Younger Dryas: A data to model comparison to constrain the strength of the overturning circulation

Transcription

1 Click Here for Full Article GEOPHYSICAL RESEARCH LETTERS, VOL. 34, L21705, doi: /2007gl031304, 2007 Younger Dryas: A data to model comparison to constrain the strength of the overturning circulation Katrin J. Meissner 1 Received 12 July 2007; revised 24 September 2007; accepted 9 October 2007; published 8 November [1] The University of Victoria Earth System Climate Model (UVic ESCM) is used to compare simulated time series of radiocarbon during the Younger Dryas (YD) with paleoceanographic records. I find that only a complete shutdown and recovery of the Atlantic Meridional Overturning Circulation (AMOC) can simulate both the rise in atmospheric CO 2 concentrations seen in ice core records and the peak and subsequent decrease in atmospheric D 14 C comparable to the peak recorded in the varved sediments of the Cariaco Basin. Simulated radiocarbon profiles in the western North Atlantic match well with data from deep-sea corals at the beginning of the YD, whereas planktonic/ benthic foraminifera records match best with a transient state during the rapid recovery of the AMOC. The steepness of the increase in atmospheric D 14 Cat12.9 ka cal could not be simulated with oceanic circulation changes only because the response time of the climate system is too slow. Citation: Meissner, K. J. (2007), Younger Dryas: A data to model comparison to constrain the strength of the overturning circulation, Geophys. Res. Lett., 34, L21705, doi: / 2007GL Introduction 1 School of Earth and Ocean Sciences, University of Victoria, Victoria, Canada. Copyright 2007 by the American Geophysical Union /07/2007GL031304$05.00 [2] The Younger Dryas (YD, cal ka) is one of the three major abrupt cold events that interrupted the warming of the Northern Hemisphere during the last deglaciation. It can be recognized in a variety of tracers in terrestrial [e.g., Mathewes et al., 1993], marine [e.g., Broecker et al., 1989] and ice records [e.g., Alley et al., 1993] across the Northern Hemisphere. Johnson and McClure [1976] first postulated that an increased freshwater flux to the St. Lawrence River associated with the diversion of continental runoff from the Mississippi River triggered the YD cold event by causing an increase in North Atlantic sea ice. Rooth [1982] subsequently proposed that this Mississippi-to-St. Lawrence routing caused the YD through its effect on the thermohaline circulation and attendant poleward ocean heat transport. [3] The YD is a relatively recent event. Radiocarbon (D 14 C) is therefore a valuable proxy that can help us to understand the trigger and mechanism of this abrupt cooling. 14 C is produced at the top of the atmosphere where neutrons interact with nitrogen atoms. The production rate is highly dependent on solar activity and the strength of the terrestrial magnetic field [Muscheler et al., 2004]. After production, 14 C oxidizes rapidly to 14 CO 2 and is incorporated in the global carbon cycle. Since radiocarbon decays with a half-life of 5730 years [Godwin, 1962], carbon reservoirs which have been isolated from the atmosphere for a long time (such as the deep ocean) contain considerably lower amounts of 14 C than newer carbon reservoirs. Changes in the oceanic circulation and ventilation will therefore alter the atmosphere-ocean 14 Cflux.Theatmospheric D 14 C is influenced both by production at the top of the atmosphere and the exchange with old carbon reservoirs. [4] Atmospheric D 14 C records are compiled from paleo proxies in lake sediments [Goslar et al., 1995,1999],corals [Fairbanks et al., 2005; Chiu et al., 2007] and varved sediments of the Cariaco Basin [Hughen et al., 2000, 2004a, 2004b]. The