Camille Li
A dissertation submitted in partial fulfillment of the requirements for the degree of
Doctor of Philosophy
University of Washington
2007
Program Authorized to Offer Degree: Department of Atmospheric Sciences
This is to certify that I have examined this copy of a doctoral dissertation by Camille Li
and have found that it is complete and satisfactory in all respects, and that any and all revisions required by the final
examining committee have been made.
Chair of the Supervisory Committee:
David Battisti
Reading Committee:
David Battisti Gerard Roe John Michael Wallace
Date:
or reproduction of this dissertation may be referred to Proquest Information and Learning, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, 1-800-521-0600, or to the author.
Signature
Date
A general circulation modelling perspective on abrupt climate change during glacial times
Camille Li
Chair of the Supervisory Committee:
Professor David Battisti Department of Atmospheric Sciences
The climate of the Last Glacial Maximum (LGM; 21 ka before present) is investigated using simulations from coupled global climate models and general circulation models. Of particular in- terest is the atmospheric circulation in the Northern Hemisphere Atlantic sector, a region known to have experienced large, rapid climate changes called Dansgaard-Oeschger (D-O) events during the last ice age. This thesis aims to develop a better understanding of these rapid climate changes by examining the mean climate state in which they occur. To this end, model simulations are used to characterize the large-scale atmospheric flow, and the sensitivity of this flow to plausible climate perturbations such as sea ice displacements, changes in sea surface temperatures and the growth of land-based ice sheets.
The Last Glacial Maximum as simulated by the Community Climate System coupled model (CCSM3) is shown to exhibit strong, steady jets and weak transient eddies compared to today’s climate. Uncoupled experiments performed with an atmosphere model suggest that the presence of the Laurentide ice sheet complex over North America is responsible for establishing the steady jet.
It is demonstrated that with the ice sheet at its maximum size, it is difficult to alter the jet, even with large perturbations to sea surface temperature and sea ice patterns. However, with the ice sheet at some intermediate size, it is possible to produce a substantial climate response in the North Atlantic region by perturbing the sea ice edge. In fact, experiments show that the climate signal arising from a retreat of sea ice is comparable to the observed changes in temperature and accumulation recorded
List of Tables . . . v
Chapter 1: Introduction . . . 1
Chapter 2: Abrupt climate shifts in Greenland due to displacements of the sea ice edge 5 2.1 Introduction . . . 5
2.2 Experimental Design . . . 7
2.3 Results . . . 9
2.4 Discussion . . . 12
Chapter 3: A simulation of Northern Hemisphere atmospheric circulation during Last Glacial Maximum . . . 14
3.1 Introduction . . . 14
3.2 Model description and methods . . . 17
3.3 Last Glacial Maximum climate in a coupled model (CCSM3) . . . 17
3.4 Comparison of LGM jets and storminess in CCSM3 with PMIP-1 simulations . . . 29
3.5 The influence of land ice, sea surface temperature and sea ice . . . 36
3.6 Conclusions . . . 43
Chapter 4: Suppression of eddy activity in glacial times . . . 44
4.1 Eddy structure . . . 44
4.2 Mechanisms for suppression of eddy activity . . . 48
Chapter 5: Evaluation of the model simulation against paleoclimate observations . . . 51
5.1 Land ice . . . 51
5.2 Sea surface temperature and sea ice . . . 52
Chapter 6: Concluding remarks . . . 56
References . . . 60 i
Appendix C: Eddy diagnostics in observations . . . 76 Appendix D: Comparison of coupled and uncoupled simulations . . . 80
ii
2.1 Seasonal cycle of North Atlantic sea ice extent (10 km
) in the LGM control ex-
periment and the two sea ice retreat scenarios . . . 7
