Our Common Future Under Climate Change

International Scientific Conference 7-10 JULY 2015 Paris, France

Menu
  • Home
  • Zoom Interactive Programme
Cliquer pour fermer

Tuesday 7 July - 16:30-18:00 UNESCO Fontenoy - ROOM II

1102 - From past to future Climate Changes

Parallel Session

Chair(s): G. Ramstein (LSCE, Gif-sur-Yvette, France), V. Masson-Delmotte (LSCE, Gif-sur-Yvette, France)

Convener(s): Y. Godderis (CNRS , Toulouse, France), H. Linderholm (University of Göteborg, Göteborg, Sweden), T. Yao (Chinese Academy of Sciences, Beijing, China)

Past climate event suggests severe and long-lasting consequences of fossil fuel burning

R.E. Zeebe (University of Hawaii at Manoa, Hawaii, United States of America)

Abstract details
Past climate event suggests severe and long-lasting consequences of fossil fuel burning

RE. Zeebe (1)
(1) University of Hawaii at Manoa, Department of Oceanography, Hawaii, United States of America

Abstract content

Carbon release rates from anthropogenic sources have reached a record high of ~10 Pg C/y in 2014. Due to uncertainties in climate system feedbacks, the impact of the rapid carbon release on the Earth system is difficult to predict. Hence geologic analogues from past transient climate changes are urgently sought after to guide future climate assessments. Throughout the past 66 Myr, the currently known climate aberration with the highest carbon release rate is the Paleocene-Eocene Thermal Maximum (PETM) - an event characterized by future-relevant total carbon release/peak warming and a surprisingly long duration of at least 50,000 years. Based on time-series analysis of stable carbon/oxygen isotope records and carbon cycle/climate modeling, we determine the initial carbon release during the PETM onset. This constrains the maximum sustained PETM carbon release rate to less than ~1 Pg C/y. Given currently available records, it follows that the present anthropogenic carbon release rate is unprecedented during the past 66 Myr by at least an order of magnitude. Future ecosystem disruptions will hence likely exceed the relatively limited extinctions observed at the PETM. Moreover, unforeseeable future responses of the climate system are possible as the Earth system has effectively entered an era of no-analogue state.

Understanding a warming world by studying the Pliocene

A. Haywood (University of Leeds, Leeds, United Kingdom)

Abstract details
Understanding a warming world by studying the Pliocene

A. Haywood (1)
(1) University of Leeds, Uk national centre for atmospheric science, Leeds, United Kingdom

Abstract content

The investigation of warm intervals of the Pliocene epoch has intensified dramatically over the last 10 years. The reasons for this are varied but undoubtedly the availability of progressively higher temporal and spatial resolution environmental records is a key driver. These reconstructions are providing new insights into environmental change during the Pliocene, which enable us to investigate local and regional climate response to an atmospheric CO2 concentration akin to the modern. Furthermore, the use of Climate and Earth System Models in a Pliocene context has been encouraged as a means to test the predictive ability of models, and to understand climate processes generating regional patterns of environmental change.

This presentation will summarise the current state of knowledge of the Pliocene Earth System derived from proxy data and model outputs. In particular it will focus on key aspects of the climate system such as the reconstruction of atmospheric CO2, ice sheet and sea-level change, surface temperature change and polar amplification, the hydrological cycle, ocean circulation and the monsoons.

Challenges in environmental reconstruction and modelling will be highlighted with a summary of emerging initiatives, which are designed to enhance our ability to use warm intervals of the Pliocene as a mechanism to understand the dynamics and drivers of a warming world.