tree-ring chronology [Stuiver et al., 1998; Friedrich et al., 2004] extendsbackto12.40calka and is therefore too young to capture the signature of the whole YD event. Radiocarbon is also used to reconstruct shifts in water masses and evaluate possible changes in the oceanic meridional heat transport. D 14 C in deep-sea corals [Robinson et al., 2005;Eltgroth et al., 2006]andtheage difference between benthic and planktonic foraminifera [Keigwin, 2004] give an insight about the change in ventilation and water masses at a certain location. [5] In the last ten years, numerous studies have been published in an attempt to understand what caused the variability in the atmospheric radiocarbon record during the YD [e.g. Beck et al., 2001]. The researchers divide into two schools: those who emphasize on changes in the production rate [e.g., Muscheler et al., 2000; Marchal et al., 2001;Goslar et al., 2000]andthosewhohighlight evidence for a change in the oceanic circulation [e.g., Goslar et al., 1995; Hughen et al., 1998; Clark et al., 2002; Adkins and Pasquero, 2004; McManus et al., 2004]. Radiocarbon was also integrated into a variety of physical or carbon cycle models: Stocker and Wright [1996] used a zonally averaged physical ocean model and found an increase in atmospheric D 14 C of 35 per mill when the Atlantic Meridional Overturning Circulation (AMOC) shuts down. With a newer version of the same model, Marchal et al. [2001] could not explain the atmospheric D 14 Cpeak with either production or circulation changes. On the other hand, several carbon cycle box models have been employed: Goslar et al. [1999] concluded using the PANDORA box model that variations in geomagnetic field are too weak to explain the D 14 Cmaximumseeninthe records. The simulations of Köhler et al. [2005] using the BICYCLE box model showed that the modelled pco 2 and d 13 Cconcentrationsweremostconsistentwithdatawhen only a small change in ocean circulation was imposed; they concluded that the D 14 C record must be caused by changes in the production rate. L of5

2 Figure 1. (a) Freshwater flux added to the St. Lawrence River Delta (blue) and strength of the overturning motion for run A (red) and B (green) in Sv. Note that the blue shaded line is only valid for simulation B. The colored stars indicate the times of the D 14 CprofilesshowninFigure2. (b) Time series of global carbon budgets (minus mean value for each reservoir) in PgC for run B only, atmosphere (black), ocean (cyan), terrestrial carbon (orange). [6] Here I present the results of a state-of-the-art three dimensional ocean model, coupled to an atmosphere, sea ice, land surface and vegetation model when forced with a freshwater event. The model incorporates a fully coupled carbon cycle and prognostic D 14 C. The results are compared to the atmospheric D 14 C record [Hughen et al., 2004a], radiocarbon in benthic foraminifera [Keigwin, 2004] and deep-sea corals [Eltgroth et al., 2006], in order to constrain the strength of the AMOC during the YD event. I concentrate on the effect circulation changes have on the atmospheric D 14 C and keep therefore the 14 Cproduction rate constant during my simulations. 2. Model and Experimental Design [7] The University of Victoria (UVic) Earth System Climate Model (ESCM, version 2.8) consists of an ocean general circulation model (Modular Ocean Model version 2 [Pacanowski, 1995])coupledtoaverticallyintegratedtwo dimensional energy-moisture balance model of the atmosphere, a dynamic-thermodynamic sea ice model based on the works by Semtner [1976], Hibler [1979] and Hunke and Dukowitz [1997], a land surface scheme (simplified version of MOSES [Cox et al., 1999])andadynamicglobal vegetation model ( TRIFFID [Meissner et al., 2003b; Cox, 2001]).Themodelincludingtheatmospheric,ocean