2.2 Comparison of LGM and reduced sea ice scenario (I) . . . 8
2.3 Seasonal cycle of temperature (top), monthly accumulation (middle) and monthly accumulation as a percentage of total annual accumulation (bottom) at the Green- land Summit . . . 11
2.4 Implied ocean heat transport (OHT) in the Atlantic Ocean for the present day, LGM control and reduced sea ice (scenario I) simulations . . . 13
3.1 Wintertime atmospheric circulation in CCSM3 simulations . . . 19
3.2 Atlantic jet and eddy characteristics in CCSM3 (coupled) and CAM3 (uncoupled) simulations of PD and LGM climate . . . 20
3.3 Meridional heat transport from observations and simulations . . . 22
3.4 Total atmospheric heat transport and heat transport by stationary waves in CCSM3 simulations . . . 22
3.5 Wintertime jet position and eddy diagnostics in CCSM3 simulations . . . 24
3.6 Wintertime zonal wind and temperature profiles by ocean sector . . . 26
3.7 Atlantic jet and eddy characteristics from reanalysis data . . . 28
3.8 Wintertime baroclinicity in CCSM3 simulations . . . 29
3.9 Wintertime jet position in simulations of present day and Last Glacial Maximum climate from PMIP-1 . . . 31
3.10 Wintertime jet position and eddy diagnostics in CAM3 baseline experiments . . . . 32
3.11 Reconstructions of land ice topography over North America at Last Glacial Maximum 34 3.12 Distribution of Atlantic jet speeds in CAM3 perturbation experiments . . . 37
3.13 Atlantic jet characteristics in CAM3 perturbation experiments . . . 40
3.14 Atlantic eddy characteristics in CAM3 perturbation experiments . . . 40
3.15 February baroclinicity and sea ice in CAM3 simulations . . . 42
4.1 1-point regression analysis of wintertime eddy structure in CCSM3 simulations . . 46
4.2 Lag regression analysis of wintertime upper level eddy structure in CCSM3 simula- tions . . . 47
iii
tion and as simulated by CCSM3 . . . 53
5.2 Sea surface conditions in the North Atlantic during Last Glacial Maximum from reconstructions (CLIMAP, GLAMAP, Meland et al. 2005) and as simulated by CCSM3 . . . 55
6.1 Global ice volume and D-O events during the last glaciation . . . 57
A.1 Determining an adequate sample size for stable eddy statistics . . . 70
B.1 Pacific jet and eddy characteristics from reanalysis data . . . 72
B.2 Pacific jet and eddy characteristics in CCSM3 (coupled) and CAM3 (uncoupled) simulations of PD and LGM climate . . . 73
B.3 Pacific jet characteristics in CAM3 perturbation experiments . . . 75
B.4 Pacific eddy characteristics in CAM3 perturbation experiments . . . 75
C.1 Wintertime jet position and eddy diagnostics from NCEP reanalysis data . . . 77
C.2 Summertime jet position and eddy diagnostics in CCSM3 simulations . . . 78
C.3 Summertime jet position and eddy diagnostics from NCEP reanalysis data . . . 79
D.1 Atlantic jet and eddy characteristics in CCSM3 (coupled) and CAM3 (uncoupled) simulations of PD and LGM climate . . . 81
iv
2.1 Temperature at 2 m reference height, accumulation and accumulation-weighted tem- perature (weighted T) at Greenland Summit in CCM3 simulations . . . 9 3.1 Jet and eddy characteristics in the Atlantic sector (15 N–65 N, 90 W–0 ) for DJFM
winter . . . 27 3.2 Uncoupled models shown in Figure 3.9 . . . 33 3.3 Boundary conditions for CAM3 perturbation experiments . . . 38 B.1 Jet and eddy characteristics in the Pacific sector (15 N–65 N, 100 E–125 W) for
DJFM winter . . . 74
v
I owe countless thanks to countless people. First, to my committee: my advisor, David Battisti, for his friendship, guidance, and the constant supply of caffeinated and literary sus- tenance; and my committee members, Mike Wallace, Gerard Roe, Cecilia Bitz and Eric Steig, for their help in shaping and improving this work, and for their Herculean effort to get all this done in time. Beyond my committee, there are many others who have gener- ously shared their experience and expertise, including Joe Barsugli, Greg Hakim, Dennis Hartmann, Nathan Kutz, Marc Michelson, Dan Schrag, Richard Seager and Eli Tziperman.
