurtic Mesokurtic Leptokurtic Mesokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic Leptokurtic AC C EP TE D M AN U SC RI PT