Climate change and variability : lessons from the past

P. Braconnot (Laboratoire des Sciences du Climat et de l'Environnement, Saclay, France)

Abstract details
Climate change and variability : lessons from the past

P. Braconnot (1)
(1) LSCE-IPSL, Gif-sur-Yvette, France

Abstract content

Past climate states provide a unique opportunity to evaluate model performance outside the range ofrecent observed climate variability. They provide test cases of our understanding of climate feedbacksand thresholds that are not observed in modern variability and that could lead to major changes in thefuture. Because of this, past climate simulations of the Last Glacial Maximum (21 000 years BP), themid‐Holocene (6000 years BP) and the last 1000 years have been considered as part of the lastmulti‐model CMIP5/PMIP3 experiments (Taylor et al. 2012, Braconnot et al. 2012), so as to put intoperspective future climate changes and provide complementary model evaluation. Using the resultsof these simulations and of mode‐data comparisons, this presentation will provide an overview ofrecent analyses of climate sensitivity and feedbacks, hydrological cycle in the tropical regions, andinterannual climate variability. It will highlight the new possibilities offered by the modeling of thebiochemical cycles and tracers or by the high resolution records that provide information oninterannual to multi‐decadal variability. Finally, it will discuss the constraints these analyses bring onthe credibility of climate model as well as the new questions that will be addressed in the next phaseof the Paleoclimate Modeling Intercomparison Project (PMIP).

Linking past, present and future climate change to adaptation in the African Sahel

A. Giannini (IRI, Columbia University, Palisades, NY, United States of America)

Abstract details
Linking past, present and future climate change to adaptation in the African Sahel

A. Giannini (1)
(1) Columbia University, International Research Institute for Climate and Society, Palisades, NY, United States of America

Abstract content

The semi-arid African Sahel has received unique attention in the climate science community and beyond since inception of persistent drought at the end of the 1960s. This attention is reflected in the efforts of Working Groups 1 and 2, summarized in the Assessments Reports of the Intergovernmental Panel on Climate Change going back to the first, published in 1990. Initially, drought in the Sahel was attributed to rapid population growth leading to mismanagement of land resources. The hypothesis of a positive bio-geophysical feedback tied human-induced baring of the soils to a reduction of precipitation, which further exacerbated the loss in vegetation cover [Charney 1975, in Q J Roy Meteor Soc]. In the more recent ~10 years the climate of the Sahel has again emerged as the focus of active research, this time as a possible "canary in the coal mine" for anthropogenic climate change. Advances in climate science have first conclusively tied persistent drought to subtle shifts in the surface temperature of the global oceans [Giannini et al. 2003], "freeing farmers of blame" in the drought, then partially attributed these shifts to the influence of greenhouse gases and aerosols [Held et al. 2005, in Proc Nat Acad Sci; Booth et al. 2012, in Nature]. However, in the meantime the region has partially recovered from drought, and is experiencing an increased frequency of flooding [Tall 2010, in Proc Env Sci], underlined, to the extent that it has been documented, by a subtle increase in the intensity of precipitation [Lodoun et al. 2013, in Env Develop; Alhassane et al. 2013, in Secheresse].

Here I present a novel interpretation for the role of the oceans in effecting precipitation change in this region: Sahel rainfall responds to the relative temperature of the North Atlantic, source of the moisture that converges in the region, with respect to the global tropical oceans. The temperature of the global tropical oceans, which is communicated first vertically through deep convection, then laterally by atmospheric waves [Chou and Neelin 2004, in J Climate; Held and Soden 2006, in J Climate; Sobel et al. 2001, in J Atmos Sci], broadly determines the threshold for convection. The temperature of the North Atlantic relative to that of the global tropical oceans measures the potential for the moist, but cool air that is converged onto the African continent from the adjacent ocean to lead to deep convection and precipitation.

This interpretation consistently explains past drought, partial recovery, and the current alternation of wet and dry states on time scales from daily to interannual. It also sheds light on the uncertainty in future projections, relating them to the uncertainty in patterns of sea surface temperature change [Giannini et al. 2013, in Env Res Lett]. This contribution aims to frame the physical context in which to discuss societal response to drought, and its applicability to adaptation to current variability and future change.

Long-term evolution of feedbacks in a GCM run to equilibrium

A. Hannart (CNRS, Buenos Aires, Argentina), J.-L. Dufresne, (CNRS, Paris, France), R. Knutti (ETH, Zurich, Switzerland), C. Li, (Max Planck Institute for Meteorology, Hamburg, Germany)

Abstract details
Long-term evolution of feedbacks in a GCM run to equilibrium

A. Hannart (1) ; JL. Dufresne, (2) ; R. Knutti (3) ; C. Li, (4)
(1) CNRS, IFAECI, Buenos Aires, Argentina; (2) CNRS, Lmd, Paris, France; (3) Max Planck Institute for Meteorology, The Ocean in the Earth System, Hamburg, Germany; (4) ETH, Zurich, Switzerland

Abstract content

Equilibrium climate sensitivity (ECS) is a common measure of Earth’s global temperature response to radiative forcing. To first order, Earth’s radiative imbalance decreases linearly with global temperature anomaly when a step forcing is applied. ECS is thereby assumed to be constant and inversely related to the linear feedback coefficient (λ). The validity of this linear approximation has been increasingly questioned recently, with several studies consistently showing that feedback are state- and time-dependent. But can these deviations from linearity significantly shift the final equilibrium state actually reached by a GCM as compared to the linear extrapolation based on the first few centuries given by ECS, or even imply a bifurcation leading to a vastly different equilibrium? What are the dominant physical processes that explain them? Here we analyze the evolution of feedbacks over a six thousand years-long integration to equilibrium of the coupled climate model ECHAM5/MPIOM under atmospheric CO2 quadrupling –the only simulation to equilibrium of a state-of-the-art GCM available to our knowledge. Consistent with previous studies, the global feedback λchanges significantly over the entire integration, but more surprisingly its evolution is markedly non-monotonous. Indeed, while λ progressively decreases (1.5 down to 0.4 Wm-2K-1) during the first thousand years of the run – which is the period partly analyzed by most previous studies and hence yields consistent findings, we observe a steep increase in λ(0.4 up to 1.8 Wm-2K-1) throughout the remaining five thousand years, until the equilibrium is reached. This evolution is predominantly driven by: (i) a steeper and steeper increase in cloud feedback over the entire simulation which is mainly associated to decreasing low level cloud fractions over high latitude oceanic areas; combined with (ii) a non-monotonous evolution of water feedback characterized by a moderate increase during the first thousand years followed by a steeper and steeper decrease until equilibrium, both being mainly associated to changes in relative humidity over tropical land areas and near the surface. We discuss implications of these results for the above two questions: while the assumption of a linear global feedback may be a useful approximation in energy balance models for the century time scale temperature response, we argue that its validity has not been demonstrated to represent the final equilibrium. In particular, our results suggest that a change in the sign of λ on the very long term, thereby implying a bifurcation potentially leading to a vastly different equilibrium, can not be discarded, given that the relevant processes are poorly understood.

Past and future aerosol emission reductions and their impact on Arctic climate

H.-C. Hansson (Stockholm University, Stockholm, France), V. D. Varma, (STockholm University, Stockholm, Sweden), N. J. A. Acosta (Stockholm University, Stockholm, Sweden), Ø. Seland (Norweigan Meteorological Institut, Oslo, Norway), T. Iversen (Norweigan Meteorological Institute, Oslo, Norway)

Abstract details
Past and future aerosol emission reductions and their impact on Arctic climate

HC. Hansson (1) ; VD. Varma, (2) ; NJA. Acosta (3) ; Ø. Seland (4) ; T. Iversen (5)
(1) Stockholm University, Environmental Sccience and Analytical Chemistry, Stockholm, France; (2) STockholm University, Meteorology, Stockholm, Sweden; (3) Stockholm University, Environmental science and analytical science, Stockholm, Sweden; (4) Norweigan Meteorological Institut, Research, Oslo, Norway; (5) Norweigan Meteorological Institute, Research, Oslo, Norway

Abstract content

INTRODUCTION

Large changes in the magnitude and spatial patterns of global aerosol emissions have occurred during the 20th century and are projected to continue over the coming century. It is not clear how the Arctic climate has been affected by changes in global and European aerosol emissions. As more and more countries adapt different strategies to reduce air pollution, it is important to examine how this will affect not only the top-of-the-atmosphere radiative forcing, but rather the climate and other parameters important for society. Of particular relevance for the Arctic are the reductions in sulfate emissions from industrial activities, domestic heating, and power production that have taken place in Europe during the latest decades. These changes provide an opportunity to in detail by available observations study how regional emissions over Europe have affected the global and specifically the Arctic climate. This to bring more confidence in projections of how future air pollution emissions will affect especially the Arctic climate.