For sympathy, support and adventures, I have been lucky to know a lot of good eggs:
the Class of 2000 (and 1999.5); denizens of the Climate Palace, past and present; the ATG staff; various and assorted officemates in various and assorted cities; the fourth floor tomato gardening hotline; friends across the fountain; and fellow punters on the river Livet.
To my family, and especially 54 Leacock, my #1 Brother et le Berger des Aubergines:
You have watched with concern, amusement, and at times, the abject horror of those caught in a Hard-Boiled Wonderland origami project gone monstrously awry. Thank you for your unflagging encouragement, despite not understanding precisely what I was doing, why I was doing it, and how it could possibly take so long.
Finally, I’d like to acknowledge the Comer Abrupt Climate Change Fellowship for fi- nancial support and for the opportunity to work with so many fun and inspiring people.
vi
CHAPTER1
INTRODUCTION
The massive ice sheets that blanket Greenland and Antarctica hold an isotopic record of past climate in their layers of compressed snow. The excavation of these layers provides an invaluable glimpse into Earth’s history. By analyzing the bubbles of air trapped in the ice and the isotopic composition of the ice itself, we are able to uncover fascinating, and often puzzling, stories of ice ages and warm periods, of the Sun’s influence on global climate, and of the link between atmospheric CO and glacial cycles.
One of the most intriguing stories is that of Dansgaard-Oeschger (D-O) events, the abrupt, millennial-scale warmings that occurred intermittently throughout much of the last glacial period.
D-O events were first discovered as excursions in the O of Greenland ice, and their signature has since been identified at geographic locations far removed from Greenland (see Rahmstorf , 2002;
V¨olker, 2002, and references therein). For example, D-O events have been observed with varying degrees of confidence in paleoclimate archives such as marine sediment cores in the extratropical and tropical Atlantic (Bond et al., 1993; Sachs and Lehman, 1999; Peterson et al., 2000), extratrop- ical Pacific (Behl and Kennett, 1996) and Arabian Sea (Schulz et al., 1998; Altabet et al., 2002) and speleothems from caves in China (Wang et al., 2001) and Brazil (Wang et al., 2004).
The proxy variables measured in these paleoclimate archives can be related to climate variables to reveal the response of the climate system during D-O events, but the paleoclimate records are tricky to interpret. In many cases, we have an imperfect understanding of the complex proxy-climate relationships. Not only is it often unclear how the proxy variables relate to climate variables, it is difficult to determine the spatial scales (local, regional, hemispheric, or global) that the proxies represent. In addition, uncertainties involved in dating, particularly for marine cores which tend to have lower temporal resolution, make synchronization of events from different records problematic.
Despite these problems, the records together tell a rich and compelling story. D-O interstadials were warm periods in the mid- to high-latitude Atlantic as inferred from reconstructions of surface temperatures (Grootes and Stuiver, 1997; Severinghaus et al., 2003) and sea surface temperatures (Bond et al., 1993; Sachs and Lehman, 1999). Greenland experienced, on average, a 50–100%
increase in snowfall (Dahl-Jensen et al., 1993) and 10–12 C annual mean warming (Severinghaus et al., 2003; Masson-Delmotte et al., 2005) with each event. The warming seems to have been more pronounced in winter than summer, leading to a reduced seasonal cycle (see Denton et al., 2005, and references therein). Evidence of abrupt climate changes away from the North Atlantic attests that D-O events were widespread. In the Cariaco Basin off the coast of Venezuela, rapid changes in tropical Atlantic marine sediments have been interpreted as north-south shifts of the ITCZ (Peterson et al., 2000) which seem to correspond to alternating wet and dry periods in Brazil (Wang et al., 2004). Corroborating evidence that the tropics were involved in D-O events comes from methane measurements (Brook et al., 1996), primarily a tropical terrestrial signal, recorded in ice cores. Even more removed geographically, Chinese speleothems (Wang et al., 2001) and Arabian Sea sediments (Schulz et al., 1998; Altabet et al., 2002) suggest that the Asian and Indian monsoons were stronger during interstadials.
