Quantitative reconstruction of the last

Climate Dynamics (2005) 25: 625–637 DOI 10.1007/s00382-005-0045-0 P Tarasov Ỉ W Granoszewski Ỉ E Bezrukova S Brewer Ỉ M Nita ặ A Abzaeva ặ H Oberhaănsli Quantitative reconstruction of the last interglacial vegetation and climate based on the pollen record from Lake Baikal, Russia Received: 20 December 2004 / Accepted: 16 May 2005 / Published online: 13 August 2005 Ó Springer-Verlag 2005 Abstract Changes in mean temperature of the coldest (Tc) and warmest month (Tw), annual precipitation (Pann) and moisture index (a) were reconstructed from a continuous pollen record from Lake Baikal, Russia The pollen sequence CON01-603-2 (53°57¢N, 108°54¢E) was recovered from a 386 m water depth in the Continent Ridge and dated to ca 130–114.8 ky BP This time interval covers the complete last interglacial (LI), corresponding to MIS 5e Results of pollen analysis and pollen-based quantitative biome reconstruction show pronounced changes in the regional vegetation throughout the record Shrubby tundra covered the area at the beginning of MIS 5e (ca 130–128 ky), consistent with the end of the Middle Pleistocene glaciation The late glacial climate was characterised by low winter and summer temperatures (Tc $ À38 to À35°C and Tw$11– 13°C) and low annual precipitation (Pann$300 mm) P Tarasov (&) Institute of Geological Sciences, Palaeontology Department, Free University, Malteserstr 74-100 Building D, Berlin, 12249, Germany E-mail: paveltarasov@mail.ru Tel.: +49-30-83870280 Fax: +49-30-83870745 However, the wide spread of tundra vegetation suggests rather moist environments associated with low temperatures and evaporation (reconstructed a$1) Tundra was replaced by boreal conifer forest (taiga) by ca 128 ky BP, suggesting a transition to the interglacial Taiga-dominant phase lasted until ca 117.4 ky BP, e.g about 10 ky The most favourable climate conditions occurred during the first half of the LI Pann reached 500 mm soon after 128 ky BP However, temperature changed more gradually Maximum values of Tc $ À20°C and Tw$16–17°C are reconstructed from about 126 ky BP Conditions became gradually colder after ca 121 ky BP Tc dropped to $ À27°C and Tw to $15°C by 119.5 ky BP The reconstructed increase in continentality was accompanied by a decrease in Pann to $400–420 mm However, the climate was still humid enough (a$0.9) to support growth of boreal evergreen conifers A sharp turn towards a dry climate is reconstructed after ca 118 ky BP, causing retreat of forest and spread of cool grass-shrub communities Cool steppe dominated the vegetation in the area between ca 117.5 ky and 114.8 ky BP, suggesting the end of the interglacial and transition to the last glacial (MIS 5d) Shift to the new glaciation was characterised by cool and very dry conditions with Tc $ À28 to À30°C, Tw$14– 15°C, Pann$250 mm and a$0.5 W Granoszewski Carpathian Branch, Polish Geological Institute, Skrzatow 1, Krakow, 31-560, Poland E Bezrukova Æ A Abzaeva Institute of Geochemistry, Siberian Branch Russian Academy of Sciences, Favorsky street 1a, Irkutsk, 664033, Russia S Brewer CEREGE CNRS/University P Ce`zanne, UMR 6635, BP80, 13545, Aix-en-Provence cedex 4, France M Nita Faculty of Earth Sciences, University of Silesia, Bedzinska 60, 41-200, Sosnowiec, Poland H Oberhaănsli GeoForschungsZentrum, Potsdam, 14473, Germany Introduction Lake Baikal (455 m a.s.l.) is the largest fresh-water basin of the world It is situated in the inner part of Eurasia in the rift valley and is surrounded by mountain ridges The lake has a maximum length of 636 km, a maximum width of 79 km and a maximum depth of 1,620 m, which make it comparable to a sea The modern climate is continental and the dominant vegetation is boreal conifer forest (taiga) Lake Baikal is highly suitable for 626 Tarasov et al.: Quantitative reconstruction of the last interglacial vegetation and climate the investigation of past climates and environments due to its age and its location in a region that is controlled by the Siberian Anticyclone in winter, and by the Atlantic, Pacific and Polar air masses during the warm season Pollen records have been frequently used to obtain palaeoclimatic information from the Pleistocene lacustrine sediments (e.g Guiot et al 1989; Nakagawa et al 2002; Frogley et al 1999; Tzedakis et al 2002), and Lake Baikal is a perfect site for the collection of pollen and spores produced by regional vegetation The size of the basin decreases the effects of local or non-climatic factors on the composition of the pollen spectra, and mountain barriers surrounding Baikal Valley substantially minimise the role of long-distant pollen transport The bottom and coastal sediments of Lake Baikal were extensively studied during the 1990s (Hutchinson et al 1992; Bezrukova 1999; Bezrukova and Letunova 2001; Grachev et al 1997; Horiuchi et al 2000; Khursevich et al 2001; Prokopenko and Williams 2004; BDP Members 2004) However, little is known about the changes in vegetation and climate of the region during the last interglacial (LI) and the glacial-interglacial transitions The last interglacial (LI) is conventionally attributed to the marine isotope stage (MIS) 5e, dated to ca 130– 116 ky BP (Martinson et al 1987; Kukla 2000; Goldberg et al 2001) It is the most recent complete warm phase with a negligible role of man and with well-defined differences in the earth orbital parameters from those of today (Berger 1978; Imbrie et al 1984) The recent interest in LI climate and environments is driven by several different theoretical and practical interests and has resulted in a number of publications (see among others Harrison et al 1995; Cheddadi et al 1998b; Karabanov et al 2000; Kubatzki et al 2000; Kukla 2000; Turner 2000; Rioual et al 2001; Klotz et al 2003; Kuăhl and Litt 2003) A predominant interest in the LI is in the eld of comparisons with the present Holocene warm interval (Kuăhl and Litt 2003) Until the nineteenth century, humans did not influence vegetation and environments in the Baikal region as much as they did in Europe and China since the Early-Mid-Holocene Thus, the reconstruction of natural climatic variability during the LI and the Holocene is possible using pollen records from Lake Baikal For the modelling community, these results provide an opportunity to examine the effects of orbitally induced extreme insolation anomalies on regional climates and vegetation (Harrison et al 1995) and are essential for data-model comparison and for evaluation of the models’ capability to simulate past climate (Harrison et al 1995; Kubatzki et al 2000) The main goal of this study was to carry out the first quantitative climate reconstruction for the Baikal Region, for the interval from 130 ky to 114.8 ky BP This is based on the detailed pollen record from the CON01603-2 deep-water core (53°57¢N, 108°54¢E) from the northern part of Lake Baikal (Granoszewski et al 2005) and an extensive surface pollen data set from northern Eurasia (see Tarasov et al 1998a, b; Andreev et al 2003; Bigelow et al 2003 and references therein) The results are discussed in terms of regional and global-scale environmental and climatic changes Data and reconstruction methods 2.1 Regional setting In the present paper we discuss the pollen record from the core CON01-603-2 (53°57¢N, 