 

METHODS

Transient climate simulations over the industrial period (1850 to present) from the Norwegian Earth system model NorESM (Kirkevåg et al., 2013) with different emission levels have be conducted and analysed. The simulated aerosol number size distribution and mass composition have been evaluated versus in-situ observations from different European measurement networks. The analysis is focused on Europe and the Arctic and how surface radiative flux and temperature changes relate to different emission scenarios. We present results from a comparison between the actual estimated emissions 1850 to 2005 NorESM simulations with simulations using constant 1980 SOx emissions for Europe while all other are the actual emissions. We will also present climate projections, 2015 - 2080 with global emission of sulfur, organic and black carbon according to Current Legislation (CLE) or Maximum Feasible Reduction (MFR), assuming CO2 emission according to RCP4.5.

 

CONCLUSIONS

The simulations show a significant change in the Arctic temperature as a result of the past air quality regulation in Europe giving strongly decreased SOx emissions especially during the 1990-ties. However the effect in Europe is quite smaller. Comparing with the actual temperature change in the Arctic implies that the sulphate aerosol contribute with about 20% of the temperature increase in the Arctic. In the projections with CO2 emissions according to RCP4.5 the future temperature will increase in the Arctic will be about 3 K but with in an air pollution emission reductions according to MFR it will be about 4 K.

These is only results from simulations with one model and have to be corroborate by other models but also by detailed studies to identify the key processes and evaluation of the models towards observations. However our results strongly imply that a continued strong temperature increase will prevail for many years forward.

REFERENCES

Kirkevåg, A., T. Iversen, Ø. Seland, C. Hoose, J. E. Kristjansson, H. Struthers, A. M. L. Ekman, S. Ghan, J. Griesfeller, E. D. Nilsson, and M. Schulz (2013). Aerosol-climate interactions in the Norwegian Earth System Model – NorESM1-M. Geophysical Model Development, 6, 207-244.

Panel discussion

Abstract details
Panel discussion
Abstract content

Deep past carbon cycle and climate crises

Y. Donnadieu (LSCE, Gif-sur-Yvette, France), G. Ramstein (LSCE, Gif-sur-Yvette, France), Y. Godderis (CNRS , Toulouse, France)

Abstract details
Deep past carbon cycle and climate crises

Y. Donnadieu (1) ; G. Ramstein (1) ; Y. Godderis (2)
(1) LSCE, Gif-sur-Yvette, France; (2) CNRS , Géosciences environnement toulouse, Toulouse, France

Abstract content

Deep past carbon cycle and climate crises

Relationships between carbon cycle and climate are intensively studied for ongoing global warming. There are also in the geological past of the Earth, key periods when huge changes of atmospheric carbon led to drastic cooling: Snowball episods at Neoproterozoïc [800-600 Ma]. Another period of interest is the so-called "terrestrialization" of the continents which also led to a large decrease of atmospheric carbon but without producing an important cooling during Late Devonian [379-359 Ma]. For the first time period, we will show how tectonics-climate perturbation altered sufficiently the carbon cycle and the climate to produce a long lasting global glaciation. It is also the same carbon cycle that allow to escape from a frozen planet. For the second time period :Late Devonian also corresponds to a drastic decrease of atmospheric CO2 associated with vegetation development on continents but paradoxically this decrease doesn’t produce a cooling of the Earth because mainly of albedo feedback.

We shall discuss how these deregulations developed and which processess associated with atmospheric carbon- tectonics - biosphere and water cycle may explain the onset and the decay of such crises.

 

Causes and consequences of mid-Holocene aridity in mid-continental Eurasia in the CMIP5 simulations

S. Harrison (University of Reading, Reading, United Kingdom), K. Izumi, (Laboratoire de Météorologie Dynamique, IPSL, CNRS, Paris, France), G. Li, (Macquarie University, North Ryde, Australia), P. Bartlein, (University of oregon, Eugene, United States of America)

Abstract details
Causes and consequences of mid-Holocene aridity in mid-continental Eurasia in the CMIP5 simulations

S. Harrison (1) ; K. Izumi, (2) ; G. Li, (3) ; P. Bartlein, (4)
(1) University of Reading, Reading, United Kingdom; (2) Laboratoire de Météorologie Dynamique, IPSL, CNRS, Paris, France; (3) Macquarie University, Department of biological sciences, North Ryde, Australia; (4) University of oregon, Department of geography, Eugene, United States of America

Abstract content

The extent of mid-continental drying in Eurasia during the mid-Holocene is an example of a persistent regional mismatch between models and observations. The CMIP5 mid-Holocene simulations show drier conditions in Eurasia, particularly between 45°-60° N, whereas several types of palaeoenvironmental data systematically show that the region was wetter than today. At the same time, the models show significantly higher summer temperature, whereas observations indicate that summers were cooler. The simulated temperature bias can be up to 4-6 °C. Temperature biases in the CMIP5 historical (20th century) simulations are linked to systematic biases in evapotranspiration. Diagnosis of the surface energy balance in the mid-Holocene CMIP5 experiments shows that the simulated increase in summer temperatures results from the simulation of too-low evaporative cooling because of water limitation.  Surface water- and energy-balance interactions play a similar role in mediating the temperature response in CMIP5 future simulations.

 

Climate Change in the Past 2000 Years and its impact on society on the Tibetan Plateau

T. Yao (Chinese Academy of Sciences, Beijing, China), X. Yang (Chinese Academy of Sciences, Beijing, China), P. Yao (Chinese Academy of Sciences, Beijing, China), N. Wang (Chinese Academy of Sciences, Beijing, China), L. Tian (Chinese Academy of Sciences, Beijing, China), B. Xu (Chinese Academy of Sciences, Beijing, China), H. Zhao (Chinese Academy of Sciences, Beijing, China), J. Gao (Chinese Academy of Sciences, Beijing, China), D. R. Joswiak (Chinese Academy of Sciences, Beijing, China)

Abstract details
Climate Change in the Past 2000 Years and its impact on society on the Tibetan Plateau

T. Yao (1) ; X. Yang (1) ; P. Yao (1) ; N. Wang (2) ; L. Tian (1) ; B. Xu (1) ; H. Zhao (1) ; J. Gao (1) ; DR. Joswiak (1)
(1) Chinese Academy of Sciences, Institute of tibetan plateau research, Beijing, China; (2) Chinese Academy of Sciences, Cold and arid regions environmental and engineering research institute, Beijing, China

Abstract content

Temperature variation on the Tibetan Plateau (TP) in the past 2000 years is reconstructed using stable oxygen isotope in five ice core records on the TP, including Dunde ice core in northeast TP, Guliya ice core in northwest TP, Dasuopu ice core in south TP and the Puruogangri and Tanggula ice cores in central TP. The integration of those ice core records reveals the synchronicity of large-scale climate changes in Tibet, such as the warming in the 7th century, 12-13th centuries and the present, and the cooling in the 3th century, 16th century, and 19th century. We referred to human historical documentary record since A.D. 620 for possible responses of social, economic and military activities to climate changes. By focusing especially on human activities and social development directly determined or indirectly influenced by climate from historical documentary record, we quantified those events into five aspects, i.e., basic resources, economic development, military strength, national coherence, and cultural and religious development, to study social development on the TP by A.D. 1900. Our results show a close Tibetan societal response to climate changes in the past 2000 years, particularly before the modern ages.