and sea ice components is described by Weaver et al. [2001]. It is driven by seasonal variations in solar insolation at the top of the atmosphere and seasonally-varying wind stress and wind fields [Kalnay et al., 1996]. The UVic ESCM also includes a fully coupled carbon cycle taking into account the terrestrial carbon fluxes and reservoirs [Meissner et al., 2003b;Matthews et al., 2005]aswellas the inorganic [Ewen et al., 2004]and organic[schmittner et al., 2007] carbon cycle in the ocean. Radiocarbon has been added as a prognostic tracer and is treated in carbon reservoirs and carbon fluxes like 12 C with the only difference being that it decays with time. Present day control simulations compare well to GEOSECS data [Meissner et al., 2003a]. [8] I integrate the model for over years into equilibrium under boundary conditions corresponding to 12.9 ka cal (orbital parameters of 12.9 ka cal [Berger, 1978], atmospheric CO 2 of 242 ppm and elevated topography [Peltier, 1994]). During this equilibrium run, I hold the atmospheric D 14 Cfixedat200permill.Attheendofthe equilibrium run, I diagnose the 14 C production rate at the top of the atmosphere needed to keep the atmospheric 14 C/ 12 C ratio constant. This procedure yields a production rate of 1.61 atoms/(cm 2 s). [9] The imposed meltwater event follows the reconstruction of Carlson et al. [2007] and consists of four different regimes in the discharge from the St. Lawrence River (see Figure 1): 0.06 Sv is added from 12.9 to 12.5 k cal; 0.12 Sv from 12.5 to 12.3 k cal; 0 Sv from 12.3 to 11.9 k cal and 0.07 Sv from 11.9 to 11.5 k cal. During the simulations of the meltwater event, the 14 C production rate is held constant and atmospheric 14 CandCO 2 are therefore calculated prognostically. Two simulations are carried out, one in which the freshwater scenario is set equal to the discharge from the St. Lawrence River (run A, red line in Figure 1a) and one in which the freshwater is added to the simulated discharge from St. Lawrence River (run B, green line in Figure 1a). The difference in freshwater forcing between run A and run Bdependsontheamountofprecipitationoverland(and therefore on the climate state); its average equals 0.03 Sv. 3. Results [10] Figure 1a shows the time series of the freshwater scenario (blue, solid line) as well as the maximum strength of the meridional overturning for runs A and B. Starting with a maximum strength of 18 Sv, the red simulation (run A) shows a weakening of 36% (11.5 Sv) followed by a short recovery and another weakening to 12.5 Sv. At the end of the event, run A recovers to its initial strength. Run B on the other hand shows a complete shut down of the AMOC and does not recover on its own after the freshwater event. As there is paleoceanographic data evidence that the thermohaline circulation recovered after the YD, I apply an unphysical salt flux of 0.12 Sv to the St. Lawrence River delta during simulation B in order to simulate CO 2 and 2of5

3 Figure 2. Simulated D 14 C profiles at 30 N, 70 W comparedtokeigwin s [2004] profile (black stars, connected by black line, error bars in gray) and Eltgroth et al. s [2006] deep-sea coral profile (pink circles connected by pink line). (a) Snapshots at 12.9, 12.8, 12.6, 12.2 and 11.7 k cal. (b) Snapshots at 11.4, 11.3, 11.2 and 11.1 k cal (see also Figure 1). D 14 C during a recovery (Figure 1a, blue dashed line). Run B reacts instantly with a recovery to the initial strength of the meridional overturning first and a further increase to a maximum of 35 Sv. [11] The global carbon budgets for run B are shown in Figure 1b. Carbon is transferred from the atmosphere (black line) and land (orange line) into the ocean (cyan line) during the time the AMOC is shut down. With the recovery