Despite a concerted and multidisciplinary research effort over the last decades, the precise mech- anism behind D-O events remains elusive. The leading hypothesis attributes D-O events to an in- ternal oscillation of the ocean thermohaline circulation (OTC) (Broecker et al., 1985) which con- sequently causes abrupt shifts in ocean heat transport (OHT) into the North Atlantic and in surface temperature in Greenland. Support for this theory comes from McManus et al. (2004), who showed a link between deep ocean circulation and the Greenland record for the Bølling-Allerød/Younger Dryas events. However, paleoclimate proxy data are currently inadequate to provide constraints that conclusively support or invalidate this hypothesis.
Models offer an additional tool to address mechanisms of abrupt climate change, although there are several serious challenges involved in simulating D-O events. First, the base climate of the Last Glacial Maximum is not well known, meaning that the initial conditions for a D-O event are not well known. Second, any hypothesis invoking internal oscillations of a complex, dynamic system does not present a straightforward mechanism to test cleanly with a model. Finally, the time scale of D-O events, on the order of thousands of years, makes simulations with general circulation models
One approach is to use simplified models that are computationally efficient compared to GCMs.
The existence of self-sustained “deep decoupling” oscillations of the OTC has been demonstrated in a simplified 3-D ocean model (Winton and Sarachik, 1993). More recently, small freshwater forcings have been used to trigger shifts in intermediate complexity climate models – simplified but fully coupled atmosphere-ocean models – which also exhibit multiple stable modes of OTC opera- tion (Ganopolski and Rahmstorf , 2001; Knutti et al., 2004). We note that these models are able to capture many temporal aspects of the paleoclimate records in Greenland, but that their “intermedi- ate complexity” renders certain aspects of their behaviour suspect. In general, they comprise highly simplified atmospheres (coarse quasigeostrophic or some sort of energy balance model) and low resolution oceans with unsophisticated representations of the processes responsible for controlling ocean circulation. It is unclear whether the OTC in these models would display regime shifts if the relevant ocean processes were better represented, or if the ocean models were coupled to more realistic atmosphere models.
In this work, we adopt an alternate approach using atmospheric general circulation models. The goal is to gain insight into the climate of the Last Glacial Maximum in order to better understand the possibilities of abrupt climate events within the context of this base state. We thus explore the climate response to perturbations in the North Atlantic region, where the D-O warmings had their largest impact.
In Chapter 2, we present a sensitivity study of sea ice, one of the most important climate ampli- fiers in polar regions, and its effect on atmospheric circulation in the North Atlantic. We demonstrate that sudden displacements in the sea ice edge produce a climate response that is consistent with the signal of D-O events recorded in ice cores at the Greenland Summit. What remains to be shown is why the sea ice might abruptly change. It could be due to abrupt shifts in OTC (discussed above) or abrupt rearrangements in the atmospheric circulation that affect sea ice via momentum fluxes or forced changes in ocean circulation and ocean heat transport.
In Chapter 3, we explore features of the atmospheric circulation in a coupled simulation of the Last Glacial Maximum. We document changes in jets and in transient activity in the North Atlantic sector during the winter season from the present day to the LGM, and observe that the latter is a world with strong, stable jets and reduced storminess. We note that the LGM as simulated
by the coupled model bears some striking differences to the LGM as simulated by the uncoupled atmosphere models from the first phase of the Paleoclimate Modelling Intercomparison Project (PMIP-1; http://www-lsce.cea.fr/pmip). A series of experiments performed using the atmospheric component of the coupled model points to three key factors that are responsible for the differences:
the size of the Laurentide ice sheet complex over North America, and the sea surface temperature and sea ice conditions in the Northern Hemisphere.
