The Cueva del Camino site (Pinilla del Valle, Madrid, Spain) is located in the upper valley of the Lozoya River in the Sierra de Guadarrama, a mountain range extending NE−SW within the Central Range System. Due to its location within a mountain range on the central Iberian Peninsula at an altitude of 1114 m a.s.l. and the numerical dating of its sediments, the palaeontological site of Cueva del Camino has proved a highly relevant location for studying the ecological changes linked to the climatic fluctuations at the end of Marine Isotope Stage (MIS) 5 and the beginning of MIS 4. Environmental reconstructions suggest a rather open, patchy landscape throughout the succession, with abundant evidence of dry meadows, scrublands and rocky habitats. The climate can be considered as generally warm, reaching mean annual temperatures (MATs) of up to 13.8°C (i.e. higher than today's by up to 2.9°C). Three cooler events can be seen throughout the succession as reflected by the presence of Rana iberica, Anguis fragilis and Coronella austriaca. The first of these events may correlate with MIS 5b; the second in the Central sector may correlate with the Stadial I pollen event occurring at the end of MIS 5a; and the third event, corresponding to the coldest MAT of the entire succession with MATs 0.9°C lower than today's, may correspond to the transition from MIS 5a to MIS 4. The evolution of mean annual precipitation (MAP) is characterized by warm periods, drier and cold periods, as well as wetter periods (up to +356 mm compared to today's MAP values), similar to what occurs today in the high-elevation areas of the neighbouring mountains. Our study gives new quantitative estimations for the climatic fluctuations in mountain environments of central Spain at the MIS 5/4 transition and their associated ecological changes.

The geomorphology of the south-western and central Lake District, England is used to reconstruct the mountain palaeoglaciology pertaining to the Lateglacial and Younger Dryas. Limitations to previous ice-mass reconstructions and consequent palaeoclimatic inferences include: (i) the use of static (steady-state) glacier reconstructions, (ii) the assumption of a single-stage Younger Dryas advance, (iii) greatly varying ice-volume estimates, (iv) inexplicable spatial variations in ELA (Equilibrium Line Altitude), and (v) a lack of robust extent chronology. Here we present geomorphological mapping based on aerial photography and the NextMap Britain Digital Elevation Model, checked by ground survey. Former glacier extents were inferred and ELAs were calculated using the Balance Ratio method of Osmaston. Independently, a time-dependant 2-D ice-flow model was forced by a regional ELA history that was scaled to the GRIP record. This provided a dynamic reconstruction of a mountain ice field that allowed for non-steady-state glacier evolution. Fluctuations in climate during the Younger Dryas resulted in multiple glacial advance positions that show agreement with the location of mapped moraines, and may further explain some of the ELA variations found in previous local and static reconstructions. Modelling based on the GRIP record predicts three phases: an initial maximum extent, a middle minor advance or stillstand, and a pronounced but less extensive final advance. The comparisons find that the reconstructions derived from geomorphological evidence are effective representations of steady-state glacier geometries, but we do propose different extents for some glaciers and, in particular, a large former glacier in Upper Eskdale.

We provide evidence for the subglacial to ice-marginal successive deposition of the Lohtaja−Kivijärvi ice lobe margin esker influenced by the changes in the meltwater delivery and proglacial water depth within the Finnish Lake District lobe trunk during the last deglaciation in Finland. The study is mostly based on the sedimentological data from the 100 km long esker chain with 15 logged sites. The long breaks in the lobe margin esker and the re-emerged deposition along the stable position of the subglacial meltwater route were related to the discontinuities and reappearances of the neighbouring eskers. This considerable variability in the meltwater discharge and debris transport under the described deglacial conditions cannot be explained by markedly decreased meltwater production due to palaeoclimatic factors or lack of debris within the trunk region. The primary control on the changes in meltwater availability and related esker deposition was thus due to the spatial and temporal changes in ice mass properties and shifting of the meltwater flow paths within the trunk. These changes were initiated by the topographically higher and partly supra-aquatic Suomenselkä watershed area with subsequent deepening of the proglacial water during the deglaciation. The understanding of the long-lived esker deposition along the former ice-stream trunk margin adds to the evaluation of palaeoglaciological reconstructions and geomorphologically based spatial models for ice-stream landscapes.

The late Quaternary glacial history of the Nun-Kun massif, located on the boundary between the Greater Himalaya and the Zanskar range in northwestern India, was reconstructed. On the basis of morphostratigraphy and ¹⁰Be dating of glacial landforms (moraines and glacial trimlines), five glacial stages were recognized and defined, namely: (i) the Achambur glacial stage dated to Marine Oxygen Isotope Stage 3 to 4 (38.7–62.7 ka); (ii) the Tongul glacial stage dated to the early part of the Lateglacial (16.7–17.4 ka); (iii) the Amantick glacial stage dated to the later part of the Lateglacial (14.3 ka, 11.7–12.4 ka); (iv) the Lomp glacial stage dated to the Little Ice Age; and (v) the Tanak glacial stage, which has the youngest moraines, probably dating to the last few decades or so. Present and former equilibrium-line altitudes (ELAs) were calculated using the standard area accumulation ratio method. The average present-day ELA of ∼4790 m above sea level in the Greater Himalaya is lower than those in the Ladakh and Zanskar ranges, namely 5380 and ∼5900 m a.s.l., respectively. The ELA in the Zanskar range is higher than in the Ladakh range, possibly due to the higher peaks in the Ladakh range that are able to more effectively capture and store snow and ice. ELA depressions decrease towards the Ladakh range (i.e. inner Plateau). Peat beds interbedded with aeolian deposits that cap the terminal moraine of Tarangoz Glacier suggest millennial-time-scale climate change throughout the Holocene, with soil formation times at c. 1.5, c. 3.4 and c. 5.2 ka, probably coinciding with Holocene abrupt climate change events. Given the style and timing of glaciation in the study area, it is likely that climate in the Nun-Kun region is linked to Northern Hemisphere climate oscillations with teleconnections via the mid-latitude westerlies.

Chronological, sedimentological and geochemical analyses of a clastic infill from Kelly Hill Cave (5K1), Kangaroo Island, document a palaeoenvironmental record that spans from the Late Pleistocene to the middle Holocene. We AMS radiocarbon-dated bone collagen and U–Th-dated speleothem to determine that fossiliferous sediments were deposited between >20 ka and 7 ka ago. Most of the 15 sedimentary layers are dominated by sand- and silt-sized quartz that is physically and geochemically comparable with surface soils in the Kelly Hill area. Late Pleistocene and Last Glacial Maximum strata are represented primarily by homogeneous, poorly sorted quartz-rich sediments that contain little organic matter, but include a thin layer composed largely of silt-sized clay pellets that resemble sediments deflated from playa lakes. Microstructures observed in petrographic slides indicate that, with the exception of one layer, all sediments experienced little reworking once deposited in the cave. Some layers display pedogenic microstructures such as redeposited clays and opaline silica infilling that indicate postdepositional modification; that is, cave-floor soil development. Overlying Holocene-aged sediments also consist mainly of quartz but have much greater organic matter content. Some of these sediments have been strongly influenced by re-precipitated organic matter that appears to have been transported into the cave via vadose drip water. The presence of dissolved organic matter in soil/vadose waters suggests a high vegetation density and acidic soils, which are congruent with the more equitable climatic conditions characteristic of the Holocene. The sediments described here provide a valuable palaeoenvironmental record that will facilitate future interpretation of associated vertebrate fossils.

The Late Pleistocene was characterized by rapid climate oscillations with alternation of warm and cold periods that lasted up to several thousand years. Although much work has been carried out on the palaeoclimate reconstruction, a direct correlation of ice-core, marine and terrestrial records is still difficult. Here we present new data from late Middle Pleniglacial to Lateglacial alluvial-fan and aeolian sand-sheet deposits in northwestern Germany. Records of Late Pleniglacial alluvial fans in central Europe are very rare, and OSL dating is used to determine the timing of fan aggradation. In contrast to fluvial systems that commonly show a delay between climate change and incision/aggradation, the small alluvial-fan systems of the Senne area responded rapidly to climatic changes and therefore act as important terrestrial climate archives for this time span. The onset of alluvial-fan deposition correlates with the climate change from warm to cold at the end of MIS 3 (29.3±3.2 ka). Strong fan progradation started at 24.4±2.8 ka and may be related to a period of higher humidity. The vertical stacking pattern of sedimentary facies and channel styles indicate a subsequrent overall decrease in water and sediment supply, with less sustained discharges and more sporadic runoffs from the catchment area, corresponding to an increasing aridity in central Europe during the Late Pleniglacial. Major phases of channel incision and fan aggradation may have been controlled by millennial-scale Dansgaard–Oeschger cycles. The incision of channel systems is attributed to unstable climate phases at cold–warm (dry–wet) or warm–cold (wet–dry) transitions. The alluvial-fan deposits are bounded by an erosion surface and are overlain by aeolian sand-sheets that were periodically affected by flash-floods. This unconformity might be correlated with the Beuningen Gravel Bed, which is an important marker horizon in deposits of the Late Pleniglacial resulting from deflation under polar desert conditions. The deposition of aeolian sand-sheet systems (19.6±2.1 to 13.1±1.5 ka) indicates a rapid increase in aridity at the end of the Late Pleniglacial. Intercalated flash-floods deposits and palaeosols (Finow type) point to temporarily wet conditions during the Lateglacial. The formation of an ephemeral channel network probably marks the warm-cold transition from the Allerød to the Younger Dryas.

Holocene palaeolimnological conditions were reconstructed by analysing fossil diatom assemblages within a lacustrine sediment core from Lake Sokoch, southern Kamchatka (Russia). Sediments of this proglacial lake cover the past 9400 years and hence represent almost the whole Holocene history. The biosiliceous muddy sample material was analysed for several geochemical and biological parameters, such as the total organic carbon and biogenic silica content, and the diatom community (quantitative and qualitative changes). Based on changes in the relative abundances of the most frequent species Aulacoseira subarctica, Staurosira martyi and Stephanodiscus alpinus and a depth-constrained cluster analyses (CONISS), five diatom assemblage zones could be identified. The oldest stage recovered lies between 9400 and 9000 cal. a BP and reflects the initial lake stage after the retreat of local glaciers, with a high detrital sediment supply, shallow-water conditions and a high diatom diversity. The next zone (9000–6200 cal. a BP) shows a more mature lake system with accumulating biogenic remains and higher water levels during climate amelioration. This is followed by the most obvious change in the diatom assemblage, delineated by an occurrence of S. alpinus, between 6200 and 2700 cal. a BP. Wet conditions in spring probably led to an enhanced fluvial runoff and eutrophic to hypertrophic conditions. The end of this period might reflect climate deterioration related to the Neoglacial epoch of the Holocene. Between 2700 and 1600 cal. a BP the sediments of Lake Sokoch reveal oligotrophic water conditions in a windy high-energy environment. The youngest interval, between 1600 cal. a BP and the Present, indicates shallow-water conditions and a very short growing season, which might reflect the Little Ice Age. The results may offer a baseline for the interpretation of Holocene palaeoenvironmental changes in Kamchatka and their relation to regional climate change from a palaeoecological perspective.

Two fluvial sediment cores taken from a floodplain of the Hawkesbury-Nepean River system in the Sydney region, eastern Australia are dated using Optically Stimulated Luminescence (OSL) to provide a reliable chronology essential for the management and planning of water resources. Nine charcoal ¹⁴C (AMS) dates constrain these OSL ages. Quartz extracted from seven OSL samples from each of the cores was measured using both single-grain and multi-grain OSL techniques. Three of the single-grain natural dose distributions appear to be well bleached, but the others appear to be incompletely bleached to various degrees. Three minimum-age models (MAM, MAM_UL and IEU) are applied to the single-grain dose distributions. We conclude that these models give consistent age estimates. For one of the cores it appears to be necessary to use a minimum-age model to obtain accurate ages, but in the other core incomplete bleaching is probably less important than postdepositional mixing and mixing during sampling. As a result, the burial age is probably best estimated using the weighted average of the individual single-grain dose estimates. The application of multi-grain OSL techniques to these samples results in an average apparent age overestimation of ∼200 years, which is significant for these samples, but negligible for sediments older than a few thousand years. The intention is that the chronology obtained in this study will be used in conjunction with a proxy flood record, derived from floodplain sediments, to gain an understanding of the long-term variability in periods of high and low rainfall in eastern Australia.

An annual (July to June) precipitation reconstruction for the period AD 1760–2010 was developed from a Picea crassifolia regional tree-ring chronology from two sites in the northern mountainous region of the Hexi Corridor, NW China. This reconstruction explains 52.1% of the actual precipitation variance during the period 1951 to 2010. Spatial correlations with gridded land-surface data reveal that our reconstruction contains a strong regional precipitation signal for the Hexi Corridor and for the southern margin of the Badain Jaran Desert. Significant spectral peaks were identified at 31.9, 11.1, 8.0, 7.0, 3.2, 2.6 and 2.2 years. A large-scale comparison indicates that our reconstruction is more consistent with climate records of a Westerly-dominated Central Asia, and that the Westerlies have a greater impact on the precipitation in this region than the Asian summer monsoon. Our reconstructed precipitation series is significantly correlated with sea-surface temperature (SST) in the tropical Atlantic Ocean (positive), the tropical Indian Ocean (positive), the western tropical Pacific Ocean (positive), and the western North Pacific Ocean (negative). The spatial correlation patterns between our precipitation reconstruction and SSTs of the Atlantic and Pacific Oceans suggest a connection between regional precipitation variations and the high-mid-latitude northern atmospheric circulations (Westerlies and Asian summer monsoon).

A database consisting of radiocarbon (¹⁴C), optically stimulated luminescence (OSL), thermoluminescence (TL) and beryllium (¹⁰Be) dates was used for timing the advance of the Late Weichselian Scandinavian Ice Sheet (SIS), determining the age of the Last Glacial Maximum (LGM) and the rate of deglaciation. The study area encompasses the southeastern sector of the last SIS between the Baltic Sea and the LGM position in the western part of the East European Plain, covering the Karelian Ice-Stream Complex (ISC) area in the east and the Baltic ISC area in the west. The linear advance and recession rates of the last SIS were estimated to be between 110 and 330 m a⁻¹ and between 50 and 170 m a⁻¹, respectively. The onset of the last SIS in the Karelian ISC area reached the western shores of Latvia not before 26 OSL ka, and in the Baltic ISC area, on the southern shores of the Gulf of Finland, not before 21 OSL ka. The last SIS reached close to the LGM position earliest in NW Belarus, not earlier than 22.6 cal. ¹⁴C ka BP, and latest in the NE of Belarus, not earlier than 19.1 cal. ¹⁴C ka BP. The Baltic ISC area between the LGM position and the western shores of Latvia was deglaciated in about 8 ka, and in the Karelian ISC area, between the LGM position and the southern shores of the Gulf of Finland, in about 2.6 ka. The whole area between the LGM position and the Baltic Sea was deglaciated between 14.2 ¹⁰Be ka and 13.3 cal. ¹⁴C ka BP.

Iceland's glaciers are particularly sensitive to climate change, and their margins respond to trends in air temperature. Most Icelandic glaciers have been in retreat since c. 1990, and almost all since 1995. Using ice-front measurements, photographic and geomorphological evidence, we examined the record of ice-front fluctuations of Virkisjökull–Falljökull, a steep high-mass-turnover outlet glacier in maritime SE Iceland, in order to place recent changes in a longer-term (80-year) context. Detailed geomorphological mapping identifies two suites of annual push moraines: one suite formed between c. 1935 and 1945, supported by lichenometric dating; the other between 1990 and 2004. Using moraine spacing as a proxy for ice-front retreat rates, we show that average retreat rates during the 1930s and 1940s (28 m a⁻¹) were twice as high as during the period from 1990 to 2004 (14 m a⁻¹). Furthermore, we show that both suites of annual moraines are associated with above-average summer temperatures. Since 2005, however, retreat rates have increased considerably – averaging 35 m a⁻¹ – with the last 5 years representing the greatest amount of ice-front retreat (∼190 m) in any 5-year period since measurements began in 1932. We propose that this recent, rapid, ice-front retreat and thinning in a decade of unusually warm summers has resulted in a glaciological threshold being breached, with subsequent large-scale stagnation of the glacier terminus (i.e. no forward movement) and the cessation of annual push-moraine formation. Breaching this threshold has, we suggest, caused further very rapid non-uniform retreat and downwasting since 2005 via a system feedback between surface melting, glacier thinning, decreased driving stress and decreased forward motion.

The stratigraphy of lake Endletvatn on northern Andøya, northern Norway, has been revisited to improve the understanding of the palaeoenvironment in the region during the Last Glacial Maximum (LGM). Four high-quality cores were analysed with respect to various lithological parameters and macrofossil content, supplemented by 47 AMS radiocarbon dates. The sediments indicate a low-energy environment with a mean sedimentation rate of 0.5 mm a⁻¹. We infer perennially frozen ground in the surroundings during the LGM. Climate proxies indicate a high arctic climate (i.e. July mean temperatures between 0 and 3°C) throughout most of the LGM. The warmest periods are marked by a rise in seed, moss and animal fossils, and often also by higher organic production in the lake. These periods took place from 21.4 to 20.1, from 18.8 to 18.1, around 17 and from 16.4 cal. ka BP onwards. The shifts between the different climatic regimes occurred rapidly – probably during one or two decades. The present data do not support recently published conclusions stating that Picea, Pinus and Betula pubescens grew on Andøya during parts of the LGM. The highest relative sea level after the final deglaciation on northern Andøya is bracketed between 36 and 38 m a.s.l. It occurred between 21.0 and 20.3 cal. ka BP, peaking around 20.7 cal. ka BP. The final deglaciation of the northern tip of Andøya occurred 22.2 cal. ka BP. Then the western margin of the Andfjorden ice stream receded to the Kjølhaugen Moraine and shortly thereafter to the Endleten Moraine. Our research confirms that northern Andøya is a key location for understanding the natural environment in northwestern Europe during the LGM.

Analyses of fossil mollusc successions have rarely been used to study the development of floodplain forests during the Holocene. The Ohře River, located in a prehistorically settled chernozem area in the Czech Republic, is partly situated in Cretaceous marlstones, yielding sediments suitable for fossilization directly in floodplain deposits. We analysed five fossil mollusc successions situated in the lower stretch of the Ohře River and compared the results with recent mollusc assemblages studied along the entire 256 km of the river. Fossil samples were composed mostly of open-country species throughout the Holocene or the whole preserved succession. Only some samples also contained woodland assemblages, but these were always greatly impoverished, with a very low frequency of strictly woodland species. Although the natural-looking appearance of the present-day floodplain forests of the lower river stretch has resulted in its being declared a nature reserve, modern floodplain forest mollusc assemblages there are also impoverished. This reduction in the distribution of strictly woodland species compared with modern assemblages in the upper stretch of the river seems to be the result of an ancient human settlement and continuous disturbances of the floodplain forest development since the Neolithic. Thus, fully developed floodplain forest assemblages occur recently only in the upper non-impacted stretch of the river. Based on the studied fossil successions we can conclude that the lower Ohře River floodplain was probably a mosaic of open and disturbed forest habitats throughout the Holocene. This area is part of a central European landscape island, where forests probably never fully developed and open patches from the early Holocene continually developed into an agricultural landscape.

We document the Lateglacial to Holocene sedimentation and vegetation history of a small, infilled crater landform in the West-Eifel Volcanic Field (WEVF; western Germany). We analysed geomorphological landform change, sedimentological and geochronological data, pollen, and plant macrofossils of a 16-m-long sediment core from the Eichholz Maar (EHM). The EHM erupted between ∼20 and 15 ka ago (MIS 2). Lacustrine siliclastic infilling was completed about 7500 years ago. Lateglacial rates of sedimentation are generally 2 to 5 times higher than in other maar lakes of the WEVF. Local factors, therefore, overprint the relative efficacy of the climate-controlled variance of sedimentation rates at the Lateglacial/Holocene transition. The predominance of local factors relates to inherent geomorphological process–response mechanisms that were triggered by the EHM eruption. Rapid crater infilling and its completion by the mid-Holocene are attributed to a combination of small storage capacity and geomorphological activity. A late Boreal interval of significant lake-level fall can, however, be attributed to a period of continental-scale climate change as recorded in other European lacustrine settings. Our findings highlight the importance of utilizing geomorphological information to reveal the relative significance of local controls as opposed to climate control when investigating small-sized lake settings with active sediment supply systems.

Glacier thermal regime is shown to have a significant influence on the formation of ice-marginal moraines. Annual moraines at the margin of Midtdalsbreen are asymmetrical and contain sorted fine sediment and diamicton layers dipping gently up-glacier. The sorted fine sediments include sands and gravels that were initially deposited fluvially directly in front of the glacier. Clast-form data indicate that the diamictons have a mixed subglacial and fluvial origin. Winter cold is able to penetrate through the thin (<10 m) ice margin and freeze these sediments to the glacier sole. During winter, sediment becomes elevated along the wedge-shaped advancing glacier snout before melting out and being deposited as asymmetrical ridges. These annual moraines have a limited preservation potential of ∼40 years, and this is reflected in the evolution of landforms across the glacier foreland. Despite changing climatic conditions since the Little Ice Age and particularly within the last 10 years when frontal retreat has significantly speeded up, glacier dynamics have remained relatively constant with moraines deposited via basal freeze-on, which requires stable glacier geometry. While the annual moraines on the eastern side of Midtdalsbreen indicate a slow steady retreat, the western foreland contains contrasting ice-stagnation topography, highlighting the importance of local forcing factors such as shielding, aspect and debris cover in addition to changing climate. This study indicates that, even in temperate glacial environments, restricted or localised areas of cold-based ice can have a significant impact on the geomorphic imprint of the glacier system and may actually be more widespread within both modern and ancient glacial environments than previously thought.

This paper presents the first integrated macroscale and microscale examination of subglacial till associated with the last-glacial (Fraser Glaciation) Cordilleran Ice Sheet (CIS). A new statistical approach to quantifying till micromorphology (multivariate hierarchical cluster analysis for compositional data) is also described and implemented. Till macrostructures, macrofabrics and microstructures support previous assertions that primary till in this region formed through a combination of lodgement and deformation processes in a temperate subglacial environment. Macroscale observations suggest that subglacial environments below the CIS were probably influenced by topography, whereby poor drainage of the substrate in topographically constricted areas, or on slopes adverse to the ice-flow direction at glacial maximum, facilitated ductile deformation of the glacier bed. Microscale observations suggest that subglacial till below the CIS experienced both ductile and brittle deformation, including grain rotation and squeeze flow of sediment between grains under moist conditions, and microshearing, grain stacking and grain fracturing under well-drained conditions. Macroscale observations suggest that ductile deformation events were probably followed by brittle deformation events as the substrate subsequently drained. The prevalence of ductile-type microstructures in most till exposures investigated in this study suggests that ductile deformation signatures can be preserved at the microscale after brittle deformation events that result in larger-scale fractures and shear structures. It is likely that microscale ductile deformation can also occur within distributed shear zones during lodgement processes. Cluster analysis of microstructure data and qualitative observations made from thin sections suggest that the relative frequency of countable microstructures in this till is influenced by topography in relation to ice-flow direction (bed drainage conditions) as well as by the frequency and distribution of voids in the till matrix and skeletal grain shapes.

The Cordilleran Ice Sheet (CIS) covered much of the mountainous northwestern part of North America at least several times during the Pleistocene. The pattern and timing of its growth and decay are, however, poorly understood. Here, we present a reconstruction of the pattern of ice-sheet retreat in central British Columbia at the end of the last glaciation based on a palaeoglaciological interpretation of ice-marginal meltwater channels, eskers and deltas mapped from satellite imagery and digital elevation models. A consistent spatial pattern of high-elevation (1600–2400 m a.s.l.), ice-marginal meltwater channels is evident across central British Columbia. These landforms indicate the presence of ice domes over the Skeena Mountains and the central Coast Mountains early during deglaciation. Ice sourced in the Coast Mountains remained dominant over the southern and east-central parts of the Interior Plateau during deglaciation. Our reconstruction shows a successive westward retreat of the ice margin from the western foot of the Rocky Mountains, accompanied by the formation and rapid evolution of a glacial lake in the upper Fraser River basin. The final stage of deglaciation is characterized by the frontal retreat of ice lobes through the valleys of the Skeena and Omineca Mountains and by the formation of large esker systems in the most prominent topographic lows of the Interior Plateau. We conclude that the CIS underwent a large-scale reconfiguration early during deglaciation and was subsequently diminished by thinning and complex frontal retreat towards the Coast Mountains.

The fine-grained (4–11 μm) quartz Optically Stimulated Luminescence (OSL) dating technique was applied to the Weinan section on the southeastern Chinese Loess Plateau (CLP) with a high luminescence sampling resolution (10- and 20-cm intervals). Fifty-eight OSL ages, spanning c. 1–74 ka, were obtained for the 10-m loess–palaeosol sequences. The reliability of the OSL dating and the constructed chronology was confirmed by comparing the OSL ages with independent dates from other studies and by correlation with palaeoclimatic time series. The closely spaced OSL ages at Weinan indicate that the mean dust accumulation rate (DAR) of L_1–1 (MIS 2, 7.0±0.9 cm ka⁻¹) is lower than those of L_1–3 (MIS 4, 19.1±6.1 cm ka⁻¹) and L_1–2 (MIS 3, 16.0±0.7 cm ka⁻¹) and that the mean DAR c. 30–20 ka ago (11.4±2.4 cm ka⁻¹) is higher than that c. 20–10 ka ago (3.0±0.1 cm ka⁻¹) in L_1–1. The subsequent calculation of the mass accumulation rate (MAR) indicates that the MAR of L_1–1 (107 g m⁻² a⁻¹) is much lower than those of L_1–2 (247 g m⁻² a⁻¹) and L_1–3 (307 g m⁻² a⁻¹). By comparing the mean DAR results during the Last Glacial at Weinan with those at a further six sites from other studies, we observed that the mean DAR of L_1–2 is higher (lower) than that of L_1–1 on the eastern (western) CLP, and that the mean DAR during MIS 2 has an evident transition from high to low at c. 20 ka on the entire CLP. Possible mechanisms for the above mean DAR changes at orbital and sub-orbital time scales are presented.

Based on geological and archaeological proxies from NW Russia and NE Estonia and on GIS-based modelling, shore displacement during the Stone Age in the Narva-Luga Klint Bay area in the eastern Gulf of Finland was reconstructed. The reconstructed shore displacement curve displays three regressive phases in the Baltic Sea history, interrupted by the rapid Ancylus Lake and Litorina Sea transgressions c. 10.9–10.2 cal. ka BP and c. 8.5–7.3 cal. ka BP, respectively. During the Ancylus transgression the lake level rose 9 m at an average rate of about 13 mm per year, while during the Litorina transgression the sea level rose 8 m at an average rate of about 7 mm per year. The results show that the highest shoreline of Ancylus Lake at an altitude of 8–17 m a.s.l. was formed c. 10.2 cal. ka BP and that of the Litorina Sea at an altitude of 6–14 m a.s.l., c. 7.3 cal. ka BP. The oldest traces of human activity dated to 8.5–7.9 cal. ka BP are associated with the palaeo-Narva River in the period of low water level in the Baltic basin at the beginning of the Litorina Sea transgression. The coastal settlement associated with the Litorina Sea lagoon, presently represented by 33 Stone Age sites, developed in the area c. 7.1 cal. ka BP and existed there for more than 2000 years. Transformation from the coastal settlement back to the river settlement indicates a change from a fishing-and-hunting economy to farming and animal husbandry c. 4.4 cal. ka BP, coinciding with the time of the overgrowing of the lagoon in the Narva-Luga Klint Bay area.

Previously only three terrestrial interglacial periods were known from southern Scandinavia, all of which could be relatively easily correlated within the central European stratigraphical framework. Here, we present a new interglacial–interstadial pollen, plant macrofossil and charcoal record from Trelde Klint, Denmark, and analyse its biostratigraphy, correlation with other European records, vegetation development, fire dynamics and absolute dating. Except for a slight truncation of the early part of the record, the pollen stratigraphy exhibits a full interglacial succession, including temperate trees (Quercus, Ulmus and Tilia) during its mesocratic stage. Macrofossil analysis allowed identification to species level for Quercus robur, Picea abies and two mosses. Conifers (Pinus and Picea) dominate the pollen record of the interglacial sequence, and the occurrence of Larix pollen in the top part of the interglacial record as well as in the interstadial sediments is especially indicative of this interglacial. The overall diversity of tree genera is rather low. These biostratigraphical features suggest that Trelde Klint is unique among Danish records, but it is similar to records from northern Germany. Numerical analyses (REVEALS and DCA) indicate that forests during the temperate stage were dense and that vegetation openness increased only towards the end of the interglacial, accompanied by increased fire occurrence. A short interstadial sequence with a dominance of Pinus and Betula and the presence of Larix is present above the interglacial deposit. We argue that lack of attention to differences in fire regimes may hamper understanding of between-site correlations of interglacial pollen records. OSL dating, using a novel feldspar technique, yields an average age of 350±20 ka for the sandy sediments above the interglacial layers at Trelde Klint, suggesting that the whole interglacial–interstadial succession belongs to Marine Isotope Stage 11.

The postglacial tree line and climate history in the Swedish Scandes have been inferred from megafossil tree remains. Investigated species are mountain birch (Betula pubescens ssp. czerepanovii), Scots pine (Pinus sylvestris) and grey alder (Alnus incana). Betula and Pinus first appeared on early deglaciated nunataks during the Lateglacial. Their tree lines peaked between 9600 and 9000 cal. a BP, almost 600 m higher than present-day elevations. This implies (adjusted for land uplift) that early Holocene summer temperatures may have been 2.3°C above modern ones. Elevational tree line retreat characterized the Holocene tree line evolution. For short periods, excursions from this trend have occurred. Between c. 12 000 and 10 000 cal. a BP, a pine-dominated subalpine belt prevailed. A first major episode of descent occurred c. 8200 cal. a BP, possibly forced by cooling and an associated shift to a deeper and more persistent snow pack. Thereafter, the subalpine birch forest belt gradually evolved at the expense of the prior pine-dominated tree line ecotone. A second episode of pine descent took place c. 4800 cal. a BP. Historical tree line positions are viewed in relation to early 21st century equivalents, and indicate that tree line elevations attained during the past century and in association with modern climate warming are highly unusual, but not unique, phenomena from the perspective of the past 4800 years. Prior to that, the pine tree line (and summer temperatures) was consistently higher than present, as it was also during the Roman and Medieval periods, c. 1900 and 1000 cal. a BP, respectively.

To provide insights into the long-term evolution of aquatic ecosystems without human interference, we here evaluate a decadal- to centennial-scale-resolution diatom record spanning about 12 ka of the Holsteinian interglacial (Marine Isotope Stage 11c). Using a partially varved sediment core from the Dethlingen palaeolake (northern Germany), which has previously been studied for palynological and microfacies signals, we document the co-evolution of the aquatic and surrounding terrestrial environment. The diatom record is dominated by the genera Stephanodiscus, Aulacoseira, Ulnaria and Fragilaria. Based on the diatom assemblages and physical sediment properties, the evolution of the Dethlingen palaeolake can be subdivided into three major phases. During the oldest phase (lasting ∼1900 varve years), the lake was ∼10–15 m deep and characterized by anoxic bottom-water conditions and a high nutrient content. The following ∼5600 years exhibited water depths >20 m, maximum diatom and Pediastrum productivity, and a peak in allochtonous nutrient input. During this phase, water-column mixing became more vigorous, resulting in a breakdown of anoxia. The youngest lake phase (∼4000–5000 years) was characterized by decreasing water depth, turbulent water conditions and decreased nutrient loading. Based on our palaeolimnological data, we conclude that the evolution of the Dethlingen palaeolake during the Holsteinian interglacial responded closely to (i) changes within the catchment area (as documented by vegetation and sedimentation) related to the transition from closed forests growing on nutrient-rich soils (mesocratic forest phase) to open forests developing on poor soils (oligocratic forest phase), and (ii) short-term climate variability as reflected in centennial-scale climate perturbations.

Thermokarst ponds are widespread in arctic and subarctic regions, but little is known about their temporal evolution prior to human observations. This paper presents a pioneer biostratigraphic study conducted at a subarctic site with limnologically contrasted ponds located on the eastern shore of Hudson Bay, Canada. Fossil diatom and visible near infrared (VNIR) derivative spectral analyses were performed on short sediment cores, confirming the occurrence of three distinct stratigraphic facies as already inferred from an anterior sedimentological study: a lacustrine upper facies (UF) and a marine lower facies (LF), separated by an organic-rich/peat transitional zone (TZ). Diatoms were almost absent from LF, but increased significantly in both TZ and UF. Identified diatom taxa were mainly benthic species (e.g. genera Fragilaria, Pinnularia), and their down-core distribution appeared to be related to dissolved organic carbon (DOC) and possibly pH conditions. Diatom-inferred DOC showed a decreasing trend towards the surface (potentially associated with an increase in pH), inverse to the general trend in this region, suggesting the action of other mechanisms on DOC, such as exhaustion of external inputs from limited catchments and the role of discontinuous peat layers (former surfaces of permafrost mounds) during the initial stages of pond formation. These bryophilous substrates in aerophilic habitats probably controlled diatom community composition. The combination of diatom and VNIR data revealed similar trends between (i) opal (amorphous silica) and diatom abundances; (ii) eukaryotic/prokaryotic algae ratio and anoxia or hypoxia in bottom waters; and (iii) limonite (iron oxide) and redox conditions in surface sediments. These findings indicate that diatom community changes and pond limnological evolution in the recent past were controlled mainly by autogenic processes (e.g. local vegetation/soil development, peat accumulation and erosion), rather than by allogenic forcing mechanisms (e.g. precipitation and temperature, geochemical leaching of the surrounding glaciomarine sediments).

The paper summarises materials on the mammal remains in northeastern Europe, dated by radiocarbon. Altogether, 23 local faunas of small mammals and 47 local faunas of large mammals were analysed. Multidimensional statistical analysis shows a strong correlation between changes in small mammal fauna composition and climate changes throughout time. The correlations with the spatial gradients, however, are less pronounced. The faunas are classified into three groups: (1) faunas of Holocene age; (2) Late Pleistocene ‘stadial’ assemblages; and (3) Late Pleistocene ‘interstadial’ assemblages. In some cases, changes in species abundance are better understood in terms of biotic interrelations rather than of climatic effects. The most pronounced change in small mammal fauna composition and structure occurred at the Preboreal/Boreal boundary, and a less conspicuous alteration took place at the LGM/Lateglacial transition. The most noticeable transformation in the large mammal fauna composition is dated to the early Holocene. Less significant changes are observed at the Middle Weichselian/LGM transition and at the LGM/Lateglacial transition. It is safely concluded that variations in the faunas of small and large mammals recorded in NE Europe during the last 35 000 years occurred synchronously and unidirectionally.

In unconsolidated sediments subject to strain, clays and silts are realigned into particular optical birefringent arrangements (plasmic fabrics), which provide information on the style and intensity of sediment deformation. A relatively new, non-destructive, optical microscopy technique for automatically recording and quantifying birefringence (previously commercialized under the name ‘Metripol’) is pioneered in this study as a valuable and innovative micromorphological tool with which to examine deformation in unconsolidated sediments. Metripol is applied to unistrial plasmic fabric in glaciotectonized and ice keel scoured sediment from the Netherlands and former Glacial Lake Agassiz (Manitoba, Canada) respectively. Colour-coded images are produced in which colour represents relative optical retardation and thus optical anisotropy through the quantity |sinδ| and optical orientation of anisotropy through the angle Ø (also indicated by linear azimuths). In this study Metripol typically demonstrates that the better developed the unistrial plasmic fabric is, the higher the |sinδ| values, the larger the areas of high |sinδ| values, and the longer and more densely populated the azimuths. In addition, some unistrial plasmic fabrics under Metripol demonstrate lower |sinδ| than previous examples and the surrounding sediment, despite being ‘perceived’ as demonstrating higher birefringence under a standard petrographic microscope. This is particularly true in clay-rich sediments and has implications for the way we currently describe and interpret unistrial plasmic fabrics in unconsolidated sediment. Finally, the identification and quantification of additional structures that would otherwise have gone undetected using a standard petrographic microscope (e.g. linear and circular structures that are likely to represent discrete shears and skelsepic plasmic fabric, respectively) highlight the potential for Metripol to gather information on the deformation history of unconsolidated sediments that is unavailable to standard techniques.

This paper describes and discusses the palaeobotanical data obtained from organic levels of two exposed deposits on the Atlantic shore of the northwestern Iberian Peninsula. Radiocarbon dating assigns these levels to a period of marine regression at the end of MIS 3 and the beginning of the Last Glacial Maximum. The pollen record shows an initial predominance of tree taxa (mainly deciduous, including the presence of Fagus far from its current limit), followed by an episode of partial forest retraction related to the end of MIS 3 and the beginning of the Last Glacial Maximum. However, the presence of numerous tree taxa in the record, even during the cold intervals associated with Heinrich events, points to the existence of sheltered refugia for these species during this period, in keeping with the conclusions of recent reviews of the vegetation dynamics of this region for the Lateglacial and the Holocene.

The results of detailed pollen-analytical investigations of a core from Lough Dargan, Co. Sligo, Ireland are presented. The pollen diagram spans much of the postglacial and documents changes in woodland composition and cover, and farming activity. Special attention is paid to prehistoric farming and to the significance of cereal-type pollen. The first sign of arable farming coincides with the Elm Decline at c. 3760 BC. This early Neolithic farming phase extended over c. 750 years, the main Landnam phase having a duration of ∼700 years. After a break of about three centuries, Neolithic farming resumed. Late Neolithic farming was at first predominantly pastoral, but later (c. 2360–2130 BC) it had a distinct arable component. In the early Bronze Age, beginning c. 2130 BC, farming increased and woodland was substantially reduced for the first time. From then until the beginning of the late Iron Age (c. 80 BC), there was a sustained and strong human impact. In the late Iron Age, a distinct lull in pastoral farming lasted for about four centuries (c. 80 BC–AD 350). This facilitated woodland regeneration that included yew. Substantial woodland clearance, and farming that included a considerable arable component, characterized the Medieval and later periods. The changes recorded at L. Dargan and other sites in the region are discussed in the light of evidence for climate change provided by regional and super-regional climate proxies. It is argued that climate may not have been a decisive factor in determining human impact and farming activity.

During the Last Glacial Maximum, the British–Irish Ice Sheet was dominated by a number of accumulation centres, including a terrestrially based, semi-independent icecap centred on Wales. The dynamics of this Welsh Ice Cap (WIC) over the last glacial period are still relatively poorly understood, with few studies taking into consideration the dynamic evolution of the icecap as a whole. Here we contrast results from two modelled reconstructions of the WIC in conjunction with the wider glacial geomorphological record to elucidate understanding of its form, extent and dynamics. Model output was analysed to yield zones of high basal motion and the spatial distribution of potential glacial erosion. We conclude that coherent flowsets of streamlined bedforms are linked to fast-flowing outlets dominated by basal sliding. Large-scale changes in dynamics are discussed, with a number of possible major advances proposed over the glacial cycle. Maximum ice thicknesses of ∼1200 m in Mid Wales indicate that all mountain summits were probably ice-covered during the Last Glacial Maximum, even if it was with a thin protective mantle of cold-based ice, leading to landscape preservation of these upland zones. The distribution, dynamism and landscape modification related to the WIC are further discussed at the regional scale. Model predictions of glacier distribution through the Younger Dryas stadial accord well with geologically reconstructed limits at this time.

A high-resolution, three-dimensional, thermomechanical ice-flow model is used to investigate the glaciodynamics of the Last Glacial Maximum Welsh Ice Cap – a large, independent ice centre of the British–Irish Ice Sheet. The model uses higher-order physics to solve longitudinal stresses, and is coupled to climate via a distributed, positive degree-day mass-balance scheme. A suite of model experiments driven by the GISP2 δ¹⁸O curve was initiated from a climatic optimum at 38.3 ka BP through to the Devensian/Holocene boundary to identify an icecap configuration compatible with available empirical evidence. An enhanced cooling from present of 11.85°C and strong precipitation suppression are required between 27.4 and 23.5 ka BP for the modelled icecap to attain well-established empirical limits, a scenario probably associated with Heinrich Event-2 and the potential collapse of thermohaline circulation in the North Atlantic. The experiments indicate ice-dispersal centres located in North and Mid Wales, the latter being essential for forcing ice southwards of the Brecon Beacons during the Last Glacial Maximum. Deglaciation of the Welsh Ice Cap was relatively rapid, occurring within one millennium. Dynamic stability is governed largely by the dominance and vigour with which fast-flowing outlet glaciers drain the icecap interior, which in turn are linked to variations in the climatic forcing. The distribution of permanently cold-based ice across the uplands and summits indicates the probable preservation of relict landscapes in these areas throughout the full glacial cycle.

The net effect of ice-flow shifts resulting in the dilution or reworking of clasts on a single preserved till sheet is often unknown yet has major implications for palaeoglaciology and mineral exploration. Herein, we analyse variations in till clast lithologies from a single till sheet, within palimpsest-type Glacial Terrain Zones in NE Manitoba, Canada, to better understand sediment–landform relationships in this area of high landform inheritance. This near-ice-divide area is known to consist of a highly fragmented subglacial landscape, resulting from spatio-temporal variations in intensity of reworking and inheritance throughout multiple glacial events (subglacial bed mosaic). We show that a seemingly homogenous ‘Keewatin’ till sheet is composed of local (>15 km) and continental-scale (∼100-km-long carbonate train and 350–600 km long Dubawnt red erratic train) fan, irregular (amoeboid) or lobate palimpsest dispersal patterns. Local dispersal is more complex than the preserved local landform flowset(s) record, but appears consistent with the overall glacial history reconstructed from regional flowset and striation analyses. The resultant surface till is a spatial mosaic interpreted to reflect variable intensities in modification (overprinting) and preservation (inheritance) of a predominately pre-existing till sheet. A multi-faceted approach integrating till composition, regional landforms, ice-flow indicators, and stratigraphic knowledge is used to map relative spatio-temporal erosion/reworking intensity.

The sediment core NP05-71GC, retrieved from 360 m water depth south of Kvitøya, northwestern Barents Sea, was investigated for the distribution of benthic and planktic foraminifera, stable isotopes and sedimentological parameters to reconstruct palaeoceanographic changes and the growth and retreat of the Svalbard–Barents Sea Ice Sheet during the last ∼16 000 years. The purpose is to gain better insight into the timing and variability of ocean circulation, climatic changes and ice-sheet behaviour during the deglaciation and the Holocene. The results show that glaciomarine sedimentation commenced c. 16 000 a BP, indicating that the ice sheet had retreated from its maximum position at the shelf edge around Svalbard before that time. A strong subsurface influx of Atlantic-derived bottom water occurred from 14 600 a BP during the Bølling and Allerød interstadials and lasted until the onset of the Younger Dryas cooling. In the Younger Dryas cold interval, the sea surface was covered by near-permanent sea ice. The early Holocene, 11 700–11 000 a BP, was influenced by meltwater, followed by a strong inflow of highly saline and chilled Atlantic Water until c. 8600 a BP. From 8600 to 7600 a BP, faunal and isotopic evidence indicates cooling and a weaker flow of the Atlantic Water followed by a stronger influence of Atlantic Water until c. 6000 a BP. Thereafter, the environment generally deteriorated. Our results imply that (i) the deglaciation occurred earlier in this area than previously thought, and (ii) the Younger Dryas ice sheet was smaller than indicated by previous reconstructions.

One of the thickest and most extensive tufa deposits in northern France occurs at Daours, about 8 km upstream of Amiens at the confluence of the River Somme and its tributary the Hallue. It covers an area of about 80 ha and reaches 10 m above the level of the present valley, probably owing to the construction of a tufa barrage, which caused periodic damming of the valley. The molluscan succession from a 4-m section through the Daours tufa provides an unusually detailed record of environmental change during the Holocene. The faunal record, supported by four radiocarbon dates, indicates that the tufa sequence at this location does not extend back to the early Holocene but began c. 7340 cal. a BP and ceased to form shortly after c. 4938 cal. a BP, consistent with the notion of a NW European ‘late Holocene tufa decline’. The molluscan record is divisible into six zones that define three aquatic phases separated by three terrestrial episodes, when the tufa surface dried out sufficiently to allow colonization by dry-ground species, including some shade-demanding elements. These events are also reflected by differences in the lithology of the tufa. The earliest aquatic phase at the base of the sequence represents the most stable and permanent water-body. The two subsequent aquatic phases appear to represent smaller, more ephemeral, water-bodies surrounded by marsh. The land snail assemblages show ecological changes within each terrestrial episode, as well as faunal differences between them. These events, which reflect changes in the local hydrology, were short-lived, each lasting for only a few hundred years. The tufa at Daours has also yielded flint artefacts and pottery, but it seems unlikely that the dynamic environmental record of the site results from anthropogenic activity.

In the present investigation, an age model of carbonate-rich cores from a seamount top in the Central Indian Basin (CIB) was constructed using both isotopic (²³⁰Th_excess, AMS ¹⁴C, oxygen isotopes) and biostratigraphic methods. The chronologies using the two methods are in good agreement, yielding a record of the late Middle Pleistocene to the Pleistocene–Holocene transition (550 to 11.5 ka). The first appearance datum (FAD) of the radiolarian Buccinosphaera invaginata (180 ka) and coccolith Emiliania huxleyi (268 ka) and the last appearance datum (LAD) of the radiolarian Stylatractus universus (425 ka) were used. A monsoon-induced productivity increase was inferred from carbonate, organic carbon and δ¹³C records in response to the Mid-Brunhes Climatic Shift (MBCS), consistent with an increased global productivity. While the coccolith diversity increased, a decrease in coccolith productivity was found during the MBCS. At nearly the same time period, earlier records from the equatorial Indian Ocean, western Indian Ocean and eastern Africa have shown an increased productivity in response to the influence of westerlies and increased monsoon. The influence of easterlies from Australia and the intensification of aridity are evidenced by increased kaolinite content and clay-sized sediments in response to the MBCS. An increased abundance of Globorotalia menardii and other resistant species beginning from marine isotope stage (MIS) 11 and the proliferation of coccolith Gephyrocapsa spp. indicate increased dissolution, which is consistent with the widespread global carbonate dissolution during this period. The relatively high carbonate dissolution during the transition period of MIS 3/2 and glacial to interglacial periods (MIS 6, 7 and 8) may be due to the enhanced flow of corrosive Antarctic Bottom Water (AABW) into the CIB.

A high-resolution record, covering 9.3–0.2 ka BP, from the sub-arctic Stjernsund (70°N) was studied for benthic foraminiferal faunas and stable isotopes, revealing three informally named main phases during the Holocene. The Early- to Mid-Holocene (9.3–5.0 ka BP) was characterized by the strong influence of the North Atlantic Current (NAC), which prevented the reflection of the Holocene Climatic Optimum (HCO) in the bottom-water temperature. During the Mid-Holocene Transition (5.0–2.5 ka BP), a turnover of benthic foraminiferal faunas occurred, Atlantic Water species decreased while Arctic-Polar species increased, and the oxygen isotope record showed larger fluctuations. Those variations correspond to a period of global climate change, to spatially more heterogeneous benthic foraminiferal faunas in the Nordic Seas region, and to regionally diverging terrestrial temperatures. The Cool Late Holocene (2.5–0.2 ka BP) was characterized by increased abundances of Arctic-Polar species and a steady cooling trend reflected in the oxygen isotopes. In this period, our record differs considerably from those on the SW Barents Sea shelf and locations farther south. Therefore, we argue that regional atmospheric cooling triggered the late Holocene cooling trend. Several cold episodes centred at ∼8.3, ∼7.8, ∼6.5, ∼4.9, ∼3.9 and ∼3.3 ka BP were identified from the benthic foraminiferal faunas and the δ¹⁸O record, which correlated with marine and atmospherically driven proxy records. This suggests that short-term cold events may result from reduced heat advection via the NAC or from colder air temperatures.

Dendrogeomorphic research has long relied on scarred trees to reconstruct the frequency of mass-movement processes. Injuries have mostly been dated macroscopically by counting the tree rings formed after wounding. Tree-ring anatomical anomalies induced by cambial injury, in contrast, have only recently been recognized as proxy records of past events. We investigated 12 sub-arctic downy birch (Betula pubescens Ehrh.) trees scarred by snow avalanches in Norway and Iceland. Earlywood vessel lumina were measured for each tree in the xylem tissue bordering the scars. Seven successive rings were examined, namely two control rings laid down prior to wounding and five rings in the wound xylem. We provide evidence that snow-avalanche-induced wounding resulted in atypically narrow earlywood vessels over at least two years. Our data demonstrate that wound-associated vessel anomalies represent tangible markers of mass-movement processes, and as such make a viable tool for reconstructing past events. Similar dendrogeomorphic studies based on tree-ring anatomy can be readily conducted with other mass-movement processes, as well as with other broad-leaved tree species. Ultimately, this new approach will foster increment coring over more invasive sampling techniques.

The Dolní Věstonice loess section in the Czech Republic is well known for its high-resolution loess–palaeosol sequence of the last interglacial–glacial climatic cycle (Upper Pleistocene). The loess section is situated in a climatic transition zone between oceanic and continental climates and is therefore of great value in reconstructing past regional climate conditions and their interaction with climate systems, in particular that of the North Atlantic. Based on a combination of optically stimulated luminescence (OSL) ages, stratigraphic field observations and magnetic susceptibility measurements, a chrono-climatic interpretation of the Dolní Věstonice loess section is presented. To establish a reliable Upper Pleistocene chronology, a quartz OSL approach was applied for equivalent dose (D_e) determination. Monomineralic quartz extracts of three distinct grain sizes, fine (4–11 μm), middle (38–63 μm) and coarse (90–200 μm), were used and compared. Within error limits, the calculated OSL ages are the same for the different grain sizes, and the OSL ages are in stratigraphic order. The established OSL chronology is in agreement with a Weichselian litho- and pedostratigraphy. The Dolní Věstonice loess section is characterized by four pedosedimentary subsequences. At the base of the profile, subsequence I is characterized by a distinct Early Glacial soil complex, OSL-dated to c. 110 to 70 ka, representing one of the most complete records of environmental change in the European loess belt. Subsequence II is allocated to the Lower Pleniglacial and is characterized by laminated sandy loess. Middle Pleniglacial subsequence III is represented by a brown soil complex, and is followed by the uppermost subsequence IV, characterized by a thick body of laminated sandy loess, indicating strong wind activity and a high sedimentation rate of more than ∼1 mm a⁻¹ during the Upper Pleniglacial. According to the OSL chronology, as well as to the sedimentological and palaeopedological investigations, it is likely that the sequence at Dolní Věstonice has recorded most of the climatic events expressed in the NGRIP δ¹⁸O reference record between 110 and 70 ka.

Cosmogenic ³⁶Cl was measured in bedrock and moraine boulders in the Za Mnichem Valley (High Tatra Mountains). The post-LGM deglaciation of the study area occurred about 15.9 ka ago. The northernmost part of the valley slopes was ice-free around 15 ka ago. The terminal moraine on the valley threshold was finally stabilized 12.5 ka ago during the Younger Dryas cold event (Greenland Stadial 1). At that time, the Za Mnichem glacier was 1.3 km long and had an area of 0.57 km². The AAR equilibrium line of the glacier was located at 1990 m a.s.l., which corresponds to an ELA depression of ∼500 m compared to today. The mean summer temperature was colder by 4°–4.5°C than the present-day temperature. The mean annual temperature was colder by 6°C than today. Such conditions suggest a decrease of the annual precipitation by ∼15–25% compared with the present-day annual average. These data indicate a probable uniform temperature change across central and western Europe, with the precipitation being the most significant factor affecting the mass balance of mountain glaciers. The spatial distribution of balance data suggests increasing continentality towards the east during the Younger Dryas.

Four cores from southwestern Sweden are presented together with their tephra geochemistry. Two cryptotephra horizons were confirmed geochemically in the cores, the Vedde Ash and the Hässeldalen Tephra. The Lateglacial Hässeldalen Tephra (11 360–11 300 cal. a BP) offers great potential as a regional isochrone to add a new degree of certainty to the deglaciation chronology of southern Sweden, including the extent of glacial Lake Bolmen. In addition, the geographical distribution of the Hässeldalen Tephra has recently been extended outside of Sweden, making it an important time-marker horizon in northern Europe. There are potential difficulties, however. Proper identification of the actual isochrone is complicated by the vertical pattern of shard distribution, which could be the result of several eruptive events, as well as by the fact that shards from the 10-ka Askja horizon (10 500–10 350 cal. a BP) were found in close stratigraphical proximity. The geochemical data presented are the result of improved EPMA methodology, which significantly reduces sodium mobilization. The results therefore have slightly altered values, which has consequences for classifying new finds when they are compared with previous data for geochemically similar tephras. Finally, potential indications of the Borrobol/Penifiler horizon are presented, although the existence of the horizon could not be confirmed geochemically. This highlights the need to retrieve cores from different locations within a basin based on an analysis of basin morphology if horizons are to be located.

Debris-covered glaciers may host several biological forms that colonize the debris cover, especially if the glacier tongue reaches sufficiently low altitudes (down to about 1700 m a.s.l. at Miage Glacier, Western Italian Alps) thus allowing also tree growth. Supraglacial trees colonizing the debris-covered tongue are strongly influenced in growth and distribution by substrate characteristics and instability. The tree age distribution at Miage Glacier presents a positive gradient towards the glacier terminus, which was found to be related to the decreasing glacier surface velocity. By analysing tree-ring growth anomalies on the glacier and at a control site at the tree line over the 20-year period 1987–2006, it was found that trees growing on the glacier presented the highest percentages of abrupt growth changes (AGCs)>+70% with respect to the four previous years. Considering tree displacement on the glacier surface over the same 20-year period and the recorded AGCs, it was found that the central-lower portion of the southern lobe towards the margins was the most unstable. The temporal analysis of AGC>+40% confirmed a period of higher glacier surface instability, reaching a maximum in the years 1988 (on lobe S) and 1989 (on lobe N), probably related to the passage of a kinematic wave in the glacier tongue. Our analysis suggests that supraglacial trees hold useful information on the glacier tongue dynamics and that both AGC>+70% and AGC>+40% may be used as a proxy for substrate instability in spatio-temporal reconstructions in the Alpine environment.

The recent discovery of a subfossil polar bear (Ursus maritimus) jawbone in the Poolepynten coastal cliff sequence, western Svalbard, and its implications for the natural history of the polar bear motivated an effort to better constrain the environmental history and age envelope of the Poolepynten sediment sequence. The focus of the present study is on the lithostratigraphy of the coastal cliffs and on re-dating the sequence using the Optically Stimulated Luminescence (OSL) dating technique. We report a revised lithostratigraphy and nine new OSL ages. It is concluded that the Poolepynten sequence contains evidence of four regional glaciation events, recorded in the strata as erosional unconformities or glacial deposits followed by shallow-marine deposition signifying transgressions and subsequent glacio-isostatic rebound and regression. Our OSL ages refine previous age determinations (¹⁴C and IRSL) and support the interpretation that the subfossil polar bear jawbone is probably of last interglacial (Eemian) age.

The varved sediment of Lake Suigetsu (central Japan) provides a valuable opportunity to obtain high-resolution, multi-proxy palaeoenvironmental data across the last glacial/interglacial cycle. In order to maximize the potential of this archive, a well-constrained chronology is required. This paper outlines the multiple geochronological techniques being applied – namely varve counting, radiocarbon dating, tephrochronology (including argon–argon dating) and optically stimulated luminescence (OSL) – and the approaches by which these techniques are being integrated to form a single, coherent, robust chronology. Importantly, we also describe here the linkage of the floating Lake Suigetsu (SG06) varve chronology and the absolute (IntCal09 tree-ring) time scale, as derived using radiocarbon data from the uppermost (non-varved) portion of the core. This tie-point, defined as a distinct (flood) marker horizon in SG06 (event layer B-07–08 at 1397.4 cm composite depth), is thus derived to be 11 255 to 11 222 IntCal09 cal. years BP (68.2% probability range).

The alternation of terrestrial and marine deposits is an indicator of past environmental and sea-level changes. The age of deposition is usually dated by means of radiocarbon. However, radiocarbon dates of molluscan shells from coastal areas may be complicated by various sources of carbon, and problematic for deposits of 40–50 ka or older. Herein, we apply the Optically Stimulated Luminescence (OSL) dating method to date samples from terrestrial and marine/coastal sediments extracted from three cores in the south Bohai Sea, China. Multiple- and single-aliquot regenerative-dose procedures using OSL signals from fine-silt (4–11 μm), coarse-silt (38–63 μm) and fine-sand (63–90 or 90–125 μm) quartz were employed to determine the equivalent dose (D _e). The results showed that: (i) OSL ages from quartz of different grain sizes and different protocols are consistent with each other; (ii) for Holocene samples, most of the radiocarbon dates agree well with OSL ages; (iii) for pre-Holocene samples, radiocarbon dates cluster at 40–50 ¹⁴ C ka BP, whereas OSL ages are in stratigraphic order from 11 ka to 176 ka. Because of the self-consistency of the quartz OSL ages, the stratigraphic agreement in the three cores, and the clustering of the radiocarbon dates, we suggest that the quartz OSL ages are more reliable with respect to dating the samples from the south Bohai Sea. Finally, the four marine strata identified in the south Bohai Sea are likely to have formed during the Holocene, Marine Isotopic Stage (MIS) 3–5, MIS 6 and probably MIS 7, respectively.

In this study we present optically stimulated luminescence (OSL) dating results obtained at one of the most important open-air Middle Palaeolithic sites in the Sierra de Atapuerca foothills – Hotel California. We also assess the possibility of obtaining extended-range OSL chronologies for a nearby Middle Pleistocene fluvial deposit using several novel methods, namely OSL dating of individual quartz ‘supergrains’, multi-grain aliquot thermally transferred OSL (TT-OSL) dating and the first application of a single-grain TT-OSL dating procedure. Four single-grain OSL ages constrain the Middle Palaeolithic occupation of Hotel California to between 71±6 and 48±3 ka. The Hotel California single-grain equivalent dose (D_e) distributions are highly overdispersed and contain several dose populations, which are probably attributable to post-depositional sediment mixing, partial bleaching and intrinsic scatter. The reliability of multi-grain aliquot OSL dating is compromised by the complex underlying D_e dispersion affecting these samples, as well as by biasing multi-grain averaging effects. Extended-range OSL and TT-OSL chronologies for the nearby Pico River terrace are consistent with each other and with broad independent age control. These experimental approaches yield a weighted average age of 348±16 ka for terrace T_A9 of the Arlanzón River sequence. Our results highlight the benefits of comparing ages obtained using several OSL methodologies to improve the robustness of luminescence chronologies. They also demonstrate the potential that single-grain OSL techniques offer for establishing improved age constraints on the many other Middle Palaeolithic sites found at Atapuerca and elsewhere across north-central Spain.

Despite the generalized occurrence of colluvial deposits in the humid tropical areas of southeastern Brazil, regional correlation is difficult because the deposits are discontinuous, and chronological data are very scarce and scattered. For the first time, colluvial deposits in the Bocaina Plateau are described, including 18 radiocarbon ages from 12 profiles. The plateau is located on the eastern flank of the Continental Rift of Southeastern Brazil and is the highest part of the Serra do Mar, with elevations up to 2000 m above sea level. Because the Bocaina Plateau is part of the summit surfaces of southeastern Brazil, it was subjected to specific climatic conditions during the Quaternary. Colluvial deposits on the lower hillslopes and edges of amphitheatres can show complex sequences with up to three intercalated dark humic horizons, corresponding to Late Pleistocene and Holocene phases of morphodynamic activity and pedogenesis. Ages of palaeosols vary from 650±50 a BP to 36 880±980 a BP. Landforms and colluvia on the Bocaina Plateau are very similar to those found on the opposite flank of the continental rift, on the Campos do Jordão Plateau. Ages of buried soils point to similar soil formation and colluviation episodes in the two plateaus. As in Campos do Jordão, the succession of erosive, depositional and pedogenetic processes on the altos campos hillslopes is probably related to late Quaternary climate changes.

Two sand wedge structures and their host sediments, from Jonzac in SW France, were successfully dated using Optically Stimulated Luminescence (OSL) measurements on both small aliquots and single grains of quartz from the 180–212 μm size fraction. One of the sand wedges clearly contains primary infilling. However, grain-size analysis and field observations do not clearly indicate whether the other feature represents a primary sand wedge or a composite sand wedge with primary and secondary infilling. OSL results and the geological setting justify using the Central Age Model (CAM) for the calculation of age estimates. Grain-size analysis and detailed investigations of OSL results revealed the contamination of one sand wedge sample with host sediment. However, age calculation using the Finite Mixture Model (FMM) provided what is considered to be a reliable age estimate for the contaminated sample. The age estimates for all samples correspond to Marine Isotope Stage (MIS) 3. While fine-grained sediments were deposited in the middle of MIS 3 (c. 43–55 ka), the sand wedges unexpectedly correspond to the end of this period (c. 33 ka) or the onset of MIS 2 (c. 27 ka). The sand wedges were probably formed during intense but short cold periods, possibly correlated with a Heinrich event (H2 and/or H3). The results help us to assess how effective luminescence dating is on sand wedges and the limitations involved in correlating sand wedge ages with Heinrich events, and contribute to the debate on the timing of cryogenic formation processes and the permafrost distribution in SW France.

Sand wedges and ice-wedge casts in the Qinghai Lake area on the northeastern Qinghai-Tibetan Plateau (QTP) occur within alluvial gravel or river terrace deposits. In this study, we report the results of quartz Optically Stimulated Luminescence (OSL) dating of the infill of five relict sand wedges and one ice-wedge cast. Combining our dating results with previously published luminescence ages of permafrost wedges in the Qinghai Lake area, we show that sand/ice wedges formed at c. 62 ka, c. 45 ka and between 30 and 15 ka, and that the mean annual air temperature (MAAT) was depressed by at least ∼3°C relative to present during the sand/ice-wedge formation periods. This new work is partially corroborated by post-LGM proxy records from lakes and aeolian deposits reported from the northeastern QTP. It also significantly extends the palaeoenvironmental record in the region in the period before the LGM, when other proxy records are rare, allowing a better understanding of the palaeoenvironmental conditions on the northeastern QTP.

The Provo shoreline of Lake Bonneville formed following the Bonneville flood, and, based on previous dating, was formed during a period of overflow from about 17.5 to 15.0 cal. ka. In many places the Provo shoreline consists of a pair of distinct shorelines, one ∼3 m higher than the other. We present data from two cuts through double beaches to show that the upper beach is younger and represents sedimentation after a lake-level rise. In addition, the lower beach deposits are internally stratified by beds that suggest three more lake-level rises during its development. The Provo beach complex thus appears to have been built during rising lake levels, which can be explained by rises in the overflow threshold by sequential landslide deposition. Evaluation of beach altitudes demonstrates that the two beach crests throughout the Bonneville basin experienced equivalent rebound from removal of the lake load, and therefore they formed after the rebound associated with the Bonneville flood occurred in early Provo time. However, radiocarbon ages on gastropods collected within the beach deposits suggest both that the sequence of five beach deposits formed from c.18.1 to c. 17.0 cal. ka, and that the Bonneville flood occurred before 18 cal. ka. These ages are discordant with previous dates on shells within offshore sands, and raise questions about the validity of radiocarbon ages for shells in Lake Bonneville as well as about the age of the Bonneville flood and Provo shoreline. The timing for maximum Provo lake depths and its association with climate stages during deglaciation remain unresolved.

Throughout northern Canada, live-collected, pre-bomb, deposit-feeding marine molluscs from calcareous sediments yield greater apparent radiocarbon ages than do suspension feeders. We explore the size of this effect in a set of 57 paired datings of deposit feeders, mainly Portlandia arctica, and suspension feeders, mainly Hiatella arctica and Mya truncata, collected from both calcareous and non-calcareous Holocene sediments. Deposit feeders from calcareous sediments are older than their suspension-feeding counterparts by as much as 2240±130 ¹⁴C years. This is attributed to the uptake of ‘old’ bicarbonate derived from calcareous bedrock. The age discrepancy between suspension and deposit feeders in calcareous terrain is non-systematic in space and time, thereby invalidating the application of a correction. In contrast, the age comparisons are concordant at sites located on the Precambrian Shield. In terrestrial environments underlain by carbonate, previous acceptance of dates on deposit feeders led to erroneous interpretations of deglaciation and relative sea-level history, in both the North American and the Eurasian Arctic. This has prompted several researchers to exclude deposit feeders from their late Quaternary reconstructions. The same chronological problem of deposit-feeding molluscs now needs to be more widely acknowledged by the marine community.

The ability of glaciers to detach and transport bedrock as glaciotectonic rafts is widely observed throughout Quaternary sections. However, the glaciological, hydrological and geological parameters controlling rafting are currently poorly constrained. There is a lack of structural and sedimentary evidence concerning rafting, and therefore the processes driving raft detachment, transport and emplacement are poorly understood. This paper contributes to our understanding by presenting a macro- and microstructural study of deformation associated with a chalk raft at West Runton, north Norfolk. Detailed thin-section analysis reveals several discrete micro-fabric orientations, representing poly-phase deformation occurring during raft transport and emplacement. A four-stage conceptual model for raft transport and emplacement is proposed, with deformation being partitioned into the relatively weaker Happisburgh Till member, the latter forming the host to the raft. Stage 1 is the main transport phase of the chalk raft, and was dominated by easterly (down-ice) directed ductile shearing. During Stage 2 a narrow ductile shear zone within the Happisburgh Till member propagated upwards through the base of the raft, leading to the detachment of an elongate block of chalk. Attenuated lenses of diamicton in this shear zone possess kinematics recording an easterly directed sense of shear. As deformation progressed, during Stage 3, the detached block impinged on the ‘high-strain’ zone wrapping the base of the raft, influencing the style of deformation partitioning and leading to localized, up-ice-directed kinematics. Stage 4 represents the final stages of raft emplacement, when the detachment zone at the base of the raft began to ‘lock-up’. These results demonstrate the relative importance of the hydrological controls associated with raft transport and emplacement underneath an actively advancing glacier. Furthermore, the model represents an example of how micromorphological analysis can reveal detailed poly-phase deformation histories in deformed glacial sediments.

Hydrofracture systems are being increasingly recognized within subglacial to ice-marginal settings and represent a visible expression of the passage of pressurized meltwater through these glacial environments. Such structures provide a clear record of the fluctuating hydrostatic pressure and of the resulting brittle fracturing of the host sediment/bedrock and the pene-contemporaneous liquefaction and introduction of sediment-fill. A detailed macro- and microstructural study of a hydrofracture system cutting Devonian sandstone bedrock exposed at the Meads of St John, near Inverness (NE Scotland), has revealed that this complex multiphase system was active over a prolonged period and accommodated several phases of fluid flow. The main conduits that fed the hydrofracture system are located along bedding within the sandstone, with the site of the wider, steeply inclined to subvertical, transgressive linking sections being controlled by the contemporaneous development of high-angle fractures and normal faults, the latter occurring in response to localized extension within the bedrock. A comparison with published engineering hydraulic fracturing data indicates that the various stages of sediment-fill deposited during a flow event can be directly related to the fluctuation in overpressure during hydrofracturing. A model is proposed linking the evolution of this hydrofracture system to the retreat of the overlying Findhorn glacier. The results of this study also indicate that the development and repeated reactivation of subglacial hydrofracture systems can have a dramatic effect on the permeability of the bed, influencing the potential for overpressure build-up within the subglacial hydrogeological system, and facilitating the migration of meltwater beneath glaciers and ice sheets.

The glacial sediment succession exposed close to the southern margin of the Late Weichselian Scandinavian Ice Sheet in Poland reveals a mosaic consisting of isolated patches of heavily deformed deposits separated by areas lacking any visible evidence of deformation. In the studied outcrop, the subglacial deforming spots composed of outwash deposits intercalated with till stringers are about 2–10 m wide and 20–60 cm thick. They rest on outwash sediments and are covered by a basal till. Based on structural and textural characteristics, the deforming spots are interpreted as previous R-channels filled with meltwater deposits. Lack of deformation in outwash sediment immediately beneath the deforming spots and in the intervening areas between the channels suggests that the ice-bed was frozen and the deformation of the channel infill was facilitated by high pore-water pressure arising because water drainage into the bed was impeded by permafrost. Channel infill deposits and the till immediately above were coevally deformed to a strain of less than 9. This study documents the possible co-existence of deforming and stable areas under an ice sheet, generated by spatially varying thermal and hydrological conditions affecting sediment rheology.

Tunnel valleys are common throughout the terrain of the Saginaw Lobe of the Laurentide Ice Sheet in southern Michigan. The set of valleys described in this paper is regularly spaced in a radial pattern behind the Kalamazoo Moraine, an ice-marginal position formed during retreat from the Last Glacial Maximum. These valleys are divided into proximal and distal groups lying north and south, respectively, of a major river valley that cross-cuts the tunnel valleys at right angles. Based on a series of rotasonic borings and core analysis, the proximal valleys are shallow, contain minimal sediment fill, and overlie fine-grained diamicton and glaciolacustrine sediment, whereas the distal valleys are deeply incised into the substrate and are partially filled with coarse sediment. The distal valleys terminate within a broad zone of high-relief, hummocky topography representing stagnation and collapse behind the Kalamazoo ice margin. The proximal valleys occur within a more subdued landscape located farther from the ice margin. Although some elements of existing genetic models are consistent with these valleys, none appears to be completely compatible with their stratigraphy and morphology. Initial incision of the valleys could have involved short-lived moderate- to high-discharge flows, followed by deposition during or after the events. The deep incision and thick, coarse sediment in distal valleys in the stagnant marginal zone probably involved supraglacial meltwater draining to the bed as the margin downwasted. Fining-upward eskers inset into the valleys were formed by flows of declining energy in small late-stage conduits.

It is a scientist's mission to try to remain unbiased. However, certain factors play a role in scientific analyses that are not controlled by conscious thought. These factors are potentially very important in areas of science where interpretations are based on a scientist's ability to identify patterns or structures. One such area is the micromorphology of glacial sediments. In this paper we investigate the role of an analyst's experience in relation to pattern perception with specific reference to turbate microstructures in glacial diamictons. An experiment was conducted on 52 participants, which demonstrated that, as may be expected, more experienced (glacial) micromorphologists tend to exhibit a higher sensitivity-to-signal, but that complete novices, if given clear instructions, can reach levels of sensitivity similar to those of experts. It also showed, perhaps more surprisingly, that response bias does not decrease with experience. We discuss psychological factors, such as the drive for success and consistency, that may have contributed to these results and investigate their possible implications in the micromorphological analysis and interpretation of glacial sediments.