108°54¢E) taken from a water depth of 386 m in the northern part of Lake Baikal, close to its east coast (Fig 1) The study area consists of the lake itself, the surrounding Baikal mountain ridges (1,700–2,500 m a.s.l.) and intermountain depressions (500–700 m a.s.l.) Differences in elevation and slope exposition cause substantial variations in temperature and precipitation The mean temperature is 14–16°C in July, and À22°C to À26°C in January Mean annual precipitation in the intermountain depressions is 250–300 mm/year, while this may be over 700 mm/year in the mountains The driest area is Olkhon Island (161 mm/year), and the slopes of KhamarDaban Ridge south of Baikal receive up to 1,300 mm/ year (Alpat’ev et al 1976; Galaziy 1993) At the coring site, precipitation values are around 400 mm/year Up to 80–85% of the annual precipitation falls during the warm season Winter precipitation is associated with Atlantic cyclones entering the area along the western periphery of the Asian High Summer precipitation is associated with cyclones developing at the Mongolian branch of the Polar Front and with Pacific monsoontype circulation amplified by the mountain relief (Alpat’ev et al 1976) The spatial distribution of the regional vegetation shows a clear influence of the climate The more humid slopes in the west are covered with taiga forest dominated by Pinus sibirica, Abies and Picea, whereas Larix and Pinus sylvestris forests occupy the dryer eastern slopes (Bezrukova 1999) Betula and Populus are common species of the lower-elevated secondary forests Dry depressions are mainly occupied by steppe patches and shrubby sub-alpine associations represented by Pinus pumila, Alnus fruticosa and Betula middendorfii cover the upper parts of the mountain slopes above 1,800 m (Dylis et al 1965; Molozhnikov 1986) 2.2 Fossil record The analysed part of the CON01-603-2 core (53°57¢N, 108°54¢E) between 725.5 cm and 608 cm depth (further referred to as the ‘Continent record’) consists mainly of diatomaceous, silty sediments, with fine and coarse lamination in its upper part (690–608 cm) (for detailed description of the core lithology see Charlet et al 2005) Low tectonic activity of the area around the coring site inferred from seismic data and the undisturbed finegrained character of the sediment indicate continuous Tarasov et al.: Quantitative reconstruction of the last interglacial vegetation and climate 627 Fig Map of northern Eurasia with Lake Baikal, the Continent pollen record and the reference pollen data set of 1,173 surface spectra used in the climate reconstruction The star indicates the coring site (CON01-603-2: 53°57¢N, 108°54¢E) and closed circles indicate surface pollen sampling sites 60 N 45 N 60 E sedimentation and make the CON01-603-2 core appropriate for palaeoclimatic reconstructions (Rioual and Mackay 2005) Based on the age model developed by correlation of the palaeomagnetic record from the core with a reference curve from ODP Site 984 (Channel 1999), the sediment accumulated during an interval between 130 ky and 114.8 ky BP (for complete results see Demory et al 2005) Detailed pollen analysis has been recently performed on the same section (for complete results see Granoszewski et al 2005) Samples of 2.5– 3.5 cm3 in volume were taken from 111 levels and treated using a standard procedure In the majority of samples, 450–1,900 terrestrial pollen grains were counted, with the exception of the upper cm layer, where 86 to 330 terrestrial pollen grains were counted due to low pollen concentration The sampling distance varies between 0.5 cm to cm, with an average of cm Estimated time intervals between the analysed levels vary from 45 years to 376 years However, for the main part of the record (between 716 cm and 618 cm), the average time interval is 116 years A simplified pollen diagram of the Continent record (Fig 2) shows the pronounced changes in pollen composition between the local pollen assemblages zones (LPAZ) described in Granoszewski et al (2005) The high values of A fruticosa and Betula sect Nanae/ Fruticosae pollen suggest that tundra communities with shrub alder, and shrub and dwarf shrub birch species were widely distributed on the landscape during the late phase of the Taz (Saale) Glaciation, between 130 ky and 128 ky BP, leading into the full interglacial Boreal tree species (e.g spruce–Picea obovata and birch–Betula sect Albae) start to play an important role within the regional vegetation with the onset of the interglacial conditions at 127.9 ky BP During the interval between 128 ky and 117.4 ky BP, the pollen spectra are dominated by arboreal pollen, suggesting that the region became forested under the interglacial climate The highest percentages of Abies sibirica and Picea pollen 120 E are registered during the period between 128 ky and 126 ky BP, indicating favourable growing conditions for fir- and spruce-dominated taiga at that time Birch was probably the most important component in the forests between 126 ky and 125.5 ky BP During the second half of the interglacial, the high percentages of Pinus sibirica-type and Pinus sylvestris-type pollen suggest a major spread of Siberian pine in the region after ca 125.5 ky BP, and of Scots pine after ca 123.7 ky BP Pollen data and the age model suggest that the interval leading out of the full interglacial was relatively short A decrease in arboreal pollen to less than 30% was accompanied by a substantial increase in Artemisia, Chenopodiaceae, Poaceae and other herbs This change occurred after 117.7 and took about three to four hundred years The vegetation during the early phase of the last glaciation, between 117.4 ky and 114.8 ky BP (MIS 5d), was similar to modern cool steppe However, the presence of Salix, Betula sect Nanae/Fruticosae, Cyperaceae pollen and Sphagnum spores (Fig 2) suggest that tundra-like associations probably occupied river valleys and upper mountain slopes, habitats with better moisture conditions 2.3 Biome reconstruction method A quantitative method of pollen-based biome reconstruction (Prentice et al 1996) allows us to investigate further and confirm the qualitative interpretation of the pollen records It is based on the objective assignment of pollen taxa to plant functional types (PFT) and to biomes on the basis of the modern ecology and distribution of plants The bioclimatic limits of PFTs and biomes defined in the BIOME1 model (Prentice et al 1992) can be used to interpret results of pollen-based biome reconstruction in climatic terms (Tarasov et al 1998a, b, 2000) 628 Tarasov et al.: Quantitative reconstruction of the last interglacial vegetation and climate Fig Simplified pollen diagram from the Continent record The equation used to calculate the affinity scores for all pollen samples was published by Prentice et al (1996): q Aik ẳ Rj dij fmaxẵ0; pjk hj ފg; PFTs and to biomes using the taxon-PFT-biome matrix published by Tarasov et al (1999) where Aik is the affinity of pollen sample k for biome i; summed for all taxa j; dij is the entry in the biome versus taxon matrix for biome i and taxon j; pjk are the pollen percentages, and hj is the universal threshold pollen percentage of 0.5% suggested for minimisation of possible noise mainly due to long-distant transport or redeposition of exotic pollen grains (Prentice