of the AMOC deep water enriched in dissolved organic carbon (DIC) is brought to the surface where a net flux of oceanic inorganic carbon into the atmosphere takes place [Meissner et al., 2007].Thiscarbonaccumulatesintheatmosphere and results in an increase of the atmospheric CO 2 concentration above initial conditions. With warmer conditions in the Northern Hemisphere due to an enhanced oceanic heat transport and higher atmospheric CO 2 concentrations, vegetation and soil carbon also increases significantly. [12] D 14 C profiles in the western North Atlantic have been reconstructed using benthic/planktonic foraminifera [Keigwin, 2004] and deep-sea corals [Eltgroth et al., 2006] and are shown in Figure 2. The most striking characteristic of Keigwin s [2004] profile is the strong contrast in ventilation at 2.3 km depth between well ventilated Glacial North Atlantic Intermediate Water and poorly ventilated glacial deep water. Also shown in Figure 2 are snapshots of D 14 CprofilesfrommyBsimulationat 30 N, 70 W. The time of each snapshot is indicated in Figure 1 using colored stars (the colors correspond to the colors of the profiles in Figure 2). None of the simulated profiles match the high D 14 CvaluesofKeigwin [2004] above 2.3 km depth. The shut down of the AMOC results in agradualdecreaseofd 14 C in the deep ocean and a slight increase of D 14 C at the surface due to a longer residence time and therefore more effective gas exchange (Figure 2a). With the recovery of the AMOC, intermediate water masses get more ventilated and the positive D 14 Canomalypropagates downward (Figure 2b). This is the only time where the model captures a strong D 14 Cgradientatintermediate depth as indicated by foraminifera data. The last profile (in blue) shows that the gradient in D 14 Caround2000to3000 meter depths completely disappeared once the AMOC has recovered and the ocean had time to re-equilibrate. This profile is also similar to the first profile (in red) before the event. The two data points of the deep-sea corals (pink circles [Eltgroth et al., 2006])match best with my first two profiles at the beginning of the YD. This is consistent with the corals age (12.8 ka). Very recent data from the California Margin [Marchitto et al., 2007]show a depletion of radiocarbon in intermediate waters during the YD. My simulations do not simulate a similar drop at the same location (23.5 N, W). As our ocean model operates on a coarse grid ( ) with simplified topography, it is not able to simulate local shifts in circulation so close to shore. [13] Figure 3a shows the atmospheric D 14 C record [Hughen et al., 2004a]comparedtothesimulatedatmospheric D 14 C during run A (red) and B (green, solid line). A weakening of the AMOC (run A) is unable to trigger significant changes in the atmospheric D 14 C. Run B (green), however, shows a peak at 244 per mill and a subsequent decline which is similar to the data. The most important difference between model and data is due to the response time of the climate system in my model. Although the AMOC collapses within 300 years, atmospheric D 14 C does not react instantly and builds up slowly over the whole period the ventilation is reduced. Once the AMOC recovers, simulated atmospheric D 14 Cdeclinesatthesamerateas shown in the paleo data (Figure 3a, green dashed line). The model response in D 14 C lags the reconstruction by 900 years, indicating that the YD freshwater event might have been shorter than the scenario used for the present simulations. [14] Finally, Figure 3b compares the CO 2 concentrations from EPICA Dome C ice cores [Monnin et al., 2001]with the simulated atmospheric CO 2 concentrations (run A in red; run B in green, solid lines). As for D 14 C the model s response lags the reconstruction by several hundred years pointing to an earlier recovery of the AMOC than possible with the meltwater scenario chosen for the present simulations (the green dashed line shows the response of run B shifted back by 900 years). Although the resumption of the AMOC provokes a rise in atmospheric CO 2 which is similar to the initial rise in CO 2 seen in data, the simulated CO 2 levels out around 260 ppm. 4. Conclusions [15] By incorporating radiocarbon as a prognostic tracer in the UVic ESCM and presuming that the production rate 3of5

