01 LD 0201

1.4.2001 - 31.3.2006

Climate transitions: Forcing and feedback mechanisms of glacial-interglacial and future climate change

G. Lohmann, K. Grosfeld, M. Dima, G. Knorr, V. Romanova,

S. Schubert, L. Könnecke, S. Zech

Geosciences Department, Earth System Modelling and Analysis, University of Bremen, P.O. 330440, 28334 Bremen

gerrit.lohmann@dkrz.de; http://www.palmod.uni-bremen.de/~gerrit/DEKLIM.html

Key words: Climate transitions, deglaciation, ocean circulation, multidecadal variability, proxy data, climate modelling, statistical data analysis

1. Summary

The oceanic thermohaline circulation (THC) has a central position in the understanding of climate variability and predictability because of its link to long-term variability and climate changes. Potential thresholds and interactions of the THC with the hydrological cycle are investigated for glacial and present day conditions. Analyses of ice-core and ocean-sediment records have shown that for the last two deglaciations, changes in temperatures over Antarctica and the tropics precede changes in Greenland ice volume. These chronologies calls into question many of the deglaciation mechanisms. Teleconnections and their role for long-term climate variability are investigated using numerical models of the Earth's system. One further focus involves the reconstruction of climate, based on observational, model and proxy data. The statistical analysis provides a synthesis of paleoclimate and simulated data.

2. Aim of the research in the framework of DEKLIM

It is of vital importance to understand whether increasing human population and industrialization have already caused, or have the potential to induce significant changes in Earth's climate. Unfortunately, the type of direct temperature measurement records which enable climate changes on a global scale are too short, and they fall already within the period of strong human impact on natural conditions. Information regarding the pre-anthropogenic state of the system can be obtained either from proxies that record past climate and environmental conditions, or by simulating the climate system with appropriate external forcing. The paleoclimate record provides an excellent test of these models as it reveals climate variations that have occurred in the past.

Within the framework of DEKLIM, we combine the different approaches of Earth System Modelling, statistical analyses of historical and proxy data, as well as work with conceptual models. We closely collaborate with people from the data side of the paleoclimate community in order to understand variations documented in proxy data with a high temporal resolution. Computer simulations and conceptual models contribute to system-analytical understanding of the dynamics in the climate system.

3. Recent and completed activities

Analyzing paleoclimatic records and models in tandem enables the evaluation of climate transitions and the analysis of forcing and feedback mechanisms of glacial-interglacial and future climate changes. Paleoclimatic time slices and specific climate transitions are simulated. An Earth System Model is used for the study of climate dynamics on decadal to multi-millennial time-scales consisting of PUMA/ECHAM (atmosphere), LSG/HAMOCC (ocean), and LPJ (land) model components. Further activities are related to teaching, organizing of workshops and conferences (EGU, AGU, Humboldt-NAS), and public relations.

4. Principal results

Stability of the THC. (Prange et al., 2002; Lohmann, 2003; Prange et al., 2003; Romanova et al., 2003; Dima and Lohmann, 2003)

One possible mechanism for a threshold in the climate system is associated with the hydrological cycle, one of the main driving forces for the THC. The stability of the glacial THC with respect to North Atlantic freshwater input is examined using a global ocean general circulation model. It is found that the quasi-equilibrium hysteresis behaviour is much less pronounced under glacial conditions than under present-day conditions. The results may help to assess the effect of iceberg invasions and meltwater events, suggesting that the THC is prone to instability during a deglaciation phase when the Atlantic meridional overturning is weakened. Under full glacial conditions, however, the THC is mono-stable and even extreme freshwater pulses are unable to exert a persistent effect on the conveyor. Characteristic patterns of atmospheric water vapor transport and their link to atmospheric variability are evaluated. Furthermore, we analyse the stability of the THC with respect to vertical mixing in the ocean.

Southern Ocean origin for resumption of Atlantic THC during deglaciation. (Knorr and Lohmann, 2003)

During the last two deglaciations Southern Hemisphere warming preceded Greenland warming and the northern Atlantic has been exposed to meltwater discharge that is known to reduce North Atlantic deep water (NADW) formation. Yet, deglaciation is accompanied by a transition from a weak glacial to a strong interglacial Atlantic THC. We utilize a three-dimensional ocean circulation model to investigate the impact of Southern Ocean warming and associated sea ice retreat onto the Atlantic THC. We show that gradual warming in the Southern Ocean induces an abrupt resumption of the THC by increased mass transport via the warm and cold water route of the global oceanic conveyor belt circulation. This effect prevails over the destabilizing effect of deglacial meltwater input to the northern Atlantic. The mechanism provides a consistent picture of Southern and Northern Hemisphere climate change in agreement with proxy records during deglaciation.