et al 1996) For a given pollen spectrum, the biome with the highest score or, when several biomes have the same score, the one defined by a smaller number of taxa is then assigned Pollen taxa from the former Soviet Union and Mongolia were assigned to the regional PFTs and to biomes by Tarasov et al (1998a, 2000) and successfully tested with modern data from northern Eurasia In addition, Tarasov et al (1998b) proposed a modification to distinguish cool steppe from warm steppe biome, by using the presence of boreal tree and arctic-alpine shrub taxa as additional criteria to assign herbaceous pollen taxa to the appropriate PFTs This is based on the reasonable assumption that the temperature requirements of the herbaceous taxa may be indicated by associated tree and shrub taxa, which can be identified by pollen analysts at a higher taxonomic level Full details of the method are presented in Tarasov et al (1998b, 1999) In total, 55 terrestrial pollen taxa identified in the Continent record (Table 1) were attributed to appropriate Quantitative pollen data may be used to reconstruct climatic variables that control the distribution of the regional vegetation, using a statistical approach known as the ‘‘best modern analogue’’ (BMA) method (Guiot et al 1989; Guiot 1990) In our study, we attempted to reconstruct the climatic variables that limit the spatial distribution of the main vegetation types (or biomes) in northern Eurasia and, in addition, are used in vegetation modelling (Prentice et al 1992) These are the mean temperature of the coldest month (Tc), the mean temperature of the warmest month (Tw) and the moisture index (a), calculated as the ratio of actual to equilibrium evapotranspiration (Prentice et al 1992) Annual precipitation (Pann) was also chosen, as its reconstruction facilitates the interpretation of moisture index, which may be influenced by changes in temperature and/or associated changes in precipitation (Tarasov et al 1999) The selected climatic variables were accurately reconstructed from the modern pollen spectra in northern Eurasia (Klimanov 1984; Tarasov et al 1999) A comparison between pollen-inferred and observed values of climatic variables performed at continental and at regional scale (Cheddadi et al 1998a, 1998b; Nakagawa et al 2002; Andreev et al 2003; Wohlfarth et al 2004) shows a reasonably good correlation between compared data sets 2.4 Best modern analogue (BMA) method Tarasov et al.: Quantitative reconstruction of the last interglacial vegetation and climate 629 Table Terrestrial pollen taxa identified in the Continent record (Granoszewski et al 2005) and used in the biome and climate reconstructions Abies, Acer*, Alnus (arboreal), Alnus fruticosa, Apiaceae, Artemisia, Asteraceae undif., Betula sect Albae, Betula sect Nanae+Fruticosae, Boraginaceae*, Brassicaceae, Cannabis, Carpinus*, Cariophyllaceae, Chenopodiaceae, Corylus, Cyperaceae, Ephedra, Ericales undif., Fabaceae, Frangula*, Fraxinus excelsior*, Gentiana*, Juniperus, Lamiaceae*, Larix, Liliaceae*, Onagraceae*, Picea, Pinus subgen Diploxylon, Pinus subgen Haploxylon, Plantago*, Plumbaginaceae*, Poaceae, Polemonium*, Polygonum, Populus, Primulaceae*, Quercus (deciduous), Ranunculaceae, Rosaceae, Rubiaceae, Rubus chamaemorus*, Rumex*, Salix, Sambucus*, Saxifraga, Scrophulariaceae, Thalictrum, Tilia*, Ulmus, Urtica, Valeriana* Taxa, which percentages in the pollen spectra not exceed 0.5% and not influence results of biome reconstruction indicated with a star Calculated root mean square error of prediction is 0.6°C for Tw and 34 mm for Pann and mean absolute error (bias) is 0.4°C and 27 mm for Tw and Pann, respectively (Solovieva et al 2005) Statistical errors of the reconstruction estimated by the RMSEP software (Line and Birks unpublished program) are close to the Klimanov (1984) estimations (e.g.±0.6°C for Tw, ±1.0°C for Tc, and ±25 mm for Pann) Another way to estimate error bars for the reconstructed climate variables is through the calculation of uncertainty range (Nakagawa et al 2002; Kuăhl et al 2002) When applied to the pollen-based reconstruction results, this method gives much larger errors For the reconstructed mean July and January temperatures of the LI at Groăbern site (Germany), 90% uncertainty range changes from 58C during the climatic optimum to 13–22°C at the beginning and at the end of the LI (Kuăhl and Litt 2003) When applied to a fossil pollen record, the BMA method uses a chord distance to determine the similarity between each fossil pollen spectrum and each spectrum in the reference modern data set (Guiot 1990) In the present study, the eight spectra that have the smallest chord distance are considered as the closest modern analogues of the analysed fossil spectrum The climate for each fossil sample is then calculated as the weighted average of the climatic variables of the selected best analogues, with the inverse chord distance as weights (Nakagawa et al 2002) This gives the reconstructed value, which is treated as the most probable The error bars for the reconstructed values are defined by the climatic variability in the set of eight BMAs and are frequently asymmetric as they take into account the fact that the selected analogue values are not normally distributed around the most probable value These confidence limits include possible errors in the modern climate observations, the natural climatic variability for the given pollen assemblage and the effect of non-climatic factors (Cheddadi et al 1998a) In the present study, we used the reference pollen data set from the large area of former Soviet Union and Mongolia (northern Eurasia) This includes 1,110 modern surface spectra, with all main vegetation types of the region well represented (Tarasov et al 1998a) The data set has already been used for quantitative vegetation and climate reconstruction in northern Eurasia (e.g Tarasov et al 1998a, 1999, 2000; Bigelow et al 2003; Andreev et al 2003; 2004) To improve the representation of the Baikal region, we have added 63 modern pollen spectra collected around Lake Baikal to the reference data set (Bezrukova, unpublished data) The spatial distribution of the 1,173 modern pollen spectra is shown in Fig In the reference data set, the sum of 81 terrestrial pollen taxa commonly identified in the pollen records from northern Eurasia is taken as 100% This taxa list includes all terrestrial taxa identified in the Continent pollen record (Table 1) and used in the climate and biome reconstruction Modern climate values at each of 1,173 modern pollen sampling sites have been calculated from the high-resolution global climatology data base that provides the 30 year average (1961–1990) of the monthly means of principal meteorological parameters on a 10 grid (New et al 2002) This has been already used for the reconstruction of the Eemian climate from European pollen and plant macrofossil records (Kuăhl et al 2002; Kuăhl and Litt 2003) Both biome reconstruction and BMA approaches have their own set of assumptions and shortcomings The biomization method provides only semi-quantitative and indirect climate information, but is ‘‘closer’’ to the actual vegetation and does not suffer as much from the no-analogue problem that more quantitative approaches The modern-analogue technique may suffer from that problem, but it can provide robust climate reconstructions,when good analogues exist The chosen approaches, thus, complement one another Results and interpretations 3.1 Biome reconstruction The results of the biome reconstruction (Fig 3) complement the qualitative interpretation of the Continent pollen record (Granoszewski et al 2005), suggesting three main phases in the development of the regional vegetation between ca 130 ky and 114.8 ky BP The biomization method is based on the idea that the biome with the highest score has a greater likelihood to be dominant in the vegetation close to the site (Prentice et al 1996) Taking into account the large area of the Lake Baikal catchment contributing pollen to the Continent record, we can assume that reconstructed change in the dominant biomes from tundra to taiga, and later taiga to cool steppe, reflects regional changes in the vegetation The capability of the method to reconstruct steppe, tundra and taiga biomes in northern Eurasia was 630 Tarasov et al.: Quantitative reconstruction of the last interglacial vegetation and climate Fig Biome scores reconstructed from the Continent pollen record: a time series of individual biome scores The arrows mark periods where cool conifer forest biome showed high affinity scores; b the dominant biome in the Baikal region; c pollen zones tested using surface pollen data from the region (Tarasov et al 1998a) Among 303 surface pollen spectra collected from the sites with natural steppe, tundra or taiga vegetation, steppe was correctly reconstructed in 97%, tundra in 72% and taiga in 89% of the cases The biomization method (Prentice et al 1996) does not allow the reconstruction of transitional vegetation types (e.g forest-steppe or forest-tundra) However, additional information can be obtained by examining the relative values of forest and non-forest biome scores (Fig 3) For example, through the whole late Glacial and interglacial time, tundra scores are greater than cool steppe, suggesting generally humid environments in the region Taiga scores are higher during the early Glacial time than those in the late Glacial, and are comparable to tundra scores This may suggest that the dominance of steppe around Baikal was not absolute and both tundra and boreal forest communities co-existed locally in the moist habitats The biome reconstruction results confirm the position of the lower and upper boundaries of the LI in the region The lower boundary of the LI is placed at the level 715.75 cm, dated to ca 128 ky BP (Fig 3) when the scores of taiga biome exceed those of the tundra Similarly the upper limit of the interglacial can be placed at 618.5 cm (ca 117.35 ky BP), when the dominant vegetation type becomes cool steppe Also of note is the reconstruction of cool conifer forest (southern taiga) biome at several levels during the first half of the interglacial between 126.8 ky and 120.8 ky BP At these levels (Fig 3), the scores of the cool conifer forest slightly exceed the scores of the taiga; mainly because of Ulmus pollen (>0.5%) was recorded in the pollen spectra In the BIOME1 model (Prentice et al 1992), elm is included in the cool temperate tree PFT, which requires warmer winter and summer temperatures than boreal evergreen and deciduous tree species Ulmus is currently present in the regional vegetation as Ulmus pumila—a tree or shrub growing in the forest-steppe area of Russia and Mongolia south of Lake Baikal (Alpat’ev et al 1976; Bezrukova 1999; Gunin et al.1999) U pumila usually grows at the floodplain terraces along the big rivers, where climate conditions are milder The first occurrence of cool conifer forest biome, which precedes by some time, the peak in taiga scores, may suggest that regional temperatures had already reached present-day values by ca 126.8 ky BP, while the complete afforestation of the basin took longer due to inertia in the vegetation response Quantitative information derived from independent records of climate may help to confirm this hypothesis The reconstructed changes in the regional biome distribution observed in the Continent record can be explained by changes in temperature and available moisture According to the bioclimatic limits used in the BIOME1 model (Prentice et al 1992), the change from continental tundra to taiga requires an increase in the mean temperature of the coldest month to above À35°C Replacement of continental taiga and cold deciduous forest by cool steppe vegetation, requires a decrease in the moisture index to below 0.65, and a mean temperature of the warmest month that does not exceed 22°C The appearance of cool temperate broadleaf elements (Corylus, Tilia and Ulmus) in the vegetation requires mean temperature of the coldest month above À19°C Thus, the biomization neatly underlines the nature of the vegetation response to moisture and temperature conditions in the study area 3.2 Quantitative climate reconstruction The results of the quantitative reconstruction of temperature and precipitation through the Continent record are shown in Fig The earlier part of the record (ca 130–128 ky) reflects the climate of the end of the Middle Pleistocene glaciation, and the transition to interglacial conditions The late glacial climate was characterised by Tc $ À38 to À35°C, Tw$11–13°C and Pann$300 mm The climate was rather humid (a$1) due to relatively low summer temperature and low evaporation Such conditions were favourable for the Tarasov et al.: Quantitative reconstruction of the last interglacial vegetation and climate 631 Fig Climate variables reconstructed from the Continent pollen record using BMA approach Most probable values (open circles) are based on the weighted average of the selected analogues and confidence intervals (horizontal lines) are taken from the climatic distribution of the selected analogues Three-point moving averages (smoothed lines) are shown for each reconstructed variable spread of tundra associations dominated by shrub alder and shrub birch in the region Between 130 ky and 129 ky BP, the assemblages consist of 50–70% A fruticosa (Fig 2) A plot of modern A fruticosa pollen percentages in the Eurasian surface pollen spectra against modern climatic variables shows that high percentages of this taxon, attributed to the arctic-alpine shrub PFT (Prentice et al 1992), appear under a very severe climate in the tundra of northern and eastern Siberia (Fig 5a) In Fig 5c Alnus pollen in surface spectra located south of the dashed line in forest and forest-steppe vegetation zones, originated from tree forms of alder (e.g Alnus glutinosa, A incana), while large values of Alnus pollen found north of this line (e.g in forest-tundra and tundra) were mainly produced by its shrub forms (e.g A fruticosa) The onset of the interglacial, marked by a transition from tundra to boreal conifer forest (taiga), was associated with a rather sharp rise of precipitation followed by a gradual increase in winter and summer temperatures Reconstructed annual precipitation exceeds 500 mm at the level dated to 127.93 ky BP Subsequently, precipitation reached the highest values observed in the record, fluctuating between 500 mm and 600 mm until 124.5 ky BP Temperature changed more gradually Maximum values of Tc > À22°C and Tw$16–17°C are reconstructed between ca 126.5 ky and 120.7 ky BP, during the interval characterised by a noticeable spread of fir and elm in the region Abies sibirica, attributed to the boreal conifer PFT (Prentice et al 1992), is a taxon which requires high levels of available moisture, winter temperatures and soil richness (Gunin et al 1999) In the Continent pollen diagram (Fig 2), the highest values of Abies pollen (16%) are recorded short before 126 ky BP Comparable values of Abies pollen in the surface pollen data set appear in a relatively mild climate (Fig 5b) Abies is a minor pollen type with highest percentages recorded in the middle and southern taiga zone (Fig 5d) Slightly higher percentages of Abies pollen are recorded at three locations in the high Arctic (Fig 5d) are due to long distant transport along the valleys of the great Siberian rivers: Ob’, Yenisej and Lena A cooling occurred after ca 121 ky BP Our reconstruction suggests a drop in Tc to $ À27°C and in T w to $15°C by 119.5 ky BP The increase in continentality of the regional climate was accompanied by a decrease in Pann to $400–420 mm However, the climate was still humid enough (a$0.9) to support the growth of taiga forest in the region A change to a dryer climate is reconstructed after ca 118 ky BP The change was rather quick and caused a retreat of forest vegetation and a spread of cool grassshrub communities Cool steppe dominated the vegetation in the area between ca 117.5 ky and 114.8 ky BP, suggesting the end of the LI and the onset of the Late Pleistocene glacial conditions (MIS 5d) The shift to the new glaciation was characterised by cooler and very dry conditions with Tc $ À28 to À30°C, Tw$14–15°C, Pann$250 mm and a$0.5 632 Tarasov et al.: Quantitative reconstruction of the last interglacial vegetation and climate Fig A plot of A fruticosa (a) and Abies (b) pollen percentages observed in the surface pollen spectra from northern Eurasia against modern climatic variables Modern pollen distribution maps for Alnus (c) and Abies (d) help to visualise the location of the best analogues for the fossil samples with highest values of A fruticosa (pollen zone CK-1) and Abies (pollen zone CK-3) The star indicates the Continent site (CS) The longer-term trends (Fig 4) in all reconstructed climatic variables suggest generally stable warm and humid conditions during the main part of the interglacial and a rapid shift into a cold phase The general trends are complicated by several short-term fluctuations A marked depression is shown in all four parameters around 126–125.7 ky BP, corresponding to a sharp decrease in Abies pollen The decrease in temper- atures and annual precipitation at ca 125 ky BP and 124.1 ky BP are both restricted to single levels and should be treated with caution The interval between 124 ky and 120.7 ky BP is characterised by stable and high summer temperature, a progressive decrease in winter temperature and in precipitation and a flat depression in the moisture availability curve, suggesting an increase in continentality This period of a relative Tarasov et al.: Quantitative reconstruction of the last interglacial vegetation and climate moisture deficit was associated with a major spread of P sylvestris in the region The increase in winter temperature and in annual precipitation reconstructed for the upper two levels, were based on counts of only 86 and 91 terrestrial pollen grains, and may not be very reliable Discussion and conclusion The last interglacial, known also as Eemian interglacial in West Europe, Mikulino in East Europe and Kazantsevo in Siberia, is well recognised in the Pleistocene succession of warm and cold intervals, controlled by changes in the earth orbital parameters and associated changes in solar insolation (Berger 1978; Imbrie et al 1984; BDP 2004) In Europe and West Siberia, sediments attributed to the LI on the basis of bio- and lithostratigraphy and geomorphology have been intensively studied (e.g Velichko 1984; Zagwijn 1996; Turner 2000 and references therein) Available pollen and plant macrofossil records have been used to reconstruct vegetation and climate of the so called ‘‘climatic optimum’’—the interval with a major representation of thermo- and hygrophilous taxa in pollen and plant macrofossil assemblages assumed to be synchronous across at least the northern hemisphere extra-tropics (e.g Grichuk 1984; Frenzel et al 1992; Velichko et al 2002) In the pollen records from Europe, this phase can be traced by a maximum in Carpinus pollen (Grichuk 1984; Turner 2000) and in Siberia by a maximum in Abies pollen curve (Grichuk 1984; Velichko et al 2002; BDP Members 2004) The relatively humid Carpinus phase has been dated to 119–123 ky BP in the pollen record from Groăbern, Germany (Kuăhl and Litt 2003) and to 122.6124.8 ky BP in the record from NW Greece (Tzedakis et al 2002) In the Continent record, the peak of Abies pollen maximum is dated to ca 126.2 ky BP, showing no evidence that climate was drier during the early half of the interglacial in the Baikal region In strong contrast with earlier studies (e.g Grichuk 1984; Frenzel et al 1992; Velichko et al 2002), the present study clearly establishes the spatial heterogeneity of the LI climate across northern Eurasia Pollen records from East Siberia and Central Asia attributed to the LI are scarce In the Baikal region, palaeoclimatic maps of the LI (Frenzel et al 1992) based on the limited number of botanical records (Rindzyunskaya and Pakhomov 1977) and modified indicatorspecies approach (Grichuk 1969), suggested that winter and summer temperatures were higher than present, by 4–6°C and 2–2.5° respectively, and precipitation was ca 100 mm above the modern levels during the spread of Abies–Picea taiga with broadleaf elements, which fits well with our results A recent multidisciplinary study of glacial-interglacial sections from Bol’shoi Lyakhovsky Island (73°20¢N, 141°30¢E) complimented by IRSL and 230 Th/U datings (Andreev et al 2004) indicated that herbaceous tundra was replaced by a shrubby tundra associations with A fruticosa and Betula nana s.l during 633 the ‘LI optimum’ Pollen-based climate reconstruction reveals the mean July temperatures were 4–5°C higher than the present during that time, suggesting that summer warming was more pronounced in northern Siberia than in its southern part, in agreement with the earlier interpretations (Frenzel et al 1992; Velichko et al 2002) A major fall in biogenic silica content and in diatom abundance recorded in BDP-96-2 core from the underwater Academician Ridge is interpreted as result of cooling occurred about 121–120 ky BP (Karabanov et al 2000) Our pollen-based reconstruction also suggests a cooling occurred after ca 121 ky BP with a drop in mean temperature of the coldest month to $ À27°C and in mean temperature of the warmest month to $15°C by 119.5 ky BP (Fig 4) More or less gradual decrease in winter temperature by 10°C at the end of MIS 5e (Fig 4) is also comparable with that found by Kuăhl and Litt (2003) and Cheddadi et al (1998b) for the French, Polish and German pollen records Our results show an agreement with the interpretation of the LI as epoch with a relatively stable climate compared to the Holocene (McManus et al 1994; Turner 2000; Kukla 2000; Kuăhl and Litt 2003) Lowamplitude and short-term oscillations, such as that reconstructed at ca 126 ky BP (major decrease in Abies pollen), cannot be excluded However, mid-Eemian severe cold episode suggested on the basis of sharp oscillation in biogenic silica and in diatom abundance recorded in the Lake Baikal sediment (Karabanov et al 2000) does not appear in our reconstruction This result is consistent with recent quantitative reconstructions from continuous pollen sequences from western and central Europe (Cheddadi et al 1998b; Rioual et al 2001; Kuăhl and Litt 2003) as well as with earlier interpretations of the LI climate based on marine and ice cores (Jouzel et al 1993; McManus et al 1994) The diatom records, as most of the other biostratigraphic evidences, are often susceptible to alternative interpretations and could reflect climate changes, but sedimentation processes, changing water chemistry may also influence them (Battarbee et al 1998) Bearing in mind such a possibility, the mid-Eemian short-term oscillation registered in the diatom records from Lake Baikal probably can be also explained by another reason rather than broad-scale temperature oscillation Comparison with the output of model simulations based on changes in solar forcing (Harrison et al 1995; Kubatzki et al 2000) shows that results of palaeonvironmental reconstructions cannot be satisfactory explained by insolation changes alone In the Baikal region, the simulated anomalies are +8°C in summer and À1°C in winter at 125 ky BP (Harrison et al 1995) Dry conditions simulated as a consequence of warmerthan-present summers caused the northward extent of steppe vegetation in central Asia The seasonal differences become less pronounced (À4°C and +1°C) at the end of the interglacial ca 115 ky BP due to the attenuation of the insolation anomalies The Continent pollen 634 Tarasov et al.: Quantitative reconstruction of the last interglacial vegetation and climate record presented here does not support this scenario However, the CCM1 simulations (Harrison et al 1995) show results that are closer to the palaeoenvironmental reconstruction from Bol’shoi Lyakhovsky Island, suggesting that the model underestimates the effect of the decreased thickness and extent of the Arctic sea ice on the winter warming in Eurasian mid-latitudes Kubatzki et al (2000) using AGCM ECHAM-1/LSG and the model of intermediate complexity CLIMBER-2 performed a number of sensitivity experiments checking role of CO2 levels (pre-industrial versus ‘‘present’’) and vegetation feedback in the climate at 125 ky BP Their conclusion that ‘‘interactive vegetation turns out to be capable of modifying the initial climate signals, leading especially to warmer winters in large parts of the northern hemisphere, as indicated by palaeodata’’ is important, although not surprising, as was shown Texier et al (1997) Figure presents climatic variables reconstructed from the Continent record (Fig 5a–c) plotted against temporal variations in d18O (&, VPDB) of Dongge Cave (China) stalagmite D3 (Yuan et al 2004), which characterise the Pacific Monsoon precipitation in the lowmiddle latitudes (Fig 6d) Changes in summer (June, July, August) insolation, considered to drive northern hemisphere temperature change and the strength of the summer monsoon are also shown (Fig 6d) Abrupt changes in 18O/16O and available 230Th ages (Yuan et al 2004) suggest that the LI monsoon started at ca 129.3±0.9 ky BP and had ended by 119.6±0.6 ky BP Both levels are easily distinguished in the precipitation curve from the Continent record (Fig 6a) This agreement suggests that our age-depth model is reliable, and indicates a relationship between precipitation rise in the Baikal region and strengthening of the summer monsoon circulation during the LI Tarasov et al (2002) found a similar pattern in the Holocene pollen sequences from Buryatia south of Lake Baikal Modern observations in the region show that July and August are the most humid months (Climatic atlas of Asia 1981) At this time of the year, the westerly flow weakens and the sub-meridional circulation of monsoon type is relatively Fig Climatic variables reconstructed from the Continent record (a–c) plotted against (d) temporal variations in summer (June, July, August) insolation and d18O (&, VPDB) of Dongge Cave (China) stalagmite D3 (Yuan et al 2004) and against (e) change in the sea level and equivalent ice volume during the LI (Lambeck and Chappell 2001) Tarasov et al.: Quantitative reconstruction of the last interglacial vegetation and climate active (Zhukov 1965) Cyclones bring warm and relatively humid subtropical air from south-east and cause heavy rains which may last several days (Alpat’ev et al 1976) Figure 6e shows the change in sea level and equivalent ice volume during the LI The estimated global sea level (Lambeck and Chappell 2001) was ca 40 m below modern by 130 ky BP and reached the highest point by ca 126.5 ky BP Higher than present sea level and decrease in the ice volume would imply a general decrease in continentality of the central Asian climate In the Continent record, this period corresponds to the highest winter temperatures of the whole interglacial (Fig 6c) After 126 ky BP the progressive decrease in winter temperature is matched by a lowering in sea level and increase in ice volume, which becomes especially pronounced after 118 ky BP The general trend in the summer temperature curve (Fig 6b) reconstructed from the Continent record follows changes in the mid-latitude solar insolation (Fig 6d) High sea level associated with decreased ice volume appears to have had a greater impact on the Siberian vegetation during the LI than direct effect of lower-than-present winter insolation Warmer-than-present winters in Euarsian Arctic as well as in Siberia during the high sea level stand can be explained by combined effect of different factors, e.g warm currents, higher cyclonic activity along the Polar Front, weakening of the Siberian winter Anticyclone and increased warming effect of the Westerly flow in winter However, the real contribution of each forcing or their combination to the reconstructed environmental change can be only evaluated with the help of modelling sensitivity experiments In conclusion, the vegetation changes derived from the Continent pollen record can be satisfactory explained by reconstructed changes in summer and winter temperatures and in available moisture during 130– 115 ky BP During the LI between 128 ky and 117.4 ky BP, the vegetation around Lake Baikal was dominated by forests of taiga type, which are associated with a generally warm and wet climate An abrupt rise in precipitation occurred at the same time as the onset of the summer monsoon, dated to ca 129.3 ky BP, and there is a significant decrease in precipitation synchronous with an attenuation of the monsoon circulation at 119.6 ky BP Reconstructed changes in winter temperature correlate well with changes in the sea level and global ice volume, while the summer temperatures derived from the Continent record track changes in the summer insolation Acknowledgements This paper is part of the CONTINENT research project (http://www.continent.gfz-potsdam.de/) supported by the European Commission under the Fifth Framework Programme (Contract no EVK2-2000-00057) W Granoszewski was financed through this project P Tarasov acknowledges Alexander von Humboldt Foundation granted his research fellowship in Alfred Wegener Institute, Potsdam We are grateful to P.J Bartlein, T Nakagawa and J.-C Duplessy, for the critical review and suggestions, which helped to improve this paper 635 References Alpat’ev AM, Arkhangel’skii AM, Podoplelov NY, Stepanov AY (1976) Fizicheskaya geografiya SSSR (Aziatskaya chast’) (in Russian) Vysshaya Shkola, Moscow Andreev AA, Grosse G, Schirrmeister L, Kuzmina SA, Novenko EY, Bobrov AA, Tarasov PE, Ilyashuk BP, Kuznetsova TV, Krbetschek M, Meyer H, Kunitsky VV (2004) Late Saalian and Eemian palaeoenvironmental history of the Bol’shoy Lyakhovsky, Island Laptev Sea region, Arctic Siberia Boreas 33:319–348 Andreev AA, Tarasov PE, Siegert C, Ebel T, Klimanov VA, Melles M, Bobrov A, Dereviagin AY, Lubinski D, Hubberten H-W (2003) Late Pleistocene vegetation and climate on the northern Taymyr Peninsula, Arctic Russia Boreas 32:484–505 Battarbee RW, Davydova NN, Digerfeldt G, Eronen M, Gaillard M-J, Gliemeroth AK, Hannon G, Harrison SP, Hofmann W, Liew PM, Lotter AF, Loeffler H, Marciniak B, Smol JP, Tarasov PE (1998) Biological records of climate change in lake sediments Palaăoklimaforschung/Palaeoclimate Res 25:161167 BDP Members (2004) High-resolution sedimentary record in a new BDP-99 core from Posol’sk Bank in Lake Baikal Russ Geol Geophys 25(2):163–194 Berger A (1978) Long term variations of daily insolations and Quaternary climatic changes J Atmos Sci 35(12):2362–2367 Bezrukova EV (1999) Paleogeografiya Pribaikal’ya v pozdnelednikov’e i golotsene (in Russian) Nauka, Novosibirsk Bezrukova EV, Letunova PP (2001) A high-resolution record of east Siberian paleoclimates in the Early and Middle Pleistocene by palynological studies of Baikal sediments from the deep borehole BDP-96-1 (in Russian, with English Abstract) Geologiya Geofizika 42:98–107 Bigelow NH, Brubaker LB, Edwards ME, Harrison SP, Prentice CI, Anderson PM, Bartlein PJ, Christensen TR, Cramer W, Kaplan JO, Lozhkin AV, Matveyeva NV, Murray DF, Mc Guire AD, Razzhivin VY, Ritchie JC, Smith B, Walker DA, Gajewski K, Wolf V, Holmquist BH, Igarashi Y, Kremenetskii K, Paus A, Pisaric MFJ, Volkova VS (2003) Climate change and Arctic ecosystems:1 Vegetation changes north of 55°N between the last glacial maximum, mid-Holocene, and present J Geophys Res 108, NO D19, 8170 DOI 10.1029/ 2002JD002558 Channell JET (1999) Geomagnetic paleointensity and directional secular variation at Ocean Drilling Program (ODP) site 984 (Bjorn Drift) since 500 ka: comparison with ODP site 983 (Gardar drift) J Geophys Res 104:22937–22951 Charlet F, Fagel N, De Batist M, Hauregard F, Minnebo B, Meischner D, SONIC Team (2005) Sedimentary dynamics on isolated highs in Lake Baikal: evidence from detailed high-resolution geophysical data and sediment cores Global Planet Change 46(1–4):125–144 Cheddadi R, Lamb HF, Guiot J, van der Kaars S (1998a) Holocene climatic change in Marocco: a quantitative reconstruction from pollen data Clim Dyn 14:883–890 Cheddadi R, Mamakowa K, Guiot J, de Beaulieu J-L, Reille M, Andrieu V, Granoszewski W, Peyron O (1998b) Was the climate of the Eemian stable? A quantitative climate reconstruction from seven European pollen records Palaeogeogr Palaeoclim Palaeoecol 143:73–85 Climatic atlas of Asia (1981) Gidrometeoizdat, Leningrad Demory F, Nowaczyk NR, Bluszcz A, Demske D, Granoszewski W, Witt A, Oberhaănsli H (2005) High-resolution magnetostratigraphy of late Quaternary sediments from Lake Baikal, Siberia: age models and time lag between marine and intracontinetal climatic responses Global Planet Change 46(1– 4):167–186 Dylis NV, Reshchikov LI, Malyshev LI (1965) Rastitel’nost In: Preobrazhenskii VS, Pomus MI, Sochava VB (eds) Predbaikal’e i Zabaikal’e (in Russian) Nauka, Moscow, pp 225–281 636 Tarasov et al.: Quantitative reconstruction of the last interglacial vegetation and climate Frenzel B, Pecsi B, Velichko AA (eds) (1992) Atlas of Palaeoclimates and Palaeoenvironments of the Northern Hemisphere, Late Pleistocene–Holocene Hungarian Academy of Sciences, Budapest, Gustav Fisher Verlag, Stuttgart Frogley MR, Tzedakis PC, Heaton THE (1999) Climate variability in Northwest Greece during the last interglacial Science 285:1886–1889 Galaziy GI (ed) (1993) Baikal Atlas (in Russian) Federal Agency for Geodesy and Cartography of Russia, Moscow Granoszewski W, Demske D, Nita M, Heumann G, Andreev AA (2005) Vegetation and climate variability during the last interglacial evidenced in the pollen record from Lake Baikal Global Planetary Change 46(1–4):187–198 Goldberg EL, Grachev MA, Edgington DN, Navez J, Andre L, Chebykin EP, Shul’pyakov IO (2001) Direct U–Th dating of the last two interglacials in sediments of Lake Baikal DAN 381:805–808 Grachev MA, Vorobyova SS, Khlystov OM, Bezrukova EV, Weinberg EV, Goldberg EL, Granina LZ, Kornakova EG, Lazo FI, Levina OV, Letunova PP, Otinov PV, Pirog VV, Fedotov AP, Yaskevich SA, Bobrov VA, Sukhorukov FV, Rezchikov VI, Fedorin MA, Zolotarev KV, Kravchinsky VA (1997) Signal of the paleoclimates of Upper Pleistocene in the sediments of Lake Baikal Russ Geol Geophys 38:957– 980 Grichuk VP (1969) Opyt rekonstruktsii nekotorykh elementov klimata Severnogo polushariya v atlanticheskii period golotsena In: Neustadt MI (ed) Holocene Na`uka, Moscow, pp 41–57 Grichuk VP (1984) Late Pleistocene vegetation history In: Velichko AA (ed) Late quaternary environments of the Soviet Union University of Minnesota Press, Minneapolis, pp 155–178 Guiot J, Pons A, de Beaulieu J-L, Reille M (1989) A 140,000 year climatic reconstruction from two European pollen records Nature 338:309–313 Guiot J (1990) Methodology of the last climatic cycle reconstruction from pollen data Palaeogeogr Palaeoclimatol Palaeoecol 80:49–69 Gunin PD, Vostokova EA, Dorofeyuk NI, Tarasov PE, Black CC (1999) Vegetation dynamics of Mongolia Geobotany 26 Kluwer, Dordrecht Harrison SP, Kutzbach JE, Prentice IC, Behling PJ, Sykes MT (1995) The response of Northern Hemisphere extratropical climate and vegetation to orbitally induced changes in insolation during the last interglacial Quat Res 43:174–184 Horiuchi K, Minoura K, Hoshino K, Oda T, Nakamura T, Kawai T (2000) Palaeoenvironmental history of Lake Baikal during the last 23000 years Palaeogeogr Palaeoclim Palaeoecol 157:95–108 Hutchinson DR, Golmshtok AJ, Zonenshain LP, Moore TC, Scholz CA, Klitgord KD (1992) Depositional and tectonic framework of the rift basins of Lake Baikal from multichannel seismic data Geology 20:589–592 Imbrie J, Hays J, Martinson DG, McIntyre A, Mix AC, Morley JJ, Pisias NG, Prell WL, Shackleton NJ (1984) The orbital theory of Pleistocene climate: support from a revised chronology of the marine d18O record In: Berger A, Imbrie J, Hays J, Kukla G, Saltzman B (eds) Milankovitch and climate Reidel, Dordrecht Holland, pp 269–305 Jouzel J, Barkov NI, Barnola JM, Bender M, Chapellaz J, Genthon C, Kotlyakov VM, Lipenkov V, Lorius C, Petit JR, Raynaud D, Raisbeck G, Ritz C, Sowers T, Stievenard M, Yiou F, Yiou P (1993) Extending the Vostok ice-core record of palaeoclimate to the penultimate glacial period Nature 364:407–412 Karabanov EB, Prokopenko AA, Williams DF, Khursevich GK (2000) Evidence for mid-Eemian cooling in continental climatic record from Lake Baikal J Paleolomnology 23:365–371 Khursevich GK, Karabanov EB, Prokopenko AA, Williams DF, Kuzmin MI, Fedenya SA, Gvozdkov AA (2001) Insolation regime in Siberia as a major factor controlling diatom production in Lake Baikal during the past 800,000 years Quat Intern 90–91:47–58 Klimanov VA (1984) Paleoclimatic reconstructions based on the information-statistical method In: Velichko AA (ed) Late quaternary environments of the Soviet Union University of Minnesota Press, Minneapolis, pp 297–303 Klotz S, Guiot J, Mosbrugger V (2003) Continental European Eemian and early Wuărmian climate evolution: comparing signals using dierent quantitative reconstruction approaches based on pollen Global Planet Change 36:277–294 Kubatzki C, Montoya M, Rahmstorf S, Ganopolski A, Claussen M (2000) Comparison of a coupled global model of intermediate complexity and an AOGCM for the last interglacial Clim Dyn 14:461–471 Kuăhl N, Gebhardt C, Litt T, Hense A (2002) Probability density functions as botanical-climatological transfer functions for climate reconstruction Quat Res 58:381392 Kuăhl N, Litt T (2003) Quantitative time series reconstruction of Eemian temperature at three European sites using pollen data Veget Hist Archaeobot 12:205–214 Kukla GJ (2000) The last interglacial Science 287:987–988 Lambeck K, Chappell J (2001) Sea level change through the Last Glacial cycle Science 292:679–685 Martinson DG, Pisias NG, Hays J, Imbrie J, Moore TC, Shackleton NJ (1987) Age dating and the orbital theory of the ice ages: development of a high-resolution to 300,000 year chronostratigraphy Quat Res 27:1–29 McManus JF, Bond GC, Broecker WS, Johnsen S, Labeyrie L, Higgins S (1994) High resolution climate records from the North Atlantic during the last interglacial Nature 371:326–329 Molozhnikov VN (1986) Plant communities of Pribaikalie (Rastitel’nye soobshchestva Pribaikal’ya) (in Russian) Novosibirsk, Nauka Nakagawa T, Tarasov P, Kotoba N, Gotanda K, Yasuda Y (2002) Quantitative pollen-based climate reconstruction in Japan: application to surface and late Quaternary spectra Quat Sci Rev 21:2099–2113 New M, Lister D, Hulme M, Makin I (2002) A high-resolution data set of surface climate over global land areas Clim Res 21:1–25 Prentice I, Cramer W, Harrison SP, Leemans R, Monserud RA, Solomon AM (1992) A global biome model based on plant physiology and dominance, soil properties and climate J Biogeogr 19:117–134 Prentice IC, Guiot J, Huntley B, Jolly D, Cheddadi R (1996) Reconstructing biomes from palaecological data: a general method and its application to European pollen data at and ka Clim Dyn 12:185–194 Prokopenko AA, Williams DF (2004) Deglacial methane emission signals in the carbon isotopic record of Lake Baikal Earth Planet Sci Lett 218:135–147 Rindzyunskaya NM, Pakhomov MM (1977) K stratigrafii chetvertichnykh otlozhenii Severo-Baikal’skogo nagor’ya (in Russian) Izvestiya AN SSSR Ser Geol 4:146–149 Rioual P, Andrieu-Ponel V, Rietti-Shati M, Battarbee RW, de Beaulieu J-L, Cheddadi R, Reille M, Svobodova H, Shemesh A (2001) High-resolution record of climate stability in France during the last interglacial period Nature 413:293–296 Rioual P, Mackay A (2005) A diatom record of centennial resolution for the Kazantsevo Interglacial stage in Lake Baikal (Siberia) Global Planet Change 46(1–4):199–219 Solovieva N, Tarasov PE, MacDonald G (2005) Quantitative reconstruction of Holocene climate from the Chuna Lake pollen record, Kola Peninsula, northwest Russia Holocene 15(1):141–148 Tarasov PE, Webb III T, Andreev AA, Afanaseva NB, Berezina NA, Bezusko LG, Blyakharchuk TA, Bolikhovskaya NS, Cheddadi R, Chernavskaya MM, Chernova GM, Dorofeyuk NI, Dirksen VG, Elina GA, Filimonova LV, Glebov FZ, Guiot J, Gunova GS, Harrison SP, Jolly D, Khomutova VI, Kvavadze EV, Osipova IM, Panova NK, Prentice IC, Saarse L, Sevastyanov DV, Volkova VS, Zernitskaja VP (1998a) Presentday and mid-Holocene biomes reconstructed from pollen and plant macrofossil data from former Soviet Union and Mongolia J Biogeogr 25:1029–1053 Tarasov et al.: Quantitative reconstruction of the last interglacial vegetation and climate Tarasov PE, Cheddadi R, Guiot J, Bottema S, Peyron O, Belmonte J, Ruiz-Sanchez V, Saadi FA, Brewer S (1998b) A method to determine warm and cool steppe biomes from pollen data; application to the Mediterranean and Kazakhstan regions J Quat Sci 13:335–344 Tarasov PE, Peyron O, Guiot J, Brewer S, Volkova VS, Bezusko LG, Dorofeyuk NI, Kvavadze EV, Osipova IM, Panova NK (1999) Last Glacial maximum climate of the Former Soviet Union and Mongolia reconstructed from pollen and plant macrofossil data Clim Dyn 15:227–240 Tarasov PE, Volkova VS, Webb III T, Guiot J, Andreev AA, Bezusko LG, Bezusko TV, Bykova GV, Dorofeyuk NI, Kvavadze EV, Osipova IM, Panova NK, Sevastyanov DV (2000) Last Glacial maximum biomes reconstructed from pollen and plant macrofossil data from Northern Eurasia J Biogeogr 27:609–620 Tarasov PE, Dorofeyuk NI, Vipper PB (2002) The Holocene dynamics of vegetation in Buryatia Stratigr Geol Corr 10:88–96 Texier D, de Noblet N, Harrison SP, Haxeltine A, Jolly D, Joussaume S, Laarif F, Prentice IC, Tarasov P (1997) Quantifying the role of biosphere-atmosphere feedbacks in climate change: coupled model simulations for 6000 years BP and comparison with palaeodata for northern Eurasia and northern Africa Clim Dyn 13:865–882 Tzedakis PC, Frogley MR, Heaton THE (2002) Duration of last interglacial conditions in northwestern Greece Quat Res 58:53–55 637 Turner C (2000) The Eemian Interglacial in the North European plain and adjacent areas Geologie Mijnbouw 79:217–231 Velichko AA (ed) (1984) Late Quaternary environments of the Soviet Union University of Minnesota Press, Minneapolis Velichko AA, Borisova OK, Zelikson EM (2002) Paradoksy klimata poslednego mezhlednikov’ya In: Spasskaya II (ed) Routes of evolutionary geography (summary and prospects), Institute of Geography (in Russian) Russian Academy of Sciences, Moscow, pp 207–239 Wohlfarth B, Schwark L, Bennike O, Filimonova L, Tarasov P, Bjoărkman L, Brunnberg L, Demidov I, Possnert G (2004) Unstable Early Holocene climatic and environmental conditions in northwestern Russia derived from a multidisciplinary study of a lake sediment sequence from Pichozero, southwestern Russian Karelia Holocene 14(5):732–746 Yuan D, Cheng H, Edvards RL, Dykovski CA, Kelly MJ, Zhang M, Qing J, Lin Y, Wang Y, Wu J, Dorale JA, An Z, Cai Y (2004) Timing, duration, and transitions of the Last Interglacial Asian Monsoon Science 304:575–578 Zagwijn WH (1996) An analysis of Eemian climate in western and central Europe Quat Sci Rev 15:451–469 Zhukov VM (1965) Klimat In: Preobrazhenskii VS, Pomus MI, Sochava VB (eds) Predbaikal’e i Zabaikal’e (in Russian) Nauka, Moscow, pp 91–126 ... locally in the moist habitats The biome reconstruction results confirm the position of the lower and upper boundaries of the LI in the region The lower boundary of the LI is placed at the level... of the area around the coring site inferred from seismic data and the undisturbed finegrained character of the sediment indicate continuous Tarasov et al.: Quantitative reconstruction of the last. .. Quantitative climate reconstruction The results of the quantitative reconstruction of temperature and precipitation through the Continent record are shown in Fig The earlier part of the record (ca