In Chapter 4, we examine the structure of transient disturbances in the LGM climate using regression analysis. These results illustrate that transient eddies are more meridionally confined and more coherent in glacial times compared to the present day. We discuss some mechanisms that may act to suppress eddy activity despite the favourable conditions for eddy growth during LGM.
In Chapter 5, we take a closer look at the land ice topography and state of the surface ocean in the coupled model simulation of the LGM. We evaluate the model results against observational records of past climate and conclude that the coupled model produces a more realistic LGM simulation than the models that participated in PMIP-1.
Finally, in Chapter 6, we discuss the implications of these results for understanding abrupt cli- mate change, and in particular, the D-O events of the last ice age.
CHAPTER2
ABRUPT CLIMATE SHIFTS IN GREENLAND DUE TO DISPLACEMENTS OF THE SEA ICE EDGE
Dansgaard-Oeschger (D-O) events are millennial-scale climate fluctuations that occurred during the last ice age. In Greenland, where they were first discovered in ice cores, they manifest themselves as a rapid (several decades) warming episode followed by period of more gradual cooling. In this chapter, an atmospheric circulation model is used to show that a reduction in sea ice extent in the North Atlantic produces a climatic response consistent with the abrupt changes in temperature and snow accumulation recorded in Greenland during D-O events. The model simulations exhibit warming that is especially pronounced in winter and an accumulation increase that is primarily due to the summer contribution, though the fractional accumulation increase is greater in winter. The seasonality of the combined temperature and accumulation changes is consistent with the oxygen isotope signal, and explains why estimates of the magnitude of D-O warming from oxygen isotopes are damped relative to direct estimates. Mechanisms for driving such displacements of sea ice could be changes in ocean thermohaline circulation (OTC) or rearrangements of the tropical atmosphere- ocean system.
2.1 Introduction
The Dansgaard-Oeschger (D-O) events that punctuated the last glacial period (110–10 kyr BP) are abrupt warming episodes recorded in Greenland ice cores (Dansgaard et al., 1993), characterized by a large annual mean temperature rise (typically 5–10 C (Ganopolski and Rahmstorf , 2001; Sev- eringhaus et al., 2003)) over several decades, with the warm conditions lasting for 200–600 years, followed by a more gradual cooling and sometimes by an abrupt return to the glacial state. Recent studies have found that the warming in Greenland is coincident with abrupt changes in other parts of the Northern Hemisphere and global tropics (Rahmstorf (2002) and references therein). A lead- ing hypothesis attributes D-O events to an internal oscillation of the ocean thermohaline circulation
(OTC) (Broecker et al., 1990). By switching the OTC from its sluggish glacial mode to one that fea- tures an increase in ocean heat transport (OHT) into the North Atlantic, this oscillation could cause an abrupt warming in Greenland (Winton, 1993; Ganopolski and Rahmstorf , 2001). The climate models used to illustrate this theory all contain simplified atmospheric dynamics that do not accu- rately simulate temperature, precipitation and wind anomalies away from regions where the SST and sea ice changes occur. Theory and more comprehensive climate models indicate that large changes in mid- and high latitude sea surface temperature (SST) alone resulting from shifts in the OTC pro- duce a temperature response near Greenland that is weak compared to the D-O jumps, regardless of whether they occur in present day or glacial boundary conditions (Fawcett et al., 1997; Seager et al., 2002; Vellinga and Wood, 2002). If OTC changes are indeed responsible for D-O events, a critical question is how such a modest climate forcing can be reconciled with the large signals recorded in Greenland ice cores.
Sea ice is a key component of the climate system that can affect it through a variety of feedbacks.
In addition to the well-known ice-albedo feedback, sea ice has an important influence on regional air temperature in winter by insulating the atmosphere from the substantial heat capacity of the ocean.
