Search

Graphical abstract, Alpine glaciers, with their valuable combination of highly sensitive response to climate and near-global extent, are powerful tools for investigating previous and present climate changes. They also represent critical water resources for areas around the globe, with the potential for far-reaching effects in a warming world. Advancements to understand and model glacial changes and the variables influencing them are therefore paramount. Many glacier models fall into one of two endmembers: either highly complex transient models requiring careful tuning of multiple parameters to individual glaciers, or basic empirical correlations of glacier area and length with few considerations for local and regional variations in characteristics. Here we detail a physical steady-state model for alpine glaciers relating directly to glacier mass balance (via the equilibrium line altitude) while retaining the simplicity of other morphology methods, and simultaneously including error estimates. We provide custom MATLAB functions as a user-friendly and generally-applicable method to estimate glacier equilibrium line altitudes from only a limited number of glacier bed topography and glacier width measurements. As a test of the model’s efficacy, we compare the model results for present-day glaciers in the Swiss Alps with previously published estimates of equilibrium line altitudes and intermediate model outputs.•The method estimates glacier equilibrium line altitudes from a limited set of bed topography measurements and constraints on glacier width.•The method is based on continuity equations, reducing the need for empirical coefficients tuned with measured data.•The method uses Monte Carlo sampling and bootstrapping to generate uncertainty bounds on the equilibrium line altitude estimates.

Understanding local patterns and large-scale processes in past climate necessitates a detailed network of temperature reconstructions. In this study, a merged temperature inference model using fossil chironomid (Diptera: Chironomidae) datasets from Finland and Poland was constructed to fill the lack of an applicable training set for East European sites. The developed weighted averaging partial least squares (WA-PLS) inference model showed favorable performance statistics, suggesting that the model can be useful for downcore reconstructions. The combined calibration model includes 212 sites, 142 taxa, and a temperature gradient of 11.3-20.1 degrees C. The 2-component WA-PLS model has a cross-validated coefficient of determination of 0.88 and a root mean squared prediction error of 0.88 degrees C. We tested the new East European temperature transfer function in chironomid stratigraphies from a Finnish high-resolution short-core sediment record and a Polish paleolake (Zabieniec) covering the past similar to 20 000 yr. In the Finnish site, the chironomid-inferred temperatures correlated closely with the observed instrumental temperatures, showing improved accuracy compared to estimates by the original Finnish calibration model. In addition, the long-core reconstruction from the Polish site showed logical results in its general trends compared to existing knowledge on the past regional climate trends; however, it had distinct differences when compared with hemispheric climate oscillations. Hence, based on these findings, the new temperature model will enable more detailed examination of long-term temperature variability in Eastern Europe, and consequently, reliable identification of local and regional climate variability of the past.

Quantifying the palaeoclimates of Qinghai–Tibetan Plateau is vital for understanding the uplift history of plateau and the evolution of Asian monsoon since Cenozoic. Recently, the Coexistence Approach (CA) has been employed to reconstruct the palaeoelevation and palaeoclimate of the plateau by several studies. However, the application of CA in mountainous areas and the realism of climate reconstructions via this method are seldom discussed, although the complexity of reconstructions is speculated. Here we reevaluated the realism of climate reconstruction using the CA with modern pollen samples from the Qinghai–Tibetan Plateau, and try to explore the possible factors influencing the precipitation and temperature reconstructions by CA. We suggest that the long-distance transport pollen as a result of the Asian summer monsoon potentially significantly affects the reconstructions both for precipitation and temperature. The precipitation complexly interacting with snowmelt and permafrost thaw leads to the discrepancy between the reconstructed precipitation and the real value. The response temperature for blossoming of dwarfed plants on the plateau is mostly likely higher than the air temperature (usually measured at 1.5 m above ground) due to energy flux or morphological adaptation of inflorescences during the growing season, causing the distortion of temperature reconstructions. Precipitation reconstruction is notoriously difficult as the establishers of CA have already suggested, but reconstructing the low temperatures may be even more challenging on Qinghai–Tibetan Plateau. Though all of the explorations in current paper are in a qualitative way, it offers an inspiration of how appropriately interpret the disagreements between CA results and the observations, and of how to obtain a reasonable reconstruction of palaeoclimate of the plateau.

Quantitative reconstructions of past climates are an important resource for evaluating how well climate models reproduce climate changes. One widely used statistical approach for making such reconstructions from fossil biotic assemblages is weighted averaging partial least-squares regression (WA-PLS). There is however a known tendency for WA-PLS to yield reconstructions compressed towards the centre of the climate range used for calibration, potentially biasing the reconstructed past climates. We present an improvement of WA-PLS by assuming that: (i) the theoretical abundance of each taxon is unimodal with respect to the climate variable considered; (ii) observed taxon abundances follow a multinomial distribution in which the total abundance of a sample is climatically uninformative; and (iii) the estimate of the climate value at a given site and time makes the observation most probable, i.e. it maximizes the log-likelihood function. This climate estimate is approximated by weighting taxon abundances in WA-PLS by the inverse square of their climate tolerances. We further improve the approach by considering the frequency ( fx ) of the climate variable in the training dataset. Tolerance-weighted WA-PLS with fx correction greatly reduces the compression bias, compared with WA-PLS, and improves model performance in reconstructions based on an extensive modern pollen dataset.

International audience; This paper presents the first continuous pollen record from the southern Namib Desert spanning the last 50,000 years. Obtained from rock hyrax middens found near the town of Pella, South Africa, these data are used to reconstruct vegetation change and quantitative estimates of temperature and aridity. Results indicate that the last glacial period was characterised by increased water availability at the site relative to the Holocene. Changes in temperature and potential evapotranspiration appear to have played a significant role in determining the hydrologic balance. The record can be considered in two sections: 1) the last glacial period, when low temperatures favoured the development of more mesic Nama-Karoo vegetation at the site, with periods of increased humidity concurrent with increased coastal upwelling, both responding to lower global/regional temperatures; and 2) the Holocene, during which time high temperatures and potential evapotranspiration resulted in increased aridity and an expansion of the Desert Biome. During this latter period, increases in upwelling intensity created drier conditions at the site. Considered in the context of discussions of forcing mechanisms of regional climate change and environmental dynamics, the results from Pella stand in clear contrast with many inferences of terrestrial environmental change derived from regional marine records. Observations of a strong precessional signal and interpretations of increased humidity during phases of high local summer insolation in the marine records are not consistent with the data from Pella. Similarly, while high percentages of Restionaceae pollen has been observed in marine sediments during the last glacial period, they do not exceed 1% of the assemblage from Pella, indicating that no significant expansion of the Fynbos Biome has occurred during the last 50,000 years. These findings pose interesting questions regarding the nature of environmental change in southwestern Africa, and the significance of the diverse records that have been obtained from the region.

A global inventory of early instrumental meteorological measurements is compiled. It comprises thousands of series, many of which have not been digitized, pointing to the potential of weather data rescue. Instrumental meteorological measurements from periods prior to the start of national weather services are designated “early instrumental data”. They have played an important role in climate research as they allow daily-to-decadal variability and changes of temperature, pressure, and precipitation, including extremes, to be addressed. Early instrumental data can also help place 21st century climatic changes into a historical context such as to define pre-industrial climate and its variability. Until recently, the focus was on long, high-quality series, while the large number of shorter series (which together also cover long periods) received little to no attention. The shift in climate and climate impact research from mean climate characteristics towards weather variability and extremes, as well as the success of historical reanalyses which make use of short series, generates a need for locating and exploring further early instrumental measurements. However, information on early instrumental series has never been electronically compiled on a global scale. Here we attempt a worldwide compilation of metadata on early instrumental meteorological records prior to 1850 (1890 for Africa and the Arctic). Our global inventory comprises information on several thousand records, about half of which have not yet been digitized (not even as monthly means), and only approximately 20% of which have made it to global repositories. The inventory will help to prioritize data rescue efforts and can be used to analyze the potential feasibility of historical weather data products. The inventory will be maintained as a living document and is a first, critical, step towards the systematic rescue and re-evaluation of these highly valuable early records. Additions to the inventory are welcomed.

The Southern Annular Mode (SAM) is the leading mode of atmospheric interannual variability in the Southern Hemisphere (SH) extra-tropics. Here, we assess the stationarity of SAM spatial correlations with instrumental and paleoclimate proxy data for the past millennium. The instrumental period shows that temporal non-stationarities in SAM teleconnections are not consistent across the SH land areas. This suggests that the influence of the SAM index is modulated by regional effects. However, within key-regions with good proxy data coverage (South America, Tasmania, New Zealand), teleconnections are mostly stationary over the instrumental period. Using different stationarity criteria for proxy record selection, we provide new austral summer and annual mean SAM index reconstructions over the last millennium. Our summer SAM reconstructions are very robust to changes in proxy record selection and the selection of the calibration period, particularly on the multi-decadal timescale. In contrast, the weaker performance and lower agreement in the annual mean SAM reconstructions point towards changing teleconnection patterns that may be particularly important outside the summer months. Our results clearly portend that the temporal stationarity of the proxy-climate relationships should be taken into account in the design of comprehensive regional and hemispherical climate reconstructions. The summer SAM reconstructions show no significant relationship to solar, greenhouse gas and volcanic forcing, with the exception of an extremely strong negative anomaly following the AD 1257 Samalas eruption. Furthermore, reconstructed pre-industrial summer SAM trends are very similar to trends obtained by model control simulations. We find that recent trends in the summer SAM lie outside the 5–95% range of pre-industrial natural variability. Fil: Dätwyler, Christoph. University of Bern. Centre for Climate Change Research. Institute of Geography and Oeschger ; Suiza Fil: Neukom, Raphael. University of Bern. Centre for Climate Change Research. Institute of Geography and Oeschger ; Suiza Fil: Abram, Nerilie J.. Australian National University; Australia Fil: Gallant, Ailie J. E.. Monash University; Australia Fil: Grosjean, Martin. University of Bern. Centre for Climate Change Research. Institute of Geography and Oeschger ; Suiza Fil: Jacques-Coper, Martín. Universidad de Concepción; Chile Fil: Karoly, David J.. University of Melbourne; Australia Fil: Villalba, Ricardo. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Mendoza. Instituto Argentino de Nivología, Glaciología y Ciencias Ambientales. Provincia de Mendoza. Instituto Argentino de Nivología, Glaciología y Ciencias Ambientales. Universidad Nacional de Cuyo. Instituto Argentino de Nivología, Glaciología y Ciencias Ambientales; Argentina

A renewed interest in historical droughts is due to the current climate change issues and also to the recent extreme drought of 2017 in Portugal. This work aims to lengthen the historical drought series and help improve projections for the future. The main droughts of the 18th century in mainland Portugal were identified using mostly direct documentary sources: individual (poems, letters, memoirs, manuscripts and printed newspapers) and institutional (ecclesiastical and administrative), as well as accounts of Pro pluvia ceremonies and processions. All the records had been assembled in the KlimHist database (http://clima.ul.pt/Klimhist-project).A summary statistical analysis of the drought evidence including its impacts was performed to assess its temporal distribution and geographical incidence. A strong precipitation variability was detected, which, similar to the present times, hardly presents trends in long series. The main years of drought during the 18th century in Portugal are highlighted, particularly the severe and long-lasting 1737–38 and 1753–54 droughts that, together with their atmospheric causes, were studied in detail. There are more drought records for Southern Portugal (SP) than for Northern Portugal (NP). It was also observed that spring March, April and May (MAM) and winter December, January and February (DJF) drought evidence are more frequent in SP, whereas in NP there are more records concerning summer June, July and August (JJA). The results obtained agree with those of others of the Iberian Peninsula (IP). Despite Portugal being a small country, our results confirm the existence of an important climatic boundary between the NP and the SP with different precipitation variability.

Studies report the good potential of Pinus heldreichii (PIHE) and Pinus peuce (PIPE) for developing long chronologies from living trees and warn that the climate signal is weak in tree-ring widths of PIHE, and particularly PIPE. The goals of the study were to develop long chronologies, and to analyze the climate&ndash, growth relationship and potential for long climate reconstructions using tree-ring widths (TRW) and stable carbon isotopes ratios (&delta, 13C) in tree rings at the northern edge of species distribution in the eastern part of Montenegro. The PIHE TRW chronology covers the period 1571&ndash, 2013 (443 years) and the PIPE TRW chronology 1521&ndash, 2013 (493 years). The temperature signal in PIHE TRW is weak and the precipitation signal is non-existent. PIPE has no climate signal in TRW. Both studied species have very similar &delta, 13C chronologies, which allows us to merge isotope chronologies into a single composite &delta, 13C Pinus chronology. The composite chronology has a strong signal related to average monthly temperature in June, July, and August and monthly values for cloudiness in July and August, with r &gt, 0.6 and r &lt, &minus, 0.6 for individual months, respectively. The climate signal was enhanced when June, July, and August values were merged into seasonal variables. The temporal stability of temperature, precipitation and cloudiness signals is consistent. The spatial extent of the &delta, 13C chronology extends over a very large region, including all surrounding countries.

The development of research into the history of tree growth and inferred summer temperature changes in Yamalia spanning the last 2000 years is reviewed. One focus is the evolving production of tree-ring width (TRW) and tree-ring maximum-latewood density (MXD) larch (Larix sibirica) chronologies, incorporating different applications of Regional Curve Standardisation (RCS). Another focus is the comparison of independent data representing past tree growth in adjacent Yamalia areas: Yamal and Polar Urals, and the examination of the evidence for common growth behaviour at different timescales. The sample data we use are far more numerous and cover a longer time-span at Yamal compared to the Polar Urals, but Yamal has only TRW, while there are both TRW and MXD for the Polar Urals. We use more data (sub-fossil and from living trees) than in previous dendroclimatic studies in this region. We develop a new TRW chronology for Yamal, more than 2000 years long and running up to 2005. For the Polar Urals we develop new TRW and MXD chronologies that show good agreement at short (

A first assessment of the main climatic drivers that modulate the tree-ring width (RW) and maximum latewood density (MXD) along the Italian Peninsula and northeastern Sicily was performed using 27 forest sites, which include conifers (RW and MXD) and broadleaves (only RW). Tree-ring data were compared using the correlation analysis of the monthly and seasonal variables of temperature, precipitation and standardized precipitation index (SPI, used to characterize meteorological droughts) against each species-specific site chronology and against the highly sensitive to climate (HSTC) chronologies (based on selected indexed individual series). We find that climate signals in conifer MXD are stronger and more stable over time than those in conifer and broadleaf RW. In particular, conifer MXD variability is directly influenced by the late summer (August, September) temperature and is inversely influenced by the summer precipitation and droughts (SPI at a timescale of 3 months). The MXD sensitivity to August–September (AS) temperature and to summer drought is mainly driven by the latitudinal gradient of summer precipitation amounts, with sites in the northern Apennines showing stronger climate signals than sites in the south. Conifer RW is influenced by the temperature and drought of the previous summer, whereas broadleaf RW is more influenced by summer precipitation and drought of the current growing season. The reconstruction of the late summer temperatures for the Italian Peninsula for the past 300 years, based on the HSTC chronology of conifer MXD, shows a stable model performance that underlines periods of climatic cooling (and likely also wetter conditions) in 1699, 1740, 1814, 1914 and 1938, and follows well the variability of the instrumental record and of other tree-ring-based reconstructions in the region. Considering a 20-year low-pass-filtered series, the reconstructed temperature record consistently deviates

Air Temperature and Precipitation Changes in the Kaffiøyra Region (NW Spitsbergen) from 1975 to 2010 Air temperature and precipitation conditions in the Kaffiøyra region in the summer season (21st July-31st August) for the period of 1975-2010 are described: 1) on the basis of data gathered in 18 expeditions during which meteorological measurements were done, and 2) on the basis of complete series of data combining both original and reconstructed data. The latter ones were obtained using data from Ny Ålesund meteorological station, which are strongly correlated with the data from Kaffiøyra. Seasonal statistics presented for air temperature and precipitation based on these two sets of data reveal only slight changes. Temperature parameters (daily mean, maximum and minimum) for summer in Kaffiøyra in the study period (1975-2010) show upward trends, which are, however, statistically significant only for the daily mean. On the other hand, precipitation totals in the study period reveal a downward trend, but not statistically significant. Such thermal-precipitation behaviour in the study part of Spitsbergen in general terms is similar to those in other parts of Spitsbergen.

Important progresses have been made in palaeoclimatological studies by using statistical methods. But they are in somewhere limited as they take the present as an absolute reference. This is particularly true for the modern analogue technique. The availability of mechanistic models to simulate the proxies measured in the sediment cores gives now the possibility to relax this constraint. In particular, vegetation models provide outputs comparable to pollen data (assuming that there is a relationship between plant productivity and pollen counts). The input of such models is, among others, climate. The idea behind paleoclimatological reconstructions is then to obtain inputs, given outputs. This procedure, called model inversion, can be achieved with appropriate algorithms in the frame of the Bayesian statistical theory. But we have chosen to present it in an intuitive way, avoiding the mathematics behind it. Starting from a relative simple application, based on an equilibrium BIOME3 model with a single proxy (pollen), the approach has evolved into two directions: (1) by using several proxies measured on the same core (e.g. lake-level status and δ13C) when they are related to a component of the vegetation, and (2) by using a more complex vegetation model, the dynamic vegetation model LPJ-GUESS. Examples presented (most of them being already published) concern Last Glacial Maximum in Europe and Africa, Holocene in a site of the Swiss Jura, an Eemian site in France. The main results are that: (1) pollen alone is not able to provide exhaustive information on precipitation, (2) assuming past CO2 equivalent to modern one may induce biases in climate reconstruction, (3) vegetation models seem to be too much constrained by temperature relative to precipitation in temperate regions. This paper attempts to organise some recent ideas in the palaeoclimatological reconstruction domain and to propose prospectives in that effervescent domain.

International audience; Late Quaternary climate variability in the southern African subtropics is still only poorly resolved, with significant complexity and apparent contradictions in the regional dataset. To more effectively interpret and synthesize key regional records, we reanalysed the data from 13 pollen sequences from the summer rainfall zone of South Africa spanning the last 45,000 years, obtaining directly comparable quantitative reconstructions of mean annual temperature and summer rainfall. Temperature reconstructions from across the region provide consistent results, with all sites reflecting trends observed in southwest Indian Ocean sea-surface temperatures in the adjacent Mozambique Channel. Precipitation reconstructions are more heterogeneous, with two distinct subregions being identified. In the northeast, long-term trends in precipitation are determined by sea-surface and continental temperature trends, revealing a positive relationship between temperature and rainfall. This long-term pattern appears to be primarily driven by high northern latitude mechanisms, with direct local insolation being subordinate. Their relative impact reversed during terminal glacial period/early Holocene, at which time direct insolation forcing became the main driver of rainfall variability. Further south, in central South Africa, precipitation variability appears also to be influenced by the latitudinal position of the Southern Hemisphere westerlies, which combine with tropical flow to create tropical-temperate trough, advecting moisture into the interior. In this region, periods of maximum precipitation coincide with periods of elevated SSTs and equatorward expansions of the westerly storm track. This study allows for a fully constrained understanding of climate dynamics along the eastern African margin for the last 45,000 years, linking dynamics to drivers and describing how the climate systems evolved across the last glacialeinterglacial transition.

International audience; We present a method of multiproxy reconstruction of the climate of Europe during the last millennium. The proxies used comprise long tree-ring width series, grape harvest dates, Greenland ice oxygen isotope series and temperature indices based on historical documents. The proxies are calibrated using gridded April to September mean temperatures for western Europe, i.e., between 10°W and 20°E and between 35°N and 55°N. They are calibrated also using the long instrumental summer temperature series of the Marseilles observatory of Longchamp, which begins in the mid-eighteenth century. The method is a combination of an analogue technique, which is able to deal with missing data, an artificial neural network technique for an optimal non-linear calibration and a bootstrap technique for calculating error bars on the reconstruction. About 70% of the temperature variance is reconstructed. The amplitude of the past temperature variations is particularly well reconstructed, which is important when considering whether the recent temperature trend is or is not within the natural variability. It appears that the temperature of the last decade of the twentieth century was reached only 14 times during the last millennium. The reconstruction is discussed with respect to other multiproxy and borehole temperature reconstructions. We conclude that a reconstruction such ours, with a specific regional focus (as opposed to the larger Northern Hemisphere) is more reliable and is in better agreement with borehole results, even allowing for the fact that only a part of the long-term variance is reconstructed. 'Little Ice Age' (c. AD 1560 - 1930) summers were 0.2±0.5°C cooler than the 1961 - 1990 period. Borehole temperatures indicate a cooling of 0.4°C which falls in the 95% confidence interval of our reconstructions.

The paper by Martin-Chivelet et al. (2011) presents a temperature reconstruction for the past 4000 yr based on the δ 13 C record of three speleothems from different caves in northern Spain. I consider that the record of these speleothems cannot provide a reliable temperature reconstruction due to the lack of: (1) an adequate calibration (2) a mechanism able to explain the observed δ 13 C signal variability in terms of surface air temperature (SAT) and (3) a clear correlation when the record is quantitatively compared with other thermal reconstructions. Instead, the δ 13 C signal in these speleothems responds to a series of factors and not a single control. Some of these controls are known to be affected by factors other than SAT and could dominate the δ 13 C signal explaining the lack of correlation with other palaeotemperature compilation records. Therefore, a palaeotemperature reconstruction from these samples is thought to be unreliable.

This report presents a study of the variability in a recently published April to September gridded reconstruction temperatures in Europe and an analysis of their extremes. This reconstruction was based on tree rings, historical documents, pollen assemblages and ice cores. The methodology used, an original spectral analog method, preserves long-term variations and the variability of temperature series, which guarantees the pertinence of the analysis of the extremes over a period as long as the last millennium. The analysis of the extremes was performed with the peak-over-threshold (POT) method over two warm periods (A.D. 1000-1350 and 1880-2007) and a cold period (A.D. 1350-1950).|We found that (1) according to the long-term variations shown in this reconstruction, the growing season temperature during the last decade has exceeded all of those observed during the Medieval period; (2) the return period of the maximum event of the Medieval period has been reduced by at least a factor of 5; and (3) all decades before AD. 1350 were warm on average but relatively heterogeneous, while the last decade was homogeneously warmer. A new result of this study concerning Europe is that this anthropogenic change is characterized by spatial homogeneity, with similar changes in both average temperatures and in the distribution of extreme events, while natural climate forcings induce warm periods with heterogeneous spatial patterns and less frequent extreme events. (C) 2011 Elsevier B.V. All rights reserved.

Dating moraines by lichenometry enabled us to reconstruct glacier recession in the Bolivian Andes since the Little Ice Age maximum. On the 15 proglacial margins studied, we identified a system of ten principal moraines that marks the successive positions of glaciers over the last four centuries. Moraines were dated by performing statistical analysis of lichen measurements based on the extreme values theory. Like glaciers in many mid-latitude mountain areas, Bolivian glaciers reached their maximal extent during the second half of the 17th century. This glacier maximum coincides with the Maunder minimum of solar irradiance. By reconstructing the equilibrium-line altitude and changes in mass-balance, we think the glacier maximum may be due to a 20 to 30% increase in precipitation and a 1.1 to 1.2 °C decrease in temperature compared with present conditions. In the early 18th century, glaciers started to retreat at varying rates until the late 19th to early 20th century; this trend was generally associated with decreasing accumulation rates. By contrast, glacier recession in the 20th century was mainly the consequence of an increase in temperature and humidity. These results are consistent with observations made in the study region based on other proxies.

Classical climate reconstruction based on statistical method implicitly assumes that the atmospheric C02 concentration, which has greatly fluctuated during the Quaternary, has no influence on the vegetation. In order to attenuate this assumption, we propose to use a process-based vegetation model (BIOME 3.5) in inverse mode to reconstruct from pollen data the most probable climate under lowered C02 concentration in the biosphere. Appropriate tools to match the model outputs with the pollen data are developed to generate a probability distribution associated with the reconstruction (Monte Carlo sampling and neural network techniques). The method is validated with modern pollen data from Europe and Northern Eurasia : it is able to reconstruct modern climate with a sometimes large error bar from pollen data. The error bar depends in fact on the tolerance of the vegetation to the corresponding climatic variable. The application to 7 1 pollen assemblages from these regions, representing the last glacial maximum (LGM : 18,000 14C-yr B.P.), is done through two experiments : modem and LGM C02 concentration. We show that there is a good agreement between previous results based on statistical methods and inverse modelling in the northern part of the studied region. In the Mediterranean region and southern Siberia, the summer temperatures and the water availability parameters could be higher than previously reconstructed, because of the C02 effect., La reconstruction climatique classique, basée sur la méthode statistique, suppose implicitement que la concentration atmosphérique en C02, qui a considérablement fluctué pendant le Quaternaire, n'a aucune influence sur la végétation. Afin de tempérer ce postulat, nous proposons d'utiliser un modèle fonctionnel de végétation (BIOME 3.5) en mode inverse pour reconstruire, à partir des données polliniques, le climat le plus probable sous faible concentration de C02 dans la biosphère. Des outils adaptés permettant de faire coïncider les sorties des modèles avec les données polliniques sont développés afin d’établir une distribution de probabilité associée à la reconstruction (échantillonnage de Monte Carlo et réseaux neuronaux). La méthode est validée avec les données polliniques actuelles d'Europe et d'Eurasie du nord : il est possible de reconstruire le climat actuel, avec parfois une marge d'erreur importante, à partir des données polliniques. L'erreur dépend en fait de la tolérance de la végétation à la variable climatique correspondante. L'application à 71 assemblages polliniques de ces mêmes régions représentant le dernier maximum glaciaire (DMG : 18,000 ans 14C B.P.) est réalisée au travers de deux expériences : concentration en C02 actuel et glaciaire. Nous montrons qu'il y a une bonne concordance entre les résultats précédents basés sur les méthodes statistiques et la modélisation inverse, dans la partie nord de la région étudiée. En région méditerranéenne et en Sibérie méridionale, les températures estivales et les paramètres de disponibilité de l'eau pourraient être plus élevés que ceux reconstruits précédemment en raison de l'effet du C02., Guiot Joël, Torre Franck, Cheddadi Rachid, Peyron Odile, Tarasov Pavel E., Jolly Dominique, Kaplan Jed O. The climate of the Mediterranean Basin and of Eurasia of the last glacial maximum as reconstructed by inverse vegetation modelling and pollen data. In: Ecologia mediterranea, tome 25 n°2, 1999. pp. 193-204.