Atmospheric teleconnections. (Rodgers et al., 2003)

We have presented a mechanism for deglaciation whereby changes in tropical SSTs can impact the stability of the Laurentide Ice Sheet via an atmospheric bridge. This study is based on atmospheric general circulation experiments whereby changes in tropical SST can drive changes in air temperature over modern-day Canada. Our tropical feedback mechanism for Northern Hemispheric deglaciation accounts for the main shortcomings of theories which rely on internal ice sheet dynamics to explain deglaciation, namely the problem that the changes in the tropics and high-latitude Southern Hemisphere occur before ice volume changes.

Climate shift in the 70th. (Grosfeld et al., 2003; Rimbu et al., 2003)

We investigate the signature of North Atlantic sea surface temperature (SST) and sea level pressure (SLP) multidecadal variability, based on statistical analysis of observed and proxy data as well as on general circulation model experiments. To aims this, we performed ensemble integrations with an atmospheric general circulation model of intermediate complexity, forced with observed global SST over the last 145 years. The signature of North Atlantic multidecadal SST variability has a monopolar structure whereas the SLP variability is found to be similar to that of the North Atlantic Oscillation (NAO). The signature of multidecadal North Atlantic variability is also well recorded in independent proxy time series of Read Sea corals and Caribbean Sea sediments. The consistency between these independent data sets from different locations of the North Atlantic realm indicates a hemispheric mode and enables a continuation of multidecadal signals such as SST variability into the past over the last centennial.

5. Main conclusions

Different forcing and feedback mechanisms of glacial-interglacial and future climate change have been evaluated. The sensitivity of the ocean circulation is strongly dependent on the climatic background condition and associated hydrological cycle. We find different teleconnections for deglaciation scenarios. One mechanisms is linked to tropical-extratropical connections in which tropical SST anomalies propagate to the region corresponding to the Laurentide Ice Sheet through the atmosphere. Another teleconnection show that changes in the Southern Ocean propagate to the high-latitude northern hemisphere via the resumption of the ocean circulation. This interhemispheric teleconnection suggests that the "Achilles Heel" and the "Flywheel" of the Atlantic overturning circulation are located at the North Atlantic and the Southern Ocean, respectively. The analysis of the climate shift in the 70th and associated multi-decadal climate variability suggests a potential predictive skill on multidecadal climate variability, which is also important for the analysis of future climate change.

6. Planned activities

Feedback mechanisms with the biosphere and the cryosphere. (work in prep.)

We have started to study the potential role of vegetation in buffering atmospheric CO2 increase during terminations using the LPJ model. This dynamical vegetation model is used to test the reconstructed variations in terrestrial biomass during the last deglaciation. In a later stage, the interaction of vegetation with atmospheric and oceanic dynamics, and the cryosphere is studied.

Model experiments will concentrate on the onset of vigorous NADW formation during the transition to the Boelling period including the feedbacks with the marine carbon cycle. Proxy data such as C14 and artificial sediment cores are simulated using the Knorr and Lohmann (2003)-scenario for the transition to the Boelling warm period.

Early detection of THC changes. (Rühlemann et al., 2003; work in prep.)

In order to trace a THC slowdown, we suggest that the temperature at mid-depth provides a suitable measure for obtaining large-scale changes of deep circulation. A benchmark test to further evaluate the suitability of the tropical intermediate-depth Atlantic as indicator of THC change is the study of past collapses of the THC, such as occurred during the last deglaciation. It is seen that strong reductions in the THC during past times (Heinrich event H1 and the Younger Dryas) were associated with rapid and intense warming of intermediate-depth waters. Numerical and conceptual climate models in combination with observations and paleoclimate data will demonstrate that a similar temperature pattern is expected for a reduction in modern thermohaline overturning.

Statistical interpretation of geological data. (Lohmann, 2002; Rimbu et al., 2003; Lohmann and Rimbu, 2003; work in prep.).

Reconstructing past climates not only affords us to understand the present climatic system, and it also provides potential analogues for future climate change. One main focus involves the reconstruction of climate, based on observational, model and proxy data. The statistical analysis provides furthermore a synthesis, comparison and interpretation of paleo and simulated data. To reconstruct climate change over thousands of years it is necessary to use the evidence provided by sediments, ice cores, pollen data, fossil and isotope records. One question is related to find key locations for climate modes (e.g., the NAO) with stable teleconnections over time.

Modelling of climate transitions.

To strengthen the model results of abrupt climate change, it is considered to perform relatively short model integrations with the ECHAM model. The results of these simulations will be compared with proxy data available.

7. Cooperation within DEKLIM and with other programs

There is a strong link with DEKLIM projects GHOST (Schneider et al., Univ. Bremen), Stable oxygen isotopes (Mulitza, Univ. Bremen), and the research group RESPIC (Fischer, AWI). Furthermore, there are several close cooperations with the BMBF founded project KIHZ, and a project MIOCENE founded by the DFG (Bickert et al., Univ. Bremen). There is also a cooperation within the Earth System Modelling of Intermediate Complexity initiative (Claussen et al., 2002). Other cooperations are with the Univ. Hamburg, MPIs Jena and Hamburg, Univ. Wisconsin and Paris.

8. Policy relevance and application

One of the great uncertainties in the prediction of future climate is the behavior of the thermohaline circulation. Based on various climate modeling studies the Intergovernmental Panel on Climate Change (IPCC) has concluded that anthropogenic increases in atmospheric greenhouse-gas concentrations will possibly cause a weakening or even a shut-down of the Atlantic THC.

It can be asked whether the present ocean observation system would detect a change in the THC in time to be useful for climate policy. While the point of no return represents a threshold of the physical system, the policy decision must be made significantly earlier because climate policy has to account for the inertia of social and technological systems. Due to difficulties and uncertainties in direct THC measurements and internal climate variability, it may be argued that an ocean monitoring program continuing on current trends would likely fail to detect a THC shut down in time to be useful for climate policy. We suggest a different strategy to detect the beginnings of a potential THC collapse by means of changes in the temperature structure (mid-depth warming). This concept is similar to the optimal fingerprint techniques in which spatial field informations are used to eliminate the noise from the climate signal.

References

Claussen, M., et al., 2002, Earth System Models of Intermediate Complexity: Closing the Gap in the Spectrum of Climate System Models, Climate Dynamics, 18, 579-586.

Dima, M., and Lohmann, G., 2003, A solar influence on the THC?, Earth and Planetary Science Letters (in revision).

Grosfeld, K., Lohmann, G., Rimbu, N., Lunkeit, F., and Fraedrich, F., 2003, Predictable Response of the Atmospheric Circulation on North Atlantic Multidecadal Variability, Journal of Climate (submitted).

Knorr, G., and Lohmann, G., 2003, Resumption of the Atlantic conveyor circulation via Southern Ocean warming during deglaciation, Nature (accepted for publication).

Lohmann, G., 2002, Meteorologische Interpretation geologischer Daten - neue Wege in der Paläoklimaforschung. Promet, 28(3/4), 147-152.

Lohmann, G., 2003, Atmospheric and oceanic freshwater transport during weak Atlantic overturning circulation, Tellus A (in press).

Lohmann, G., and Rimbu, N., 2003, Climate signature of solar irradiance variations: Analysis of long-term instrumental and historical data, International Journal of Climatology (submitted).

Prange, M., V. Romanova, and G. Lohmann, 2002, The glacial thermohaline circulation: stable or unstable? Geophysical Research Letters, 29 (21), 2028-2031.

Prange, M., Lohmann, G., and Paul, A., 2003, Influence of vertical mixing on the thermohaline hysteresis: Analyses of an OGCM. J. Phys. Oceanogr., 33(8), (to appear).

Rimbu, N., Lohmann, G., Felis, T., and Pätzold, J., 2003, Shift in ENSO teleconnections recorded by a Red Sea coral. J. Climate, 16(9), 1414-1422.

Rodgers, K., Lohmann, G., Lorenz, S., Schneider, R., and Henderson, G., 2003, A Tropical Mechanism for Northern Hemisphere Deglaciation. Geochem., Geophys., Geosyst., 4(5), 1046, doi: 10.1029/2003GC0000508.

Romanova, V., Prange, M., and Lohmann, G.,2003, On the stability of the glacial THC and its dependence on the background hydrological cycle. Climate Dynamics (in revision).

Rühlemann, C., Mulitza, S., Lohmann, G., Paul, A., Prange, M., and Wefer, G., 2003, Abrupt warming of the intermediate-depth Atlantic Ocean in response to thermohaline circulation slowdown during the last deglaciation. PAGES NEWS, 11( 1), 17-19.