When sea ice is absent, the ocean absorbs heat in summer and releases it back to the atmosphere in winter, thereby moderating the extreme cold of the polar night in high latitudes. The existence of such feedbacks points to the possibility of rapid changes in sea ice cover resulting from relatively weak forcing (Maykut and Untersteiner, 1971). Several studies have proposed displacements of the sea ice edge as a mechanism for D-O events (Dansgaard et al., 1989; Alley et al., 1993; Broecker, 2000; Denton et al., 2005) with the suggestion of rapid, switch-like changes in sea ice cover caused by either subtle shifts in wind stress near the ice edge, or by small OTC variability (Gildor and Tziperman, 2003; Kaspi et al., 2004).
We demonstrate through atmospheric general circulation model (AGCM) experiments that re- moval of sea ice in the North Atlantic produces a response large enough to account for the magnitude of the D-O warming signal observed in Greenland. Furthermore, these sea ice changes are consistent with the snow accumulation and oxygen isotope records from the Greenland ice cores. While the sea ice studies described above use dynamical oceans to capture the timescale and timing pertinent to the OTC mechanism for abrupt climate events, the approach in our sensitivity study is to impose sea surface conditions upon a simulated atmosphere and to ask how large an effect they will have in
representation of atmospheric dynamics allows for more credible estimates of what Greenland will experience given these displacements in North Atlantic sea ice.
2.2 Experimental Design
We use the National Center for Atmospheric Research (NCAR) Community Climate Model (CCM3) AGCM to simulate scenarios with reduced North Atlantic sea ice cover and warmer North Atlantic SST relative to a Last Glacial Maximum (LGM) climate configured with Peltier ice sheets (Peltier, 1994), revised CLIMAP SST (Crowley, 2000) and sea ice (Sarnthein et al., 2003a), 21 kyr BP insolation, 200 ppm CO and 350 ppb CH . Spectral smoothing lowers the LGM Greenland Summit by 900 geopotential meters. Extrapolating to the true summit elevation using the relatively linear
“land surface lapse rate” of Greenland for each month of the year yields a more concordant (colder) annual mean LGM Summit temperature of -50 C. Assuming that snowfall at the Summit scales with saturation vapour pressure, the associated error in accumulation between the spectral and true Summit elevations is less than 10%.
In a suite of reduced sea ice experiments, the atmospheric response in the vicinity of Greenland is relatively insensitive to the details of the prescribed sea ice changes. We will focus our discussion on results from two of these experiments in which sea ice cover is modified as shown in Figure 2.1:
Scenario I has maximum sea ice extent corresponding to the LGM perennial ice line; min- imum sea ice extent corresponding to the present day perennial ice line; and, for the in-
month area (106 km2 )
J F M A M J J A S O N D
0 5 10 15
LGM I II
Figure 2.1: Seasonal cycle of North Atlantic sea ice extent (10 km
) in the LGM control experiment and the two sea ice retreat scenarios.
tervening months, sea ice distributions interpolated on a sinusoid between these extremes.
Qualitatively, the scenario exhibits significant winter ice retreat and moderate summer ice retreat in the North Atlantic. The sea ice reduction, while somewhat arbitrary, is compa- rable to the sea ice variability associated with small (2 Sv) changes in Atlantic overturning in an LGM climate simulated by NCAR’s fully coupled Community Climate System Model CCSM3 (Otto-Bliesner et al., 2006).
Scenario II has sea ice extent identical to LGM except during the extended summer season from May to September, when sea ice retreats to 80 N around Greenland as it does in the present day climate. Although this scenario is unrealistic, it provides important clues about the seasonality of temperature and accumulation changes in response to displacements of the sea ice edge.
Maps of the prescribed SST and sea ice boundary conditions for the LGM and scenario I appear in Figure 2.2a. All quoted temperature and accumulation changes correspond to at least several standard deviations of the internal variability in the reduced sea ice scenarios, and are significant at the 95% confidence level.
2.3 Results
Reduced sea ice scenario I (see Table 2.1) shows warming in the North Atlantic region with an an- nual mean temperature change of 7 C around the Greenland Summit (Fig. 2.2b). This warming is comparