Search

Aim Recent studies in southern Africa identified past biome stability as an important predictor of biodiversity. We aimed to assess the extent to which past biome stability predicts present global biodiversity patterns, and the extent to which projected climatic changes may lead to eventual biome changes in areas with constant past biome. Location Global. Taxon Spermatophyta; terrestrial vertebrates. Methods Biome constancy was assessed and mapped using results from 89 dynamic global vegetation model simulations, driven by outputs of palaeoclimate experiments spanning the past 140 ka. We tested the hypothesis that terrestrial vertebrate diversity is predicted by biome constancy. We also simulated potential future vegetation, and hence potential future biome patterns, and quantified and mapped the extent of projected eventual future biome change in areas of past constant biome. Results Approximately 11% of global ice-free land had a constant biome since 140 ka. Apart from areas of constant Desert, many areas with constant biome support high species diversity. All terrestrial vertebrate groups show a strong positive relationship between biome constancy and vertebrate diversity in areas of greater diversity, but no relationship in less diverse areas. Climatic change projected by 2100 commits 46%–66% of global ice-free land, and 34%–52% of areas of past constant biome (excluding areas of constant Desert) to eventual biome change. Main conclusions Past biome stability strongly predicts vertebrate diversity in areas of higher diversity. Future climatic changes will lead to biome changes in many areas of past constant biome, with profound implications for biodiversity conservation. Some projected biome changes will result in substantial reductions in biospheric carbon sequestration and other ecosystem services. SIGNIFICANCE STATEMENT Using global biome patterns inferred from simulations made using the LPJ-GUESS dynamic global vegetation model, we show that a substantial fraction of areas that are simulated to have supported the same biome throughout the last glacial-interglacial cycle are projected to experience biome change as a consequence of 21st century climatic changes. We further show that, with the exception of some desert areas, areas of the highest past biome constancy correspond to areas of the highest terrestrial vertebrate diversity. As a result, the projected biome changes are likely to have disproportionately large negative impacts upon global biodiversity.

Assessments of conservation status are typically based on short-term extinction risk, but the value of indicators that compare the current state of species (e.g., abundance or distribution) to potential baselines is increasingly recognised. The use of baselines in conservation legislation is hindered by ambiguity in how baselines should be determined and interpreted, leading to inconsistent application. Here, we explored the use of species’ potential ranges as a consistent means of quantifying baselines for assessing species’ distributions, a key component of conservation status. Using breeding birds of Great Britain (GB) as a case study, we simulated where bird species would be expected to occur today in a modelled world without human land use. We calculated indices that contrasted these potential human-free ranges with realised ranges. Our analyses revealed that 42% of GB birds have wider realised than potential ranges and 28% have narrower realised than potential ranges. These indices could lead to reassessments of current conservation priorities. Eighteen species assigned ‘least concern’ status by the GB regional IUCN Red List had much narrower realised than potential ranges, suggesting that their ranges are in a more degraded state than currently recognised by Red List criteria. Some of these species are not under active conservation management and could be candidates for higher prioritisation. Our approach provides a systematic means of quantifying range baselines that is not reliant on variable historic data or expert opinion and, thereby, provides a step forward in resolving a major contemporary problem in conservation assessment: how to set baselines in conservation consistently. The insights produced are also of wider scientific and cultural relevance, revealing where species would likely exist today in the absence of historic human impacts. This could be used to identify areas where targeted restoration actions might lead to the return of historically extirpated species, or even to novel colonists.

Insight into global biome responses to climatic and other environmental changes is essential to address key questions about past and future impacts of such changes. By simulating global biome patterns 140 ka to present, we aimed to address important questions about biome changes during this interval.\ud Location\ud \ud Global.\ud Taxon\ud \ud Spermatophyta.\ud Methods\ud \ud Using the LPJ‐GUESS dynamic global vegetation model, we made 89 simulations driven using ice‐core atmospheric CO2 concentrations, Earth's obliquity, and outputs from a pre‐industrial and 88 palaeoclimate experiments run using HadCM3. Experiments were run for 81 time slices between 1 and 140 ka, seven ‘hosing’ experiments also being run, using a 1‐Sv freshwater flux to the North Atlantic, for time slices corresponding to Heinrich Events H0–H6. Using a rule‐based approach, based on carbon mass and leaf area index of the LPJ‐GUESS plant functional types, the biome was inferred for each grid cell. Biomes were mapped, and the extent and total vegetation biomass of each biome, and total global vegetation biomass, estimated.\ud Results\ud \ud Substantial changes in biome extents and locations were found on all vegetated continents. Although the largest magnitude changes were in Eurasia, important changes were seen in tropical latitudes and the Southern Hemisphere. Total global extent of most biomes varied on multi‐millennial (orbital) time scales, although some (e.g. Tropical Raingreen Forest) responded principally to the c. 100‐kyr glacial–interglacial cycle and others (e.g. Temperate Broad‐leaved Evergreen Forest) mainly to the c. 20‐kyr precession cycle. Many also responded to millennial contrasts between stadial (‘hosing’) and interstadial climates, with some (e.g. Tropical Evergreen Forest) showing stronger responses than to the multi‐millennial changes.\ud Main conclusions\ud \ud No two time slices had identical biome patterns. Even equivalent Holocene and last interglacial time slices, and the last and penultimate glacial maxima, showed important differences. Only a small proportion of global land area experienced no biome change since 140 ka; many places experienced multiple biome changes. These modelling experiments provided little evidence for long‐term biome stability.

International audience; The Eemian interglacial represents a natural experiment on how past vegetation with negligible human impact responded to amplified temperature changes compared to the Holocene. Here, we assemble 47 carefully selected Eemian pollen sequences from Europe to explore geographical patterns of (1) total compositional turnover and total variation for each sequence and (2) stratigraphical turnover between samples within each sequence using detrended canonical correspondence analysis, multivariate regression trees, and principal curves. Our synthesis shows that turnover and variation are highest in central Europe (47-55°N), low in southern Europe (south of 45°N), and lowest in the north (above 60°N). These results provide a basis for developing hypotheses about causes of vegetation change during the Eemian and their possible drivers.

Millennial-scale Dansgaard Oeschger (DO) variability at northern high latitudes has influenced climatic and environmental conditions in the Mediterranean during the last glacial period. There is evidence that the hemispheric transmission of the DO variability occurred at the end of DO event 25; however, the exact timing and the trigger that activated the environmental response in the Mediterranean remains incompletely understood. Here, we provide evidence that the clear millennial-scale teleconnection between Greenland and the Mediterranean started at similar to 111.4 ka BP and was initiated by a sub-millennial scale cooling in Greenland (GI-25b). High-resolution sediment proxies and the pollen record of Lago Grande di Monticchio (MON), Italy, reflect climatic instability during the last millennium of the last interglacial, which was characterised by a first and short cooling episode (MON 1) at 111.44 +/- 0.69 ka BP, coinciding with the Greenland cold sub-event GI-25b in duration and timing (within dating uncertainties). MON and Greenland (NorthGRIP ice core) also agree in recording a subsequent warm rebound phase that abruptly culminated in the stadial MON 2/GS-25, marking the transition into the last glacial period. Our results show that the GI-25b triggered an early environmental response at MON to centennial-scale climate change in Greenland as a prelude to the millennial-scale teleconnection that was maintained during the glacial period.

Impacts of tephra deposition on vegetation are recorded in a series of 10 high temporal resolution absolute pollen diagrams from Lago Grande di Monticchio, each diagram spanning a single tephra deposition event during the last glacial–interglacial cycle. Sediment accumulation rates determined by counting and measurement of annual laminations (“varves”) provide an accurate and precise chronology, enabling the minimum recovery time after a tephra fall to be determined. In most cases, pollen accumulation rate was reduced after the tephra fall, indicating reduced vegetation productivity. Tephra deposition events also led to changes in vegetation composition, although these varied in magnitude. The magnitude and duration of the impacts upon the vegetation were related to the thickness of the tephra layer deposited, the thickest layers examined (>250 mm) having minimum recovery times of up to a century and thicker layers generally having greater impacts upon pollen productivity and vegetation composition. Tephra chemistry also influenced the persistence of the impact. The nature of the prevailing vegetation prior to the tephra fall influenced the degree and persistence of the impact. Tephra layers

Palaeoclimatic reconstructions from lake sediment biological records can be challenging, due to variation in non-climatic factors, which alter ecosystem responses. To consider this, it is important to replicate a study regionally, so as to gain information on spatial variability of ecosystem response and the influence of site-specific conditions. Previous pollen-based palaeoclimatic records from three well-dated Arctic lake sites highlight the response of regional Scots Pine ( Pinus sylvestris) and Mountain Birch ( Betula pubescens ssp. czerepanovii) forest-tundra transition to Holocene climatic variability and suggest the northernmost Peninsulas of Finnmark to be climatically sensitive. This study analysed dated sediment sequences between c. 3970 and c. 6200 cal. yr BP from these three previously published shallow lakes: Liten Čap’pesjav’ri (LCJ), over Gunnarsfjorden (OGF) and over Kobbkrokvatnet (OKV), for freshwater diatoms. Diatom assemblages showed an increase in the planktonic diatoms relative to benthic diatoms, with an onset towards higher abundances of small centric planktonic diatoms at OGF (between c. 5270 and 5350 cal. yr BP) and OKV (between c. 5280 and 5350 cal. yr BP). Additionally, a diatom compositional shift was detected at LCJ between c. 5180 and 5300 cal. yr BP. Trend analysis found the main diatom compositional change at all three sites to occur at 5300 cal. yr BP. However, this synchronous diatom shift during the mid-Holocene varies in magnitude within the three Finnmark lakes. The abrupt planktonic:benthic diatom ratio changes are independent of the Pinus:Betula ratios and are likely to be a result of lake ice-cover changes, allowing longer growing seasons, greater water column stability and higher nutrient concentrations from surrounding catchments. This study highlights the significant differences in aquatic ecosystem and terrestrial vegetation response to climatic changes, with diatom assemblages at these sites either responding before the regional tree-line shift or to a separate climatic event within the mid-Holocene.

Concern is growing that global climate change will have widespread impact on the world’s terrestrial ecosystems, but future impacts are imperfectly constrained by ecosystem models and direct observations. All available data, including records of past ecological change, need to be utilized to assess the range of potential future outcomes. Here we show that pervasive ecosystem transformations occurred in response to warming and associated climatic changes during the last glacial-to-interglacial transition, which was of comparable magnitude to climatic change projected to occur in the next 100 to 150 years under high-emission scenarios. We used data from 596 published paleoecological records to examine compositional and structural changes in global terrestrial vegetation since the last glacial period, and to project the magnitude of ecosystem transformations under different emission scenarios in the future. Our results indicate that terrestrial ecosystems are highly sensitive to temperature change, and suggest that without major reductions in greenhouse-gas emissions to the atmosphere (i.e., in line with those targeted in the 2015 Paris Agreement), most terrestrial ecosystems worldwide are at risk of major transformation, with accompanying disruption of ecosystem services and impacts on biodiversity.

Precisely-dated records of palaeovegetation and reconstructed palaeoclimate are presented from three lakes in northernmost Finnmark. The lakes lie adjacent to the southern shore of the Barents Sea and are located along a west–east transect. The three records are used to reconstruct spatial patterns in regional vegetation and climatic history since 13,900 cal yr BP. Longer-term shifts in treeline position and in the position of the Pinus–Betula ecotone are recorded. In addition, especially during the regional Holocene thermal maximum, the latter exhibited strong periodic fluctuations. The number and strength of these fluctuations that were recorded at each of the three sites differed systematically, with fewer and weaker fluctuations seen at the easternmost site, in particular. The patterns revealed are used to test the hypothesis that variations in the strength of the North Cape Current have been of primary importance as the proximal driver of climatic variability in the region since deglaciation. The results provide strong support for this hypothesis during the Holocene, the strong periodic fluctuations during the regional Holocene thermal maximum in particular being consistent with the proposed mechanism. During the Lateglacial and earliest Holocene the patterns are less clear, but nonetheless also consistent with the proposed mechanism. Further work on precisely-dated marine sediment cores will be necessary to understand the factors leading to the periodic and longer-term variations in strength of the North Cape Current.

We present a method and software for reconstructing palaeoclimate from pollen data with a focus on accounting for and reducing uncertainty. The tools we use include: forward models, which enable us to account for the data generating process and hence the complex relationship between pollen and climate; joint inference, which reduces uncertainty by borrowing strength between aspects of climate and slices of the core; and dynamic climate histories, which allow for a far richer gamut of inferential possibilities. Through a Monte Carlo approach we generate numerous equally probable joint climate histories, each of which is represented by a sequence of values of three climate dimensions in discrete time, i.e. a multivariate time series. All histories are consistent with the uncertainties in the forward model and the natural temporal variability in climate. Once generated, these histories can provide most probable climate estimates with uncertainty intervals. This is particularly important as attention moves to the dynamics of past climate changes. For example, such methods allow us to identify, with realistic uncertainty, the past century that exhibited the greatest warming. We illustrate our method with two data sets: Laguna de la Roya, with a radiocarbon dated chronology and hence timing uncertainty; and Lago Grande di Monticchio, which contains laminated sediment and extends back to the penultimate glacial stage. The procedure is made available via an open source R package, Bclim, for which we provide code and instructions.

Aim To test the following hypotheses: that Fynbos species had more extensive distributions at the Last Glacial Maximum (LGM), extending onto the exposed ‘Agulhas Plain’; that genetically distinct British taxa could have persisted through the LGM on adjacent areas of exposed shelf. Location Southern Africa; Europe. Methods Climatic response surfaces were fitted for 14 Fynbos and two European birds. These models were used to hindcast species distributions for palaeoclimates simulated using a fully coupled atmosphere–ocean general circulation model. LGM annual net primary productivity (ANPP) of two plant functional types upon which the European birds depend were simulated using a dynamic vegetation model and compared with potential LGM bird distributions. Results Fynbos birds' potential LGM distributions mostly extended southwards onto the exposed Agulhas Plain and were consistently more extensive than at present. This contrasts with conventional expectations for temperate species based upon Northern Hemisphere evidence. North-west European taxa potentially had LGM ranges on exposed shelf and ice-free areas to the west and south-west of the British Isles ice sheet. This is consistent with available genetic evidence, supporting the long-standing hypothesis that these taxa persisted through the LGM in this region. In both regions, results allow the generation of new testable hypotheses about species evolution and palaeobiogeography. Main conclusions Bird species found today in the Fynbos probably had more extensive ranges under glacial conditions, with the potential ranges of many species extending onto the Agulhas Plain. Bird taxa restricted today to the British Isles probably survived the LGM with limited distributions on exposed shelf and ice-free areas south-west of the British Isles ice sheet. Areas of shelf exposed under glacial conditions are likely to have been important components of glacial distributions of species in both the Northern and Southern Hemisphere. The contrasting history of Northern and Southern Hemisphere species has important conservation implications, especially in relation to conserving intra-specific genetic diversity.

In order to investigate the potential role of vegetation changes in megafaunal extinctions during the later part of the last glacial stage and early Holocene (42–10 ka BP), the palaeovegetation of northern Eurasia and Alaska was simulated using the LPJ-GUESS dynamic vegetation model. Palaeoclimatic driving data were derived from simulations made for 22 time slices using the Hadley Centre Unified Model. Modelled annual net primary productivity (aNPP) of a series of plant functional types (PFTs) is mapped for selected time slices and summarised for major geographical regions for all time slices. Strong canonical correlations are demonstrated between model outputs and pollen data compiled for the same period and region. Simulated aNPP values, especially for tree PFTs and for a mesophilous herb PFT, provide evidence of the structure and productivity of last glacial vegetation. The mesophilous herb PFT aNPP is higher in many areas during the glacial than at present or during the early Holocene. Glacial stage vegetation, whilst open and largely treeless in much of Europe, thus had a higher capacity to support large vertebrate herbivore populations than did early Holocene vegetation. A marked and rapid decrease in aNPP of mesophilous herbs began shortly after the Last Glacial Maximum, especially in western Eurasia. This is likely implicated in extinction of several large herbivorous mammals during the latter part of the glacial stage and the transition to the Holocene.

Aim We used a combination of new and previously published palaeoecological data to test three hypotheses: (1) that wooded steppe persisted in the Great Hungarian Plain throughout the Holocene; (2) that wooded steppe and steppe were most extensive between c. 9900 and 8300 cal. yr bp (the ‘Boreal steppe’ period); and (3) that Southern Continental, Pontic and Eastern Sub-Mediterranean steppe species reached the region during the early Holocene via the ‘Lower Danube Corridor’. Location Sarlo-hat oxbow lake, Hungary and the Eastern European wooded steppe zone. Methods Holocene sediments deposited in the Sarlo-hat oxbow lake were subjected to pollen and microcharcoal analyses. Twelve radiocarbon age estimates were obtained to determine sediment chronology. In addition, previously published palaeoecological data from the Great Hungarian Plain were compiled, analysed and compared with previous studies in other regions of steppe and wooded steppe in eastern Europe. Results Palynological data from two sediment cores extending to c. 11,400 cal. yr bp indicate the persistent dominance of the landscape by temperate deciduous wooded steppe throughout the Holocene, although with varying canopy composition. Warm-continental steppe grasslands and saline tall-grass meadows developed on edaphically constrained areas, which remained steppe-dominated throughout the Holocene. The extent of steppe grasslands did not increase between 9900 and 8300 cal. yr bp. After c. 3100 cal. yr bp, anthropogenic activities led to the development of cultural steppe. Thermophilous steppe species of the Southern Continental, Pontic and Sub-Mediterranean floristic elements probably reached the Great Hungarian Plain principally via the Lower Danube Corridor during the late glacial interstadial and Holocene. Eurythermic members of these elements, however, probably survived the Last Glacial Maximum in favourable microsites, extending their ranges during the Holocene from these local sources. Main conclusions Our results confirm the Holocene persistence of wooded steppe in the Great Hungarian Plain, disprove the ‘Boreal steppe’ theory, and suggest an Early Holocene period of greater vegetation openness between 11,400 and 9900 cal. yr bp. Evidence for the post-glacial immigration of south-eastern steppe elements into the Carpathian Basin is equivocal: the last glacial/interglacial presence of several southern steppe species suggests that the Hungarian Plain hosted suitable habitats for them during warm and cold phases alike.

In 1986/1987 the remains of several mammoths, Mammuthus primigenius (Blumenbach), were discovered on the spoil heap of an actively working gravel pit at Condover, Shropshire, England. The discovery of the remains posed two questions that could be addressed by analyses of biological proxies. First, as none of the bones was found in situ it was necessary to confirm the stratum in which the remains occurred. Second, what was the environment in which these animals lived and died? A range of biological indicators was used to address these questions, including pollen, spore and algal, plant macrofossil, invertebrate, anuran and biological mineral analyses. Multivariate statistical analyses of palynological and Pediastrum data, along with evidence from the Coleopteran assemblages, support the attribution of the mammoth bones to a unit of dark grey clayey sandy silt, although they may have lived at the time of the overlying green detritus mud. The palaeobiological data supports the correlation of these sediments to the Devensian Late-glacial. The mammoths entered this basin at the start of the Late-glacial Interstadial (Greenland Interstadial 1e) (ca. 14 830—3930 cal. year BP; 12 300 ± 110 14C year BP) and became mired in soft cohesive sediments. Palaeotemperature reconstructions, based on the Coleopteran assemblages, from the time when the mammoths actually became mired, show that the climate was temperate with mean July temperatures between 15 and 19°C and mean January temperatures between −13 and +6°C. Biological indicators from the sediments encasing the mammoths indicate that the landscape surrounding the basin was treeless and dry, contrasting with rich vegetation within the basin itself that had possibly attracted the mammoths to the site. Evidence of sedimentary disturbance suggests that the mammoths caused large-scale bioturbation of the deposits making palaeoenvironmental interpretations difficult. Fossils of terrestrial blowflies, carcass and dung beetles show that some of the decaying corpses must have lain exposed on the land surface for sufficient time for the soft parts to have rotted away and skin and bones to have become desiccated before many of them sank into the dark grey clayey sandy silt. Copyright © 2009 John Wiley & Sons, Ltd.

In 1986 remains of an adult woolly mammoth, Mammuthus primigenius (Blumenbach), were discovered at Norton Farm Pit, Condover, south of Shrewsbury, UK. Preliminary stratigraphical investigations indicated that this individual dated to the Devensian Late-glacial Interstadial, then the first evidence for survival of mammoth in Britain following the Last Glacial Maximum. Initial radiocarbon analysis confirmed this interpretation. Subsequent excavations in 1987/1988 recovered the remains of a further three juvenile mammoth individuals. All of these remains were found in the spoil heaps of overburden (ex situ) and their true stratigraphical context had to be reconstructed from the remnants surviving in the Pit. The 1987/1988 excavations enabled stratigraphical investigation of the site and submission of samples for radiocarbon (14C) dating, including the use of ultrafiltration pretreatment for bone samples, with the aims of reconstructing the geological and palaeoenvironmental evolution of the site and the sedimentary context of the unstratified mammoth remains. These results are presented here. This investigation indicates that the woolly mammoth remains at Condover derive from a dead-ice landscape dominated by eskers, kames and kettle-hole basins, and that the sedimentary sequence in which the mammoth remains were found forms the infilling of a kettle-hole basin. The sedimentary infilling and formation of the kettle-hole basin through ice block melt-induced subsidence were syngenetic. 14C determinations indicate that basin infill was initiated prior to Greenland Interstadial 1, and probably in Greenland Stadial 2 i.e. before 14.7 ka BP and that it continued until the early Holocene, around 8 ka BP. The sedimentological and 14C data indicate that the unstratified mammoth remains can be attributed to a dark grey clayey sandy silt (Unit C1), which accumulated during the earlier part of Greenland Interstadial 1 (14 to 14.5 ka BP) within an actively subsiding slow-flowing, beaded, fluvial network characterized by channels and pools/lakes, and with relatively shallow marginal slopes. The sedimentary architecture indicates survival of the buried ice block into Greenland Interstadial 1 and final melting only towards the end of the Interstadial at ca. 12.65 ka BP. Copyright © 2009 John Wiley & Sons, Ltd.

The sediments of Lago Grande di Monticchio, southern Italy, extend continuously from the present back to the penultimate glacial stage and have an independent lamination-based chronology of high precision and accuracy. Results are presented here from a detailed palynological investigation of that part of the sediment sequence that extends from the last millennia of the penultimate glacial stage to the first stadial following the Last Interglacial (LI). Quantitative palaeovegetation and palaeoclimate reconstructions made from the palynological data are also presented. The onset of the LI is dated to 127.20 ka BP, a date that is consistent with other recent estimates; the duration of the LI is estimated to have been 17.70 ka. The palaeovegetation record indicates a transition period of 3.35 ka at the end of the penultimate glacial stage prior to the onset of the LI; no Younger Dryas-like oscillation is recorded, although the transition was interrupted by a brief event, lasting ca 250 years, during which pollen of woody taxa was reduced in abundance. Steppe vegetation dominated during the latter part of the penultimate full-glacial stage, but was replaced progressively by wooded steppe during the transition. Although the development of forest cover marked the onset of the LI, the forests were relatively open or discontinuous during the first 2.65 ka, closing progressively thereafter and generally dominating between 123.00 and 109.50 ka BP. The end of the LI is dated to 109.50 ka BP, after which date forest cover became discontinuous and wooded steppe or steppe dominated during the 1.90 ka of the subsequent stadial. As might be expected, given the location of the lake, the composition of the LI forests differs markedly from those recorded from northern Europe, as well as from those recorded at other localities in southern Europe. The palaeoclimate reconstructions reveal complex changes in seasonality, the maximum coldest month mean temperatures being between 125.70 ka BP and 123.00 ka BP, whereas maxima for both annual temperature sum and the ratio of actual to potential evapotranspiration were between 120.60 ka BP and 115.80 ka BP. Reconstructed zonal mean values for all three climatic variables in the zones in which they peak exceed values at the locality today. Comparison with other palaeovegetation records of the LI from Europe reveals that forest cover generally opened up north of the Alpine region probably ca 115 ka BP, coinciding with a marked decrease in sea surface temperatures in the Nordic Seas; this probably corresponds to a marked shift in forest composition at Lago Grande di Monticchio at 115.80 ka BP with an associated reconstructed decrease of ca 5 °C in coldest month mean temperature. Nonetheless, forest continued to dominate at Lago Grande di Monticchio until 109.50 ka BP. Such comparisons also reveal considerable complexity in the geographical and altitudinal patterns of change in palaeovegetation during the LI; such complexity is to be expected given the parallel complexity of Holocene changes. Systematic comparisons between reconstructions of palaeoclimate are hampered by a lack of consistency in approach and in the variables reconstructed. Further insight into this complexity of palaeoclimate development during the LI requires a synthesis of the available data and application of a consistent reconstruction approach that also provides robust estimates of the uncertainty in the reconstructed values.

The dating of depths in two or more cores is frequently followed by a study of the synchroneity or otherwise of events reflected in the cores. The difficulties most frequently encountered are: (a) determining precisely the depths associated with the events; and (b) determining the ages associated with the depths. There has been much progress in recent years in developing tools for the study of uncertainties in establishing chronologies. This has not yet been matched by similar progress in modelling event/depth relationships. This paper proposes a simple and flexible approach, showing how uncertain events can be married to uncertain chronologies. Difficulties in studying event/depth/age relationships typically involve a confounding of two different problems. First, what exactly do we mean by an ‘event’ – a point in history, a single depth in the core corresponding to a single time, or a depth/time range? Sometimes ‘event’ is in fact a shorthand for a space-time process. Do the data reflect more than one type of event/process? This can reflect vagueness in definition. Second, what are the sources and implications of the uncertainties? Here we illustrate the issues involved by examination of several features seen in north European Holocene pollen records. The Alnus rise is regarded as a diachronous early Holocene event; in contrast the Ulmus decline is widely seen as a near synchronous event in the mid-Holocene. The third feature we examine is the interval between the Ulmus decline and the first occurrence of Cerealia-type pollen. The evidence for these events lies in cores of lake sediment from which are determined: (a) the proportions of pollen at many depths; and (b) radiocarbon age estimates from, usually, fewer depths. For this illustration we focus on six sites. We draw attention to a new and flexible method (implemented in the free R software package Bchron; [Haslett, J., Parnell, A., 2008. A simple monotone process with application to radiocarbon dated depth chronologies. Journal of the Royal Statistical Society: Series C 57 (4), 399–418]) for the establishment of the uncertainties surrounding the dating of samples in such cores. We illustrate its flexibility by assessing the synchroneity of past events.

A new global synthesis and biomization of long (> 40 kyr) pollen-data records is presented and used with simulations from the HadCM3 and FAMOUS climate models and the BIOME4 vegetation model to analyse the dynamics of the global terrestrial biosphere and carbon storage over the last glacial–interglacial cycle. Simulated biome distributions using BIOME4 driven by HadCM3 and FAMOUS at the global scale over time generally agree well with those inferred from pollen data. Global average areas of grassland and dry shrubland, desert, and tundra biomes show large-scale increases during the Last Glacial Maximum, between ca. 64 and 74 ka BP and cool substages of Marine Isotope Stage 5, at the expense of the tropical forest, warm-temperate forest, and temperate forest biomes. These changes are reflected in BIOME4 simulations of global net primary productivity, showing good agreement between the two models. Such changes are likely to affect terrestrial carbon storage, which in turn influences the stable carbon isotopic composition of seawater as terrestrial carbon is depleted in 13C.

The sediments of a small lake on Nordkinnhalvoya, Finnmark, Norway, were investigated in order to test the hypothesis that this region was sensitive to centennial–millennial climatic fluctuations during the Holocene related to changes in ocean circulation. Sedimentation at the site began during the Younger Dryas, although the site chronology, developed using a series of 14C age measurements, reveals an early Holocene hiatus in accumulation. Pollen analysis confirmed that the regional vegetation responded to Holocene climatic variability at centennial–millennial time scales and provided data used to make quantitative palaeoclimate reconstructions. The latter indicate that marked changes in seasonality characterised Holocene climatic fluctuations. Intervals with warmer summers, higher temperature sums and higher precipitation, but cooler winters and generally reduced moisture availability, alternated with intervals with cooler summers, lower temperature sums, lower precipitation, warmer winters and greater moisture availability. The former conditions were more prevalent between ca 8950 and 3950 cal BP, whereas the latter were predominant before ca 8950 and since ca 3950 cal BP. Sediment geochemistry indicates minerogenic material deposited in the lake was probably derived from two or more distinct sources or transport pathways that differed in their responses to palaeoclimatic conditions. A series of cryptotephras were located, although the small size of the shards rendered them unsuitable for electron microprobe analyses. Time-series analysis of pollen analytical and sediment geochemical data indicates that each exhibits statistically significant periodic behaviour (at periods of ca 190, 410, 1050, 1650 and 1810 yr). The periods detected suggest this behaviour may reflect regional expression of climate system responses to solar variability and/or of effects upon tides and ocean circulation of periodic lunar orbital variation. Comparison with records of fluctuations in ocean thermohaline circulation strength indicate some concordance with respect to timing of warmer and cooler intervals, but also some differences. The 8.2 ka event, that is evident in marine records from the Barents Sea, is clearly expressed by both the palaeovegetation and geochemical records. Distinctive temporal behaviour of the palaeovegetation and of different geochemical components indicates complexity in the underlying causes and mechanisms of regional climatic variability; ocean circulation variability alone cannot account for the complex climatic variability observed.

Detailed pollen, charcoal and loss on ignition profiles were analysed from Llyn Cororion, North Wales, UK. The chronology was based on 11 radiocarbon dates. This site is particularly important for this region because its high-resolution record improves the spatial and temporal resolution of records of Holocene vegetation change in an area characterized by a highly variable environment. An early Holocene phase of Juniperus-Betula scrub was succeeded by Betula-Corylus woodland. Quercus and Ulmus were established by c. 8600 14C yr BP, with Pinus dominating at c. 8430 14C yr BP. Local disturbance then allowed the spread of Alnus; Tilia was a common component of the forest by 5650 14C yr BP. Charcoal and pollen records suggest that by c. 2600 14C yr BP there was progressive deforestation, increased use of fire and spread of grassland; the first cereal grain was recorded at c. 2900 14C yr BP. Compared with data from upland Snowdonia, the results show that within a topographically diverse region there were significant local variations in forest composition. These variations developed as a response to interactions between many environmental parameters and were further complicated by the influence of human activity. In an area such as North Wales it is therefore unlikely that one site can be representative of regional Holocene vegetational development. The site is additionally important because it contributes to the data available for meta-analyses of environmental change in the North Atlantic region, particularly as detailed pollen diagrams from coastal lake sites around western Europe are rare.

Establishing phase relationships between earth-system components during periods of rapid global change is vital to understanding the underlying processes. It requires records of each component with independent and accurate chronologies. Until now, no continental record extending from the present to the penultimate glacial had such a chronology to our knowledge. Here, we present such a record from the annually laminated sediments of Lago Grande di Monticchio, southern Italy. Using this record we determine the duration (17.70 +/- 0.20 ka) and age of onset (127.20 +/- 1.60 ka B.P.) of the last interglacial, as reflected by terrestrial ecosystems. This record also reveals that the transitions at the beginning and end of the interglacial spanned only approximately 100 and 150 years, respectively. Comparison with records of other earth-system components reveals complex leads and lags. During the penultimate deglaciation phase relationships are similar to those during the most recent deglaciation, peaks in Antarctic warming and atmospheric methane both leading Northern Hemisphere terrestrial warming. It is notable, however, that there is no evidence at Monticchio of a Younger Dryas-like oscillation during the penultimate deglaciation. Warming into the first major interstadial event after the last interglacial is characterized by markedly different phase relationships to those of the deglaciations, warming at Monticchio coinciding with Antarctic warming and leading the atmospheric methane increase. Diachroneity is seen at the end of the interglacial; several global proxies indicate progressive cooling after approximately 115 ka B.P., whereas the main terrestrial response in the Mediterranean region is abrupt and occurs at 109.50 +/- 1.40 ka B.P.

Pollen records and pollen-based climate reconstructions from the Italian peninsula (central Mediterranean) show clear signals of vegetation change linked to variations in water availability in the Mediterranean basin over the past 5 million years. Profound vegetation changes occurred in four major steps from the Pliocene to the present. The subtropical taxa that dominate Pliocene assemblages declined and then disappeared between 3–2.8 and 1.66 Ma (at around 2.8 Ma in the North and later in the South), progressively being replaced by temperate Quercus forests at mid altitude. In the south Italy, Quercus expanded more at around 1.4–1.3 Ma and Fagus proportions increased after 0.5 Ma. Conifer forest (first mainly composed of Tsuga then by Abies and Picea) began to expand at 2.8 Ma, probably rather at high altitude, beginning at 2.8 Ma. Mediterranean-type forest, rare during the Early Pleistocene, developed and increased in diversity during the Middle Pleistocene. Open landscapes, with higher abundances of steppic taxa, became more frequent and extensive at the onset of Glacial/Interglacial (G/I) cyclicity around 2.6 Ma and gradually expanded with more and more marked glacials. Climate reconstructions done on selected pollen records from southern Italy suggest a decline in winter temperature and annual precipitation from the early Pleistocene to the Holocene. Specifically, both precipitation and winter temperature reconstructions show changes in interglacial maxima and glacial minima at around 3–2.8 Ma, 2 Ma, 1.3–1.4 Ma and 0.5 Ma. This critical review provides evidence that the North–South precipitation gradient, with drier conditions in the South, has been a consistent feature of the Italian peninsula since the beginning of the Pleistocene.

SummaryWe consider the problem of reconstructing prehistoric climates by using fossil data that have been extracted from lake sediment cores. Such reconstructions promise to provide one of the few ways to validate modern models of climate change. A hierarchical Bayesian modelling approach is presented and its use, inversely, is demonstrated in a relatively small but statistically challenging exercise: the reconstruction of prehistoric climate at Glendalough in Ireland from fossil pollen. This computationally intensive method extends current approaches by explicitly modelling uncertainty and reconstructing entire climate histories. The statistical issues that are raised relate to the use of compositional data (pollen) with covariates (climate) which are available at many modern sites but are missing for the fossil data. The compositional data arise as mixtures and the missing covariates have a temporal structure. Novel aspects of the analysis include a spatial process model for compositional data, local modelling of lattice data, the use, as a prior, of a random walk with long-tailed increments, a two-stage implementation of the Markov chain Monte Carlo approach and a fast approximate procedure for cross-validation in inverse problems. We present some details, contrasting its reconstructions with those which have been generated by a method in use in the palaeoclimatology literature. We suggest that the method provides a basis for resolving important challenging issues in palaeoclimate research. We draw attention to several challenging statistical issues that need to be overcome.

European vegetation during representative “warm” and “cold” intervals of stage-3 was inferred from pollen analytical data. The inferred vegetation differs in character and spatial pattern from that of both fully glacial and fully interglacial conditions and exhibits contrasts between warm and cold intervals, consistent with other evidence for stage-3 palaeoenvironmental fluctuations. European vegetation thus appears to have been an integral component of millennial environmental fluctuations during stage-3; vegetation responded to this scale of environmental change and through feedback mechanisms may have had effects upon the environment. The pollen-inferred vegetation was compared with vegetation simulated using the BIOME 3.5 vegetation model for climatic conditions simulated using a regional climate model (RegCM2) nested within a coupled global climate and vegetation model (GENESIS-BIOME). Despite some discrepancies in detail, both approaches capture the principal features of the present vegetation of Europe. The simulated vegetation for stage-3 differs markedly from that inferred from pollen analytical data, implying substantial discrepancy between the simulated climate and that actually prevailing. Sensitivity analyses indicate that the simulated climate is too warm and probably has too short a winter season. These discrepancies may reflect incorrect specification of sea surface temperature or sea-ice conditions and may be exacerbated by vegetation–climate feedback in the coupled global model.

Oxygen isotope stage 3 (OIS3), an interstade between approximately 60,000 and 25,000 yr B.P., presents an ideal opportunity to compare high-resolution climate simulations with the geologic record. To facilitate this comparison, the results of a mesoscale climate model (RegCM2) embedded in the GENESIS GCM are utilized to drive a vegetation model (BIOME 3.5). The BIOME output is then compared with OIS3 compilations derived from pollen. The simulated biomes agree well with the pollen-based biomes in southern Europe; however, disagreements occur in the northern part of the domain. The most striking mismatch involves the distribution of tundra. The models fail to have tundra extend to its observed position as far south as 50°N in central Europe during OIS3. The model also fails to have permafrost extend southward to its observed position between 50°N and 55°N in western Europe during OIS3. A variety of sensitivity experiments are performed to investigate these mismatches. These experiments demonstrate the importance of annual and summer temperatures and the length of the winter season in creating improved matches between the model results and the inferred distributions of vegetation and permafrost in northern Europe.

The Holocene pollen record from Lago Grande di Monticchio reveals environmental changes throughout the Holocene. Comparison of this record with other records of Holocene environmental variability indicates that there is a temporal correlation between these changes and north Atlantic events. However, the nature of this relationship and the underlying mechanism(s) is yet to be determined. It is evidently complex with the apparent coincidence of sea-surface temperature cooling events with both species expansions and disappearances. The picture is further complicated by the interaction between human activity and natural environmental changes.

We test the hypothesis that climatic changes since 1800 have resulted in unrealized potential vegetation changes that represent a "climatic debt" for many ecosystems. Caledonian pinewoods, an EU priority forest type, are used as a model system to explore potential impacts of two centuries of climatic change upon sites of conservation importance and surrounding landscapes. Using methods that estimate topographic microclimate, current and preindustrial climates were estimated for 50 m grid cells and simulations made using a dynamic vegetation model. Core Caledonian pinewood areas are now less suitable for growth of pine and more favorable for oak than in 1800, whereas landscapes as a whole are on average more favorable for both. The most favorable areas for pine are now mainly outside areas designated to conserve historical pinewoods. A paradigm shift is needed in formulating conservation strategies to avoid catastrophic losses of this habitat, and of many others globally with trees or other long-lived perennials as keystone species.

The high resolution, independently dated palynostratigraphic record from Lago Grande di Monticchio spanning the last 101,530 calendar years provides a basis for re-evaluating the correlations between palaeoenvironmental events in Europe during the last glacial as well as a means to determine their chronology. The interstadial events recorded from various localities throughout Europe are combined into a single scheme of correlations and a chronological framework is suggested. This provides a basis for assessing the evolution of the vegetation and environmental patterns across Europe during the last glacial, revealing a general tendency for the latitudinal climatic gradient to have been steeper than it has been during the Holocene. The vegetation has also generally been subject to greater moisture deficiency, either because of prevailing climatic conditions or as a reflection of the lower atmospheric concentration of CO2. The scheme of correlations is shown to be capable of being extended beyond Europe, not only to marine and ice-core records from the North Atlantic region, but also to a palynological record from north-west North America.

Three interstadial events recorded in a pollen diagram from Lago Grande di Monticchio, southern Italy (40°56′40″N, 15′36″E), can be correlated with Dansgaard/Oeschger events 12, 14 and 20 of the GRIP (Summit) Greenland Ice Core. Based on a chronology of varve counts, the events are dated respectively to the intervals 37,600–40,000, 43,600–50,000 and 72,000–74,600 calendar yr BP. The interstadials are of unequal duration, the longest occupying 6400 years. They appear to begin and end abruptly. Each is characterised by a sequence of pollen changes in the mesic forest component. Similarities in these sequences of mesic forest trees imply similar patterns of climatic change within each interstadial. The sequences resemble that of the late-glacial interstadial at Monticchio. Although tree pollen abundance values reach as much as 60% during the interstadials, there is always a substantial proportion of herb pollen. Tree growth may have been constrained by low CO2 levels during the glacial phase of the last glacial–interglacial cycle. This record illustrates the potential of long continental records to present opportunities for detailed ecological and chronological analyses of sub-interglacial warming events during oxygen-isotope Stages 3 and 4.

Palynological and chronological data are presented for the Lateglacial from three cores taken at different locations in the basin of Lago Grande di Monticchio. Plant macrofossil data from one core provide complementary evidence of the palaeovegetation. Quantitative palaeoclimate reconstructions are made from the pollen data using pollen-climate response surfaces. The chronological framework for Late-glacial events as well as the palaeoclimate are compared with other sources of evidence of Lateglacial palaeoenvironments. Surface conditions in the North Atlantic appear to be more directly linked to climatic conditions in Italy than is the temperature in Greenland. Nonetheless, the palynological record exhibits high-frequency variability that reflects high-frequency climatic changes paralleling those seen in the ice core records from Greenland.

High temporal resolution pollen and diatom analyses carried out on sediments from Lago Grande di Monticchio span the interval 23 700–21 200 calendar yr. BP, a brief interstadial during marine Oxygen Isotope Stage 2. Both records exhibit marked changes that are interpreted as responses to climatic changes. The diatoms and terrestrial vegetation appear to respond at the same time; any relative lag in the response of the vegetation was less than the ca. 60 yr resolution of the two records. The interval coincides, at least in part, with the Campo Imperatore Stade, when the Gran Sasso ice sheet reached its maximum extent and water level was high in the Fucino Lake. Correlation of the marked environmental oscillation with one of the Dansgaard–Oeschger Events recorded by stable oxygen isotope records from the Greenland ice–cap is proposed. This follows an interval interpreted as having a cold dry climate and correlated with a Heinrich Event in the North Atlantic. Together the two records enable a multifactorial interpretation of the palaeoclimate changes that characterise the oscillation, providing additional insight to that obtained from studies of ice and ocean-sediment cores, especially with respect to the seasonality of temperature and precipitation. Copyright © 1999 John Wiley & Sons, Ltd.

Summary The relationship between the diversity of higher plant macrofossils in surface sediments of lakes and the surrounding vegetation is examined in two mountain regions; Grodalen in central Norway and the south-east Cairngorms in Scotland. Two lake sediment cores from each area were also analysed to examine vegetation history and to estimate changes in biodiversity through the Holocene. The diversity of present day vegetation in each region was estimated using both quadrat data and classified satellite images of the study areas. The mean surface sample macrofossil representation of species recorded in quadrats collected within 250 m of the lakes was c. 17%. This figure drops to only c. 2% when the satellite imagery of the same area is used to provide a maximal species list. The macrofossil data from the Norwegian cores show that deglaciation in this region occurred earlier on the mountain summit than in the valley and that the maximum tree line elevation was during the interval 9100–4400 14C yr BP. In the Cairngorms the maximum tree line elevation was prior to c. 4500 14C yr BP. The changes in higher plant diversity recorded at these sites through the Holocene show that c. 4000 14C yr BP the reduction in the tree line resulted in decreased β-diversity at higher altitudes but an increase at the lower altitude as the forest cover opened up. Under conditions of climatic warming it is likely areas that come to lie below the tree line will experience reduced diversity and that a permanent loss of biodiversity would result from a severe reduction in the area above the tree line.

Summary Pollen analytical, plant macrofossil and dendrochronological evidence together are used to reconstruct the late-glacial and Holocene vegetation history of the Highlands of Scotland. During the late-glacial period trees were almost, if not entirely, absent from the Highlands; vegetation analogous to that of Arctic regions of Europe today was present throughout the region. During the Holocene, in contrast, forest cover was extensive, with tree lines higher up the mountains and pine forests present even in the far northern Highlands during one millennium or so of especially favourable conditions with warmer, and sometimes drier, summers than today. During recent millennia the climate has been more similar to that of the present and the Highland vegetation was a mosaic of communities similar to that which we can find today in parts of western Scandinavia. During recent centuries extensive logging and deforestation, along with intensified grazing and management, have resulted in the present predominant...

Whereas fossil evidence indicates extensive treeless vegetation and diverse grazing megafauna in Europe and northern Asia during the last glacial, experiments combining vegetation models and climate models have to-date simulated widespread persistence of trees. Resolving this conflict is key to understanding both last glacial ecosystems and extinction of most of the mega-herbivores. Using a dynamic vegetation model (DVM) we explored the implications of the differing climatic conditions generated by a general circulation model (GCM) in “normal” and “hosing” experiments. Whilst the former approximate interstadial conditions, the latter, designed to mimic Heinrich Events, approximate stadial conditions. The “hosing” experiments gave simulated European vegetation much closer in composition to that inferred from fossil evidence than did the “normal” experiments. Given the short duration of interstadials, and the rate at which forest cover expanded during the late-glacial and early Holocene, our results demonstrate the importance of millennial variability in determining the character of last glacial ecosystems.

A high-resolution palynological study of a 51 m core from Lago Grande di Monticchio, southern Italy, has provided a palaeonvironmental record for the last glacill. A annual lamination based chronology, supported by radiometric and tephrochronological dates, provides an absolute timescale for this record that spans 76,300 years. Correlations are established between the pollen stratigraphy, the GRIP ice core δ18O record and foraminiferal assemblages from Atlantic core V23-81. Both Dansgaard-Oeschger and Heinrich events are reflected by changes in the pollen stratigraphy. Revised dates are estimated for Heinrich events H1–H6. A quantitative palaeoclimate reconstruction based upon the pollen data provides evidence of the climate changes in southern Italy associated with these and other fluctuations during the last glacial.

In southern Italy, vegetation contemporary with the end of the last glacial maximum, from 15,000 to 12,000 years ago, is shown by pollen-analysis to have been treeless and steppe-like in character. At 12,500 BP (years before present), Betula (birch) expanded into the steppe, quickly followed by Quercus (oak), Fagus (beech), Tilia (lime) and other tree genera of mesic forest. High percentages of Tilia point to a rich mesic forest that was contemporary with the ‘Allerod’ interstadial of northern Europe. A major decline in mesic trees with an accompanying return of Betula and steppe genera dated to 10,500 years ago identifies a ‘Younger Dryas ’ climatic reversal. Betula and steppe genera were replaced by forest of Quercus and other mesic trees, notably Ulmus (elm), as the Holocene began. In the later Holocene, ca. 4000 years ago, Abies (fir), Carpinus betulus (hornbeam) and Taxus (yew) appeared. Abies and Taxus became extinct locally about 2500 years ago, either because of climatic change, or perhaps because of the effects of early agriculture. The Full-glacial climate is thought to have been cold and summer-dry with mainly winter precipitation. The Lateglacial ‘Bolling-Allerod’ Interstadial was summer-wet and warm. The response-surface based climate reconstruction indicates an early Holocene climate with markedly colder winter conditions than today, about −5°C compared with 3.9°C today as a mean temperature for the coldest month. The annual temperature sum is reconstructed as somewhat higher than today, 3500 degree days as compared with a calculated value of 2900 for today. The later Holocene had a climate like today's. Rainfall, and variation in its seasonal distribution, has been a critical determinant of the vegetation cover. The fossil pollen record at Laghi Di Monticchio has been complemented by diatom and plant macrofossil studies which provide evidence of former lake environments as well as data on the upland forest. Lake levels remained high during the Full- and Lateglacial with encroachment of shore vegetation during the Holocene. The sediments also have an exceptionally rich record of tephra falls which are of importance in dating and core correlation. Twenty-one macroscopically visible tephras occur in sediments of the last 15,000 years.

Paleoecology can provide valuable insights into the ecology of species that complement observation and experiment-based assessments of climate impact dynamics. New paleoecological records (e.g., pollen, macrofossils) from the Italian Peninsula suggest a much wider climatic niche of the important European tree species Abies alba (silver fir) than observed in its present spatial range. To explore this discrepancy between current and past distribution of the species, we analyzed climatic data (temperature, precipitation, frost, humidity, sunshine) and vegetation-independent paleoclimatic reconstructions (e.g., lake levels, chironomids) and use global coupled carbon-cycle climate (NCAR CSM1.4) and dynamic vegetation (LandClim) modeling. The combined evidence suggests that during the mid-Holocene (;6000 years ago), prior to humanization of vegetation, A. alba formed forests under conditions that exceeded the modern (1961–1990) upper temperature limit of the species by ;5–78C (July means). Annual precipitation during this natural period was comparable to today (.700–800 mm), with drier summers and wetter winters. In the meso-Mediterranean to sub-Mediterranean forests A. alba co-occurred with thermophilous taxa such as Quercus ilex, Q. pubescens, Olea europaea, Phillyrea, Arbutus, Cistus, Tilia, Ulmus, Acer, Hedera helix, Ilex aquifolium, Taxus, and Vitis. Results from the last interglacial (ca. 130 000–115 000 BP), when human impact was negligible, corroborate the Holocene evidence. Thermophilous Mediterranean A. alba stands became extinct during the last 5000 years when land-use pressure and specifically excessive anthropogenic fire and browsing disturbance increased. Our results imply that the ecology of this key European tree species is not yet well understood. On the basis of the reconstructed realized climatic niche of the species, we anticipate that the future geographic range of A. alba may not contract regardless of migration success, even if climate should become significantly warmer than today with summer temperatures increasing by up to 5–78C, as long as precipitation does not fall below 700–800 mm/yr, and anthropogenic disturbance (e.g., fire, browsing) does not become excessive. Our finding contradicts recent studies that projected range contractions under global-warming scenarios, but did not factor how millennia of human impacts reduced the realized climatic niche of A. alba.

This paper evaluates the evidence for millennial-scale variability in pollen records of the last glacial (Marine Isotope Stages 4, 3, and 2; 73.5–14.7 calendar ka BP) from the European continent, taking into account information derived from long, continuous terrestrial records, the fragmentary northern European terrestrial record, and marine pollen records of the European continental margins. Pollen records from these numerous European sites provide evidence for multiple intervals of relatively warm and humid conditions during the last glacial, which promoted the establishment of grassland and shrub tundra in northwestern Europe, shrub- and forest-tundra in northeastern Europe, open boreal forest in central western Europe and the Alpine region, and open temperate forest in southern Europe. The northern limit for temperate forest development during these intervals was at ∼45°N, with a subsequent northward transition to tundra across a latitudinal band of ∼15° in western and central Europe, and a greater northward extension of boreal forest in eastern Europe, with boreal forest elements detected close to their present-day limits at ∼70°N. A much smaller number of sites with sufficiently high temporal resolution provide evidence that warming intervals correspond to millennial-scale variability as recorded in Greenland ice cores. A synthesis of sites providing high-resolution terrestrial and marine records from Europe is undertaken in order to examine geographical and temporal patterns in the expression of Dansgaard–Oeschger (D–O) cycles in the European vegetation. Detailed comparison of temperate forest development at these sites during four specific D–O cycles (D–O 16–17, 14, 12 and 8) reveals contrasts between vegetation response at southernmost European latitudes (below 40°N) and at latitudes above 40°N. At southernmost latitudes, marked forest development occurred during all four D–O cycles including D–O 16–17 and 8, while at latitudes above 40°N, forest development was stronger during D–O 14 and 12 than either D–O 16–17 or 8.

Aim: To obtain palaeobotanical evidence enabling evaluation of the viability of the hypothesis that the ‘oriental’ element of the Balkan flora reached south-east Europe from Turkey prior to the Holocene, probably via the Thracian Plain during a late Quaternary glacial stage but no later than the late Weichselian. Location: Ezero wetland, northern Thracian Plain, Bulgaria. Methods: We undertook analyses of pollen and microspores, plant macrofossils, wood fragments and molluscs recovered from sediments deposited in the Ezero wetland during the late Weichselian and Weichselian late-glacial. Sediment chronology was determined using radiocarbon age estimates. Results: Six metres of sediments were recovered from the basin, of which the lower 3 m, extending from c. 15,450 cal yr bp to the early Allerod, was analysed. A major hiatus occurred after c. 13,900 cal yr bp, the overlying sediments being of late Holocene age. Palaeobotanical evidence indicates predominantly open vegetation during the Weichselian late-glacial, although macrofossil remains of woody taxa demonstrate the local presence of patches of wooded steppe and gallery forest. Changes in the composition of the steppe vegetation, and in the nature of the sediments deposited in the basin, indicate changes in climatic conditions, especially in the hydrological regime and in the moisture available to vegetation. After an initially relatively moister phase, the final centuries of the late Weichselian were drier, as was a short interval that may correlate with the Older Dryas. Moister conditions characterize intervals corresponding to the Bolling and Allerod sub-units of the Weichselian late-glacial interstadial. Although the pollen evidence is thus consistent with that from previous studies of this period in south-east Europe and south-west Asia, indicating predominantly open steppe vegetation, the macrofossil evidence indicates the persistent local presence of woody taxa. The woody taxa recorded include Celtis tournefortii-type and Juniperus cf. J. excelsa, two taxa today characteristic of the wooded steppes of Anatolia and members of the ‘oriental’ element of the southern Balkan flora, as well as Rosaceae Subfams. Maloideae and Prunoideae, Alnus and Fraxinus. Main conclusions: The late Weichselian vegetation of the northern Thracian Plain included patches of wooded steppe that supported members of the ‘oriental’ element of the modern Balkan flora. The presence of such taxa renders viable the hypothesis that they could have reached south-east Europe from Turkey via the Thracian Plain during glacial times. Such hypotheses in historical biogeography can be evaluated critically using the evidence obtained from plant macrofossil analyses in combination with that from pollen analysis.

Maar- and crater lakes have, due to their specific formation and characteristic morphology, a high potential for the development and preservation of seasonally laminated sediments (varves). Within the Long Gang Volcanic Field (LGVF) in N.E. China eight maar- or crater lakes with water depths between 15 and 127 m are located. The seasonal climate with strong influence of the East Asian summer—and winter monsoon and the position of the LGVF at the northern margin of the East Asian dust storm tracks make these lakes key archives for palaeoclimatic studies. In the course of a site survey for a deep lake coring water samples and short sediment cores were investigated from all of these lakes. Detailed pollen investigations of the last 900 years from Lake Sihailongwan, an ideally bowl-shaped maar lake with 50 m water depth, revealed a very stable vegetation cover, except for the last 150 years which shows increasing anthropogenic influence. No evidence could be found in the pollen record for a climatic deterioration comparable to the European Little Ice Age. However, climatic variations can be revealed from high-resolution geochemical and sedimentological investigations on a decadal to annual scale.

International audience

The palynostratigraphic record from Lago Grande di Monticchio (S. Italy) is presented. An independent sedimentation-rate-based chronology provides an age of 101,530 calendar years BP for the lowermost sediments sampled. This chronology enables assessment of the rapidity of vegetation changes during the last glacial; contrary to widely held views the vegetation showed numerous large magnitude changes in periods of

Oxygen-isotope records from Greenland ice cores1,2 indicate numerous rapid climate fluctuations during the last glacial period. North Atlantic marine sediment cores show comparable variability in sea surface temperature and the deposition of ice-rafted debris3,4,5. In contrast, very few continental records of this time period provide the temporal resolution and environmental sensitivity necessary to reveal the extent and effects of these environmental fluctuations on the continents. Here we present high-resolution geochemical, physical and pollen data from lake sediments in Italy and from a Mediterranean sediment core, linked by a common tephrochronology. Our lacustrine sequence extends to the past 102,000 years. Many of its features correlate well with the Greenland ice-core records, demonstrating that the closely coupled ocean–atmosphere system of the Northern Hemisphere during the last glacial4 extended its influence at least as far as the central Mediterranean region. Numerous vegetation changes were rapid, frequently occurring in less than 200 years, showing that the terrestrial biosphere participated fully in last-glacial climate variability. Earlier than 65,000 years ago, our record shows more climate fluctuations than are apparent in the Greenland ice cores. Together, the multi-proxy data from the continental and marine records reveal differences in the seasonal character of climate during successive interstadials, and provide a step towards determining the underlying mechanisms of the centennial–millennial-scale variability.