AMOC Evolution in the Last Deglaciation:Forcing Mechanism, Thermohaline Instability and Implications

Z. Liu (University of Wisconsin-Madison, Madison, United States of America)

Abstract details
AMOC Evolution in the Last Deglaciation:Forcing Mechanism, Thermohaline Instability and Implications

Z. Liu (1)
(1) University of Wisconsin-Madison, Atospheric and Oceanic Sciences, Madison, United States of America

Abstract content

The forcing mechanism and instability of the Atlantic Meridional Overtuning Circulation (AMOC) over the last 21,000 years is studied using transient simulations under realistic forcings in the NCAR-CCSM3.  First, in addition to the strong millennial AMOC variability forced by melting water fluxes, the background AMOC is determined by two opposing effects: the intensification by the rising atmospheric CO2 and the reduction by the retreating ice sheet, both through the sea ice feedbacks in the North Atlantic. As a result, the AMOC strength does not change significantly after the deglaciation. Second, the model AMOC exhibits a monostable behavior.  This monostable AMOC, which has been observed in almost all state-of-art coupled general climate models (CGCMs),  is likely to be caused by a systematic model bias that is associated with the  tropical bias, the resulted freshwater flux and AMOC freshwater export. This AMOC over-stabilization bias needs to be improved in these CGCMs to allow for a credible projection of AMOC evolution in the future.

Different vegetation responses to climatic droughts in the Mediterranean basin

C. Gouveia (Instituto Dom Luiz (IDL), Lisboa, Portugal), R. Trigo (Instituto Dom Luiz (IDL), Lisboa, Portugal), S. Beguería, (Estación Experimental de Aula Dei, Zaragoza, Spain), S. M. Vicente-Serrano (Instituto Pirenaico de Ecología, Zaragoza, Spain)

Abstract details
Different vegetation responses to climatic droughts in the Mediterranean basin

C. Gouveia (1) ; R. Trigo (1) ; S. Beguería, (2) ; SM. Vicente-Serrano (3)
(1) Instituto Dom Luiz (IDL), Faculdade de Ciências da Universidade de Lisboa, Lisboa, Portugal; (2) Estación Experimental de Aula Dei, Csic, Zaragoza, Spain; (3) Instituto Pirenaico de Ecología, Zaragoza, Spain

Abstract content

A number of recent studies have identified a significant increase in the frequency of drought events in the Mediterranean basin. Climatic droughts are relatively frequent in the Mediterranean region as a consequence of the large interannual variability of precipitation, and long periods with low precipitation. Several studies characterized climatic droughts in the Mediterranean region, emphasizing the spatial and temporal complexity of this phenomenon.

The use of remote sensing data holds a great potential since it allows analyzing large regions with a consistent dataset with high spatial and temporal resolution. Several studies analyzed the impact of droughts on the natural vegetation and crops using remote sensing data. The majority of the studies about drought impacts on the vegetation cover considered droughts as a precipitation shortage regarding the normal climatology. Nevertheless, recent drought episodes in the frame of a warming scenario have shown that large evapotranspiration rates associated to high temperatures are equally relevant and may trigger severe droughts. Therefore, in the present scenario of temperature rise it is necessary to acknowledge the impacts of both reduction in precipitation and increase in evapotranspiration rates.

The aim of the present work is to analyze in detail the impacts of drought episodes on vegetation in the Mediterranean basin behavior using NDVI data from (from GIMMS) for entire Mediterranean basin (1982-2006) and the multi-scale drought index (the Standardised Precipitation-Evapotranspiration Index (SPEI).

Correlation maps between fields of monthly NDVI and SPEI for at different time scales (1-24 months) were computed in order to identify the regions and seasons most affected by droughts. Affected vegetation presents high spatial and seasonal variability, with a maximum in summer and a minimum in winter. During February 50% of the affected pixels corresponded to a time scale of 6 months, while in November the most frequent time scale corresponded to 3 months, representing more than 40% of the affected region. Around 20% of grid points corresponded to the longer time scales (18 and 24 months), persisting fairly constant along the year. The strongest control of droughts on vegetation dynamics is obtained during February and May for drier clusters in areas with low water balance values. Accordingly the wet and cold seasons present low water balance values that implies shorter time scales over dry cluster, whereas high water balance values implies longer time scales over Central and Atlantic clusters.

The occurrence of most affected areas over regions presenting low water balance values highlights the strong dependence of vegetation with climate variability. Furthermore, this conclusion is reinforced by the strong control of drought on vegetation activity observed for Arid and Steppe clusters located over areas with higher absolute values of water balance. The projected increase in frequency of drought episodes emphasize the need for an early warning drought system covering the entire Mediterranean basin. However, our results highlight that this requirement is dependent of vegetation types, season of the year and relative location of the regional sector considered. We are confident that our results will provide a useful tool for drought management plans and will play a relevant role in mitigating the impact of such episodes within the context of climate change.

 

16:30

Observed large leads of Atlantic circulation slowdowns with respect to tropical precipitation events: A challenge for current climate models

P. Burckel, (LSCE/IPSL - CNRS, Gif-sur-Yvette, France), C. Waelbroeck (LSCE/IPSL - CNRS, Gif-sur-Yvette, France), S. Pichat, (LGL-TPE, Lyon, France), J. Gherardi (Ifremer, Issy les Moulineaux Cedex, France), H. Arz, (Leibniz Institute for Baltic Sea Research, Rostock, Germany), J. Lippold, (Oeschger Centre for Climate Change Research, Bern, Switzerland), T. Dokken, (BCCR, Bergen, Norway), F. Thil, (LSCE/IPSL - CNRS, Gif-sur-Yvette, France)

Abstract details
Observed large leads of Atlantic circulation slowdowns with respect to tropical precipitation events: A challenge for current climate models

C. Waelbroeck (1) ; P. Burckel, (1) ; S. Pichat, (2) ; J. Gherardi (3) ; H. Arz, (4) ; J. Lippold, (5) ; T. Dokken, (6) ; F. Thil, (1)
(1) LSCE/IPSL - CNRS, Gif-sur-Yvette, France; (2) LGL-TPE, Ecole normale supérieure de lyon, Lyon, France; (3) Ifremer, DAEI, Issy les Moulineaux Cedex, France; (4) Leibniz Institute for Baltic Sea Research, Rostock, Germany; (5) Oeschger Centre for Climate Change Research, University of bern, Bern, Switzerland; (6) BCCR, University of bergen, Bergen, Norway

Abstract content

Marine cores MD09-3257 and GeoB3910 were retrieved off northern Brazil at ~4°S, 36°W and 2340 m water depth; their chronology is derived from a corrected and calibrated radiocarbon-based age-depth model. As in other marine cores from this area, XRF measurements show marked Ti/Ca and Fe/Ca peaks over the last glacial that have been interpreted to reflect increased terrigenous input due to increased precipitation and runoff from the adjacent continent. These terrigenous peaks can be shown to be coeval with precipitation events recorded in speleothem d18O records from South America in the 0-30°S latitudinal band, which are considered to reflect southward shifts of the intertropical convergence zone during Greenland stadials.

New sedimentary 231Pa/230Th measurements from core MD09-3257 show that large changes took place in the overlying water mass flow rate over the last glacial in conjunction with millennial precipitation events. Furthermore, our C. wuellerstorfi δ13C data indicate that water ventilation was reduced at ~2340 m in the tropical Atlantic during Greenland stadials, in phase with the reduction in water mass flow revealed by Pa/Th data. Our results thus demonstrate that major slowdowns of the Atlantic Meridional Overturning Circulation (AMOC) upper circulation cell took place during Greenland stadials.

Because both rainfall events and changes in ocean circulation are recorded in the same sediment core, we were able to reliably determine that the AMOC started to slowdown 1420 ± 250 and 690 ± 180 (1σ) y before the onset of two large South American precipitation events associated with Heinrich stadials.

Our data open new prospects concerning causal mechanisms of rapid climate changes. More specifically, current climate models simulate a rapid response of the tropical climate to AMOC changes, whereas our results indicate that there is a large lead of AMOC changes with respect to tropical climate changes. Therefore, more work, both data- and model-wise, is necessary in order to achieve a better understanding of the highly nonlinear behavior of the climate system observed in climate archives and ensure that comprehensive climate models used for climate projections capture the full extent of ocean-ice sheets-atmosphere interactions.