Vegetation changes and human settlement of Easter Island during the last millennia: a multiproxy study of the Lake Raraku sediments

Earlier palynological studies of lake sediments from Easter Island suggest that the island underwent a recent and abrupt replacement of palm-dominated forests by grasslands, interpreted as a deforestation by indigenous people. However, the available evidence is inconclusive due to the existence of extended hiatuses and ambiguous chronological frameworks in most of the sedimentary sequences studied. This has given rise to an ongoing debate about the timing and causes of the assumed ecological degradation and cultural breakdown. Our multiproxy study of a core recovered from Lake Raraku highlights the vegetation dynamics and environmental shifts in the catchment and its surroundings during the late Holocene. The sequence contains shorter hiatuses than in previously recovered cores and provides a more continuous history of environmental changes. The results show a long, gradual and stepped landscape shift from palm-dominated forests to grasslands. This change started c. 450 BC and lasted about two thousand years. The presence of Verbena litoralis, a common weed, which is associated with human activities in the pollen record, the significant correlation between shifts in charcoal influx, and the dominant pollen types suggest human disturbance of the vegetation. Therefore, human settlement on the island occurred c. 450 BC, some 1500 years earlier than is assumed. Climate variability also exerted a major influence on environmental changes. Two sedimentary gaps in the record are interpreted as periods of droughts that could have prevented peat growth and favoured its erosion during the Medieval Climate Anomaly and the Little Ice Age, respectively. At c. AD 1200, the water table rose and the former Raraku mire turned into a shallow lake, suggesting higher precipitation/evaporation rates coeval with a cooler and wetter Pan-Pacific AD 1300 event. Pollen and diatom records show large vegetation changes due to human activities c. AD 1200. Other recent vegetation changes also due to human activities entail the introduction of taxa (e.g. Psidium guajava, Eucalyptus sp.) and the disappearance of indigenous plants such as Sophora toromiro during the two last centuries. Although the evidence is not conclusive, the American origin of V. litoralis re-opens the debate about the possible role of Amerindians in the human colonisation of Easter Island.

Stomatal proxy record of CO2 concentrations from the last termination suggests an important role for CO2 at climate change transitions

A new stomatal proxy-based record of CO₂ concentrations ([CO₂]), based on Betula nana (dwarf birch) leaves from the Hässeldala Port sedimentary sequence in south-eastern Sweden, is presented. The record is of high chronological resolution and spans most of Greenland Interstadial 1 (GI-1a to 1c, Allerød pollen zone), Greenland Stadial 1 (GS-1, Younger Dryas pollen zone) and the very beginning of the Holocene (Preboreal pollen zone). The record clearly demonstrates that i) [CO₂] were significantly higher than usually reported for the Last Termination and ii) the overall pattern of CO₂ evolution through the studied time period is fairly dynamic, with significant abrupt fluctuations in [CO₂] when the climate moved from interstadial to stadial state and vice versa. A new loss-on-ignition chemical record (used here as a proxy for temperature) lends independent support to the Hässeldala Port [CO₂] record. The large-amplitude fluctuations around the climate change transitions may indicate unstable climates and that " tipping-point" situations were involved in Last Termination climate evolution. The scenario presented here is in contrast to [CO₂] records reconstructed from air bubbles trapped in ice, which indicate lower concentrations and a gradual, linear increase of [CO₂] through time. The prevalent explanation for the main climate forcer during the Last Termination being ocean circulation patterns needs to re-examined, and a larger role for atmospheric [CO₂] considered.

The anatomy of Last Glacial Maximum climate variations in south Westland, New Zealand, derived from pollen records

We present pollen records from three sites in south Westland, New Zealand, that document past vegetation and inferred climate change between approximately 30,000 and 15,000 cal. yr BP. Detailed radiocarbon dating of the enclosing sediments at one of those sites, Galway tarn, provides a more robust chronology for the structure and timing of climate-induced vegetation change than has previously been possible in this region. The Kawakawa/Oruanui tephra, a key isochronous marker, affords a precise stratigraphic link across all three pollen records, while other tie points are provided by key pollen-stratigraphic changes which appear to be synchronous across all three sites. Collectively, the records show three episodes in which grassland, interpreted as indicating mostly cold subalpine to alpine conditions, was prevalent in lowland south Westland, separated by phases dominated by subalpine shrubs and montane-lowland trees, indicating milder interstadial conditions. Dating, expressed as a Bayesian-estimated single 'best' age followed in parentheses by younger/older bounds of the 95% confidence modelled age range, indicates that a cold stadial episode, whose onset was marked by replacement of woodland by grassland, occurred between 28,730 (29,390-28,500) and 25,470 (26,090-25,270) cal. yr BP (years before AD, 1950), prior to the deposition of the Kawakawa/Oruanui tephra. Milder interstadial conditions prevailed between 25,470 (26,090-25,270) and 24,400 (24,840-24,120) cal. yr BP and between 22,630 (22,930-22,340) and 21,980 (22,210-21,580) cal. yr BP, separated by a return to cold stadial conditions between 24,400 and 22,630 cal. yr BP. A final episode of grass-dominated vegetation, indicating cold stadial conditions, occurred from 21,980 (22,210-21,580) to 18,490 (18,670-17,950) cal. yr BP. The decline in grass pollen, indicating progressive climate amelioration, was well advanced by 17,370 (17,730-17,110) cal. yr BP, indicating that the onset of the termination in south Westland occurred sometime between ca 18,490 and ca 17,370 cal. yr BP. A similar general pattern of stadials and interstadials is seen, to varying degrees of resolution but generally with lesser chronological control, in many other paleoclimate proxy records from the New Zealand region. This highly resolved chronology of vegetation changes from southwestern New Zealand contributes to the examination of past climate variations in the southwest Pacific region. The stadial and interstadial episodes defined by south Westland pollen records represent notable climate variability during the latter part of the Last Glaciation. Similar climatic patterns recorded farther afield, for example from Antarctica and the Southern Ocean, imply that climate variations during the latter part of the Last Glaciation and the transition to the Holocene interglacial were inter-regionally extensive in the Southern Hemisphere and thus important to understand in detail and to place into a global context. © 2013 Elsevier Ltd. All rights reserved.

Pleistocene and Holocene glacier fluctuations upon the Kamchatka Peninsula

This review summarises landform records and published age-estimates (largely based upon tephrochronology) to provide an overview of glacier fluctuations upon the Kamchatka Peninsula during the Holocene and, to a lesser degree, earlier phases of glaciation. The evidence suggests that following deglaciation from the Last Glacial Maximum (LGM), the peninsula experienced numerous phases of small-scale glacial advance. During the Late Glacial, moraine sequences appear to reflect the former presence of extensive glaciers in some parts of the peninsula, though little chronological control is available for deposits of this period. During the Holocene, the earliest and most extensive phase of advance likely occurred sometime prior to c. 6.8 ka, when glaciers extended up to 8 km beyond their current margins. However, these deposits lack maximum age constrains, and pre-Holocene ages cannot be discounted. Between c. 6.8 ka and the onset of ‘Neoglaciation’ c. 4.5 ka, there is little evidence of glacial advance upon the peninsula, and this period likely coincides with the Holocene climatic optimum (or ‘hypsithermal’). Since c. 4.5 ka, numerous moraines have been deposited, likely reflecting a series of progressively less extensive phases of ice advance during the Late Holocene. The final stage of notable ice advance occurred during the Little Ice Age (LIA), between c. 1350 and 1850 C.E., when reduced summer insolation in the Northern Hemisphere likely coincided with solar activity minima and several strong tropical volcanic eruptions to induce widespread cooling. Following the LIA, glaciers upon the peninsula have generally shown a pattern of retreat, with accelerated mass loss in recent decades. However, a number of prominent climatically and non-climatically controlled glacial advances have also occurred during this period. In general, there is evidence to suggest that millennial scale patterns in the extent and timing of glaciation upon the peninsula (encompassing much of the last glacial period) are governed by the extent of ice sheets in North America. Millennial-to-centennial scale fluctuations of Kamchatkan glaciers (encompassing much of the Holocene) are governed by the location and relative intensity of the Aleutian Low and Siberian High pressure systems. Decadal scale variations in glacier extent and mass balance (particularly since the LIA) are governed by inter-decadal climatic variability over the North Pacific (as reflected by the Pacific Decadal Oscillation), alongside a broader trend of hemispheric warming.

Phylogeographical analysis of two cold-tolerant plants with disjunct Lusitanian distributions does not support in situ survival during the last glaciation

Aim: We used a combination of modelling and genetic approaches to investigate whether Pinguicula grandiflora and Saxifraga spathularis, two species that exhibit disjunct Lusitanian distributions, may have persisted through the Last Glacial Maximum (LGM, c. 21 ka) in separate northern and southern refugia.

Location: Northern and eastern Spain and south-western Ireland.

Methods: Palaeodistribution modelling using maxent was used to identify putative refugial areas for both species at the LGM, as well as to estimate their distributions during the Last Interglacial (LIG, c. 120 ka). Phylogeographical analysis of samples from across both species' ranges was carried out using one chloroplast and three nuclear loci for each species.

Results: The palaeodistribution models identified very limited suitable habitat for either species during the LIG, followed by expansion during the LGM. A single, large refugium across northern Spain and southern France was postulated for P. grandiflora. Two suitable regions were identified for S. spathularis: one in northern Spain, corresponding to the eastern part of the species' present-day distribution in Iberia, and the other on the continental shelf off the west coast of Brittany, south of the limit of the British–Irish ice sheet. Phylogeographical analyses indicated extremely reduced levels of genetic diversity in Irish populations of P. grandiflora relative to those in mainland Europe, but comparable levels of diversity between Irish and mainland European populations of S. spathularis, including the occurrence of private hapotypes in both regions.

Main conclusions: Modelling and phylogeographical analyses indicate that P. grandiflora persisted through the LGM in a southern refugium, and achieved its current Irish distribution via northward dispersal after the retreat of the ice sheets. Although the results for S. spathularis are more equivocal, a similar recolonization scenario also seems the most likely explanation for the species' current distribution.

The taphonomy of Last Glacial–Interglacial Transition (LGIT) distal volcanic ash in small Scottish lakes

An extensive micro-tephrostratigraphic survey of three small lakes in the Scottish Inner Hebrides was conducted encompassing the Last Glacial–Interglacial Transition (LGIT). The lakes are highly contrasting in terms of lake area to catchment ratio, the presence or absence of stream inlets draining the catchment, and in the complexity of the catchment drainage network. A suite of distal Icelandic volcanic ashes was consistently detected in all three lakes, with three, namely Penifiler Tephra, Vedde Ash and Ashik Tephra, being common to all the lakes. These ashes were chosen to examine the taphonomic intercomparability of ash location and concentration among the lakes. Findings reveal that the part played by catchment inlets in determining ash concentration and within-basin location applies to microtephra layers as much as it does in studies of macrotephra layer thickness. The position of ash concentration maxima is also shown to vary significantly for different LGIT periods and may be a consequence of lake-level changes, especially during the early Holocene. High-resolution stratigraphic analysis through the Vedde Ash visible macrotephra at Loch Ashik reveals a high degree of complexity in taphonomic behaviour between the different geochemical components, with possible implications for the correct interpretation of the isochron position. The detection of multiple intact ash isochrons and the taphonomic processes responsible for their deposition should prove useful in future tephrostratigraphic surveys, as well as having applications within other palaeolimnological disciplines.

Three new distal tephras in sediments spanning the Last Glacial-Interglacial Transition in Scotland

Three new microtephras are reported from a number of lake sites from the Inner Hebrides and Scottish mainland. One occurs stratigrapically in the middle of Greenland Interstadial 1 (GI-1) and has been named the Penifiler Tephra. It is rhyolitic and possesses a geochemical signature that is very similar to that of the Borrobol Tephra, which also occurs in three of the sequences reported here, but which lies close to the lower boundary of GI-1. The second occurs stratigraphically in the early Holocene below the Saksunarvatn Ash and is named the Ashik Tephra. This tephra is geochemically bimodal, with a rhyolitic component comparable to the An Druim Tephra that occurs later in the Holocene, and a basaltic component which is similar to the Saksunarvatn Ash. A third tephra occurs stratigraphically above the Saksunarvatn Ash and is provisionally named the Breakish Tephra. The consistent inter-site correlation demonstrated for these new tephras at several sites enhances the regional tephrostratigraphic framework, and increases the potential for correlating palaeoenvironmental events during GI-1 and the early Holocene. However, the occurrence of multiple tephras with similar geochemistry in close stratigraphic and temporal proximity has implications for the rigour with which tephrostratigraphic investigations must be performed.

The latest LGM culmination of the Garda Glacier (Italian Alps) and the onset of glacial termination. Age of glacial collapse and vegetation chronosequence

In the deglacial sequence of the largest end moraine system of the Italian Alps, we focused on the latest culmination of the Last Glacial Maximum, before a sudden downwasting of the piedmontane lobe occupying the modern lake basin. We obtained a robust chronology for this culmination and for the subsequent deglacial history by cross-radiocarbon dating of a proximal fluvioglacial plain and of a deglacial continuous lake sedimentation. We used reworked dinocysts to locate sources of glacial abrasion and to mark the input of glacial meltwater until depletion. The palynological record from postglacial lake sediments provided the first vegetation chronosequence directly reacting to the early Lateglacial withdrawal so far documented in the Alps.

Glacier collapse occurred soon after 17.46 +/- 0.2 ka cal BP, which is, the Manerba advance culmination. Basin deglaciation of several overdeepened foreland piedmont lakes on southern and northern sides of the Alps appears to be synchronous at millennial scale and near-synchronous with large-scale glacial retreat at global scale. The pioneering succession shows a first afforestation step at a median modeled age of 64 years after deglaciation, while rapid tree growth lagged 7 centuries. Between 16.4 +/- 0.16 and 15.5 +/- 0.16 ka cal BP, a regressive phase interrupted forest growth marking a Lateglacial phase of continental-dry climate predating GI-1. This event, spanning the most advanced phases of North-Atlantic H1, is consistently radiocarbon-framed at three deglacial lake records so far investigated on the Italian side of the Alps. Relationships with the Gschnitz stadial from the Alpine record of Lateglacial advances are discussed

Presence of cave bears in western Austria before the onset of the Last Glacial Maximum: new radiocarbon dates and palaeoclimatic considerations

Tischoferhohle and Pendling-Barenhohle near Kufstein, Tyrol, are among the only locations where remains of cave bear, Ursus spelaeus-group, were found in the western part of Austria. One sample from each site was radiocarbon-dated four decades ago to ca. 28 C-14 ka BP. Here we report that attempts to date additional samples from Pendling-Barenhohle have failed due to the lack of collagen, casting doubts on the validity of the original measurement. We also unsuccessfully tried to date flowstone clasts embedded in the bone-bearing sediment to provide maximum constraints on the age of this sediment. Ten cave bear bones from Tischoferhohle showing good collagen preservation were radiocarbon-dated to 31.1-39.9 C-14 ka BP, again pointing towards an age underestimation by the original radiocarbon-dated sample from this site. These new dates from Tischoferhohle are therefore currently the only reliable cave bear dates in western Austria and constrain the interval of cave occupation to 44.3-33.5 cal ka BP. We re-calibrate and re-evaluate dates of alpine cave bear samples in the context of available palaeoclimate information from the greater alpine region covering the transition into the Last Glacial Maximum, eventually leading to the demise of this megafauna.