4 of 14 C was constant during the Younger Dryas (YD) event, several conclusions can be drawn. Only a complete shutdown of the Atlantic Meridional Overturning Circulation (AMOC) can simulate a peak in atmospheric D 14 Cofa magnitude comparable to the peak seen in paleo records [Hughen et al., 2004a]. However, the steepness of the increase in D 14 Cat12.9 ka cal [Hughen et al., 2004a] cannot be simulated with ocean circulation changes only because the response time of the climate system is too slow. The rapid decrease of D 14 C after the peak is simulated in rate and magnitude if the AMOC recovers rapidly from a complete shut-down state. A quick recovery of the AMOC also results in a rise in atmospheric CO 2 which is comparable to ice core data [Monnin et al., 2001].Radiocarbon profiles in the western North Atlantic match well with data from deep-sea corals at the beginning of the YD [Eltgroth et al., 2006], whereas planktonic/benthic foraminifera records [Keigwin, 2004]matchbestwithatransientresponseduring the rapid recovery of the AMOC from a complete shutdown state. [16] Acknowledgments. The author would like to thank Michael Eby and Ed Wiebe for their technical support, as well as two anonymous reviewers for their helpful and constructive comments. Katrin J. Meissner is grateful for research grant support under the University Faculty Award programm (NSERC). Figure 3. (a) Time series of simulated atmospheric D 14 C for runs A (red line) and B (green solid line) compared to the paleo record (black dots [Hughen et al., 2004a]).Also shown is the imposed freshwater perturbation in blue. The green dashed line corresponds to the green simulation shifted back by 900 years to coincide the decrease in observed atmospheric D 14 C with the resumption of the AMOC in our simulation. (b) Time series of simulated atmospheric CO 2 concentrations (runs A and B in red and green, respectively) compared to the paleo record of EPICA Dome C (black dots [Monnin et al., 2001]). The green dashed line corresponds to the green simulation shifted back by 900 years to coincide the increase in observed atmospheric CO 2 concentrations with the resumption of the AMOC in our simulation. References Adkins, J. F., and C. Pasquero (2004), Deep ocean overturning: Then and now, Science, 306, Alley, R. B., D. A. Meese, C. A. Shuman, A. J. Gow, K. C. Taylor, P. M. Grootes, J. W. C. White, M. Ram, E. D. Waddington, P. A. Mayewski, and G. A. Zielinski (1993), Abrupt increase in Greenland snow accumulation at the end of the Younger Dryas event, Nature, 362, , doi: /362527a0. Beck, J. W., et al. (2001), Extremely large variations of atmospheric 14 C concentration during the last glacial period, Science, 292, Berger, A. L. (1978), Long-term variations of daily insolation and quaternary climatic changes, J. Atmos. Sci., 35, Broecker, W. S., J. P. Kennett, B. P. Flower, J. T. Teller, S. Trumbore, G. Bonani, and W. Wolfli (1989), Routing of meltwater from the Laurentide ice-sheet during the Younger Dryas cold episode, Nature, 341, Carlson, A. E., P. U. Clark, B. A. Haley, G. P. Klinkhammer, K. Simmons, E. J. Brook, and K. J. Meissner (2007), Geochemical proxies of freshwater routing during the Younger Dryas event, Proc. Natl. Acad. Sci. U.S.A., 104, Chiu, T. C., R. G. Fairbanks, L. Cao, and R. A. Mortlock (2007), Analysis of the atmospheric 14 Crecordspanningthepast50000yearsderived from high-precision 230 Th/ 234 U/ 238 U, 231 Pa/ 235 Uand 14 Cdatesonfossil corals, Quat. Sci. Rev., 26(1 2), Clark, P. U., N. G. Pisias, T. F. Stocker, and A. J. Weaver (2002), The role of the thermohaline circulation in abrupt climate change, Nature, 415, Cox, P. M. (2001), Description of the TRIFFID dynamic global vegetation model, Hadley Centre Tech. Note, 24, 16 pp. Cox, P. M., R. A. Betts, C. B. Bunton, R. L. H. Essery, P. R. Rowntree, and J. Smith (1999), The impact of new land surface physics on the GCM simulation of climate and climate sensivity, Clim. Dyn., 15, Eltgroth, S. F., J. F. Adkins, L. F. Robinson, J. Southon, and M. Kashgarian (2006), A deep-sea coral record of North Atlantic radiocarbon through the Younger Dryas: Evidence for intermediate water/deepwater reorganization, Paleoceanography, 21, PA4207, doi: /2005pa Ewen, T. L., A. J. Weaver, and M. Eby (2004), Sensitivity of the inorganic carbon cycle to future climate warming in the UVic coupled model, Atmos. Ocean, 42, Fairbanks, R. G., R. A. Mortlock, T. C. Chiu, L. Cao, A. Kaplan, T. P. Guilderson, T. W. Fairbanks, A. L. Bloom, and M. J. Nadeau (2005), Radiocarbon calibration curve spanning 0 to years BP based on paired 230 Th/ 234 U/ 238 Uand 14 Cdatesonpristinecorals,Quat. Sci. Rev., 24, Friedrich, M., S. Remmele, B. Kromer, J. Hofmann, M. Spurk, K. F. Kaiser, C. Orcel, and M. Küppers (2004), The year Hohenheim oak and pine tree-ring chronology from central Europe A unique annual record for radiocarbon calibration and paleoenvironment reconstructions, Radiocarbon, 46, Godwin, H. (1962), Half life of radiocarbon, Nature, 195, 984. Goslar, T., et al. (1995), High concentration of atmospheric 14 Cduringthe Younger Dryas cold episode, Nature, 377, Goslar, T., B. Wohlfarth, S. Bjoerck, G. Possnert, and J. Bjoerck (1999), Variations of atmospheric 14 CconcentrationsovertheAllerod-Younger Dryas transition, Clim. Dyn., 15, Goslar, T., M. Arnold, N. Tisnerat-Laborde, J. Czernik, and K. Wieckowski (2000), Variations of Younger Dryas atmospheric radiocarbon explicable without ocean circulation changes, Nature, 403, Hibler, W. D. (1979), A dynamic thermodynamic sea ice model, J. Phys. Oceanogr., 9, Hughen, K. A., J. T. Overpeck, S. J. Lehman, M. Kashgarian, J. Southon, L. C. Peterson, R. Alley, and D. M. Sigman (1998), Deglacial changes in 4of5

5 ocean circulation from an extended radiocarbon calibration, Nature, 391, Hughen, K. A., J. R. Southon, S. J. Lehman, and J. T. Overpeck (2000), Synchronous radiocarbon and climate shifts during the last deglaciation, Science, 290, Hughen, K. A., S. J. Lehman, J. R. Southon, J. T. Overpeck, O. Marchal, and C. Herring (2004a), 14 Cactivityandglobalcarboncyclechanges over the past years, Science, 303, Hughen, K. A., J. R. Southon, C. J. H. Bertrand, B. Frantz, and P. Zermeño (2004b), Cariaco basin calibration update: Revisions to calendar and 14 C chronologies for core PL07-58PC, Radiocarbon, 46, Hunke, E. C., and J. K. Dukowitz (1997), An elastic-viscous-plastic model for sea ice dynamics, J. Phys. Oceanogr., 27(9), Johnson, R. G., and B. T. McClure (1976), A model for Northern Hemisphere continental ice sheet variation, Quat. Res., 6, Kalnay, E., et al. (1996), The NCEP/NCAR 40-year reanalysis project, Bull. Am. Meteorol. Soc., 77(3), Keigwin, L. D. (2004), Radiocarbon and stable isotope constraints on Last Glacial Maximum and Younger Dryas ventilation in the western North Atlantic, Paleoceanography, 19, PA4012, doi: /2004pa Köhler, P., H. Fischer, G. Munhoven, and R. E. Zeebe (2005), Quantitative interpretation of atmospheric carbon records over the last glacial termination, Global Biogeochem. Cycles, 19, GB4020, doi: / 2004GB Marchal, O., T. F. Stocker, and R. Muscheler (2001), Atmospheric radiocarbon during the Younger Dryas: Production, ventilation, or both?, Earth Planet. Sci. Lett., 185, Marchitto, T. M., S. J. Lehman, J. D. Ortiz, J. Flückinger, and A. van Green (2007), Marine radiocarbon evidence for the mechanism of deglacial atmospheric CO 2 rise, Science, 316, Mathewes, R. W., L. E. Heusser, and R. T. Patterson (1993), Evidence for a Younger Dryas-like cooling event on the British Columbia coast, Geology, 21, Matthews, H. D., A. J. Weaver, and K. J. Meissner (2005), Terrestrial carbon cycle dynamics under recent and future climate change, J. Clim., 18, McManus, J. F., R. Francois, J. M. Gherardi, L. D. Keigwin, and S. Brown- Leger (2004), Collapse and rapid resumption of Atlantic meridional circulation linked to deglacial climate changes, Nature, 428, Meissner, K. J., A. Schmittner, A. J. Weaver, and J. F. Adkins (2003a), Ventilation of the North Atlantic Ocean during the Last Glacial Maximum: A comparison between simulated and observed radiocarbon ages, Paleoceanography, 18(2), 1023, doi: /2002pa Meissner, K. J., A. J. Weaver, H. D. Matthews, and P. M. Cox (2003b), The role of land surface dynamics in glacial inception: A study with the UVic Earth System Model, Clim. Dyn., 21, Meissner, K. J., M. Eby, A. J. Weaver, and O. A. Saenko (2007), CO 2 threshold for millennial-scale oscillations in the climate system: Implications for global warming scenarios, Clim. Dyn., in press. Monnin, E., A. Indermühle, A. Dällenbach, J. Flückinger, B. Stauffer, T. F. Stocker, D. Raynaud, and J. M. Barnola (2001), Atmospheric CO 2 concentrations over the last glacial termination, Science, 291, Muscheler, R., J. Beer, G. Wagner, and R. C. Finkel (2000), Changes in deep-water formation during the Younger Dryas event inferred from 10 Be and 14 Crecords,Nature, 408, Muscheler, R., J. Beer, and P. W. Kubik (2004), Long-term solar variability and climate change, in Solar Variability and Its Effects on Climate, Geophys. Monogr. Ser., vol.141,editedbyj.m.papandp.fox,pp , AGU, Washington, D. C. Pacanowski, R. C. (1995), MOM 2 documentation, user s guide and reference manual, Tech. Rep. 3, GFDL Ocean Group, Geophys. Fluid Dyn. Lab., Princeton, N. J. Peltier, W. R. (1994), Ice age paleotopography, Science, 265, Robinson, L. F., J. F. Adkins, L. D. Keigwin, J. Southon, D. P. Fernandez, S. L. Wang, and D. S. Scheirer (2005), Radiocarbon variability in the western North Atlantic during the last deglaciation, Science, 310, Rooth, C. (1982), Hydrology and Ocean Circulation, Prog. Oceanogr. vol. 11, pp , Pergamon, New York. Schmittner, A., A. Oschlies, H. D. Matthews, and E. D. Galbraith (2007), Future changes in climate, ocean circulation, ecosystems and biogeochemical cycling simulated for a business-as-usual CO 2 emission scenario until year 4000 A. D., Global Biogeochem. Cycles, doi: / 2007GB002953, in press. Semtner, A. J. J. (1976), A model for the thermodynamic growth of sea ice in numerical investigations of climate, J. Phys. Oceanogr., 6(3), Stocker, T. F., and D. G. Wright (1996), Rapid changes in ocean circulation and atmospheric radiocarbon, Paleoceanography, 11(6), Stuiver, M., P. J. Reimer, E. Bard, J. W. Beck, G. S. Burr, K. A. Hughen, B. Kromer, G. McCormac, J. van der Plicht, and M. Spurk (1998), IntCal98 radiocarbon age calibration, cal BP, Radiocarbon, 40, Weaver, A. J., et al. (2001), The UVic Earth System Climate Model: Model description, climatology, and applications to past, present and future climates, Atmos. Ocean, 4, K. J. Meissner, School of Earth and Ocean Sciences, University of Victoria, PO Box 3055, Stn CSC, Victoria, BC, Canada, V8W 3P6. (katrin@ocean.seos.uvic.ca) 5of5