Sea-level changes in Iceland and the influence of the North Atlantic Oscillation during the last half millennium

We present a new, diatom-based sea-level reconstruction for Iceland spanning the last -500 years, and investigate the possible mechanisms driving the sea-level changes. A sea-level reconstruction from near the Icelandic low pressure system is important as it can improve understanding of ocean-atmosphere forcing on North Atlantic sea-level variability over multi-decadal to centennial timescales. Our reconstruction is from Viarhólmi salt marsh in Snæfellsnes in western Iceland, a site from where we previously obtained a 2000-yr record based upon less precise sea-level indicators (salt-marsh foraminifera). The 20th century part of our record is corroborated by tide-gauge data from Reykjavik. Overall, the new reconstruction shows ca0.6m rise of relative sea level during the last four centuries, of which ca0.2m occurred during the 20th century. Low-amplitude and high-frequency sea-level variability is super-imposed on the pre-industrial long-term rising trend of 0.65m per 1000 years. Most of the relative sea-level rise occurred in three distinct periods: AD 1620-1650, AD 1780-1850 and AD 1950-2000, with maximum rates of ~3±2mm/yr during the latter two of these periods. Maximum rates were achieved at the end of large shifts (from negative to positive) of the winter North Atlantic Oscillation (NAO) Index as reconstructed from proxy data. Instrumental data demonstrate that a strong and sustained positive NAO (a deep Icelandic Low) generates setup on the west coast of Iceland resulting in rising sea levels. There is no strong evidence that the periods of rapid sea-level rise were caused by ocean mass changes, glacial isostatic adjustment or regional steric change. We suggest that wind forcing plays an important role in causing regional-scale coastal sea-level variability in the North Atlantic, not only on (multi-)annual timescales, but also on multi-decadal to centennial timescales.

From Beijing with love: The global dimension of Bertolucci’s The Last Emperor

Famous for being the first foreign feature film that obtained permission to shoot in the Forbidden City, The Last Emperor (1987) is also one of the most ambitious and expensive independent productions of its time, awarded four Golden Globes and nine Academy Awards, including Best Picture. In addition, The Last Emperor can be considered as one of the first attempts of cinematic collaboration between West and East, in a period of cultural and economic transformations witnessed by China. This article aims to offer an overview of the production history of The Last Emperor, focusing on the co-production collaborations and the outcomes of a western auteur’s gaze on Chinese history. Questions of Orientalism, travel narrative and critical reception are taken into account in order to engage with the transnational implications of Bertolucci’s film and the western fascination with China.

Hydroclimatic shifts in northeast Thailand during the last two millennia - the record of Lake Pa Kho

The Southeast Asian mainland is located in the central path of the Asian summer monsoon, a region where paleoclimatic data are still sparse. Here we present a multi-proxy (TOC, C/N, δ¹³C, biogenic silica, and XRF elemental data) study of a 1.5m sediment/peat sequence from Lake Pa Kho, northeast Thailand, which is supported by 20 AMS ¹⁴C ages. Hydroclimatic reconstructions for Pa Kho suggest a strengthened summer monsoon between BC 170-AD 370, AD 800-960, and after AD 1450; and a weakening of the summer monsoon between AD 370-800, and AD 1300-1450. Increased run-off and a higher nutrient supply after AD 1700 can be linked to agricultural intensification and land-use changes in the region. This study fills an important gap in data coverage with respect to summer monsoon variability over Southeast Asia during the past 2000 years and enables the mean position of the Intertropical Convergence Zone (ITCZ) to be inferred based on comparisons with other regional studies. Intervals of strengthened/weaker summer monsoon rainfall suggest that the mean position of the ITCZ was located as far north as 35°N between BC 170-AD 370 and AD 800-960, whereas it likely did not reach above 17°N during the drought intervals of AD 370-800 and AD 1300-1450. The spatial pattern of rainfall variation seems to have changed after AD 1450, when the inferred moisture history for Pa Kho indicates a more southerly location of the mean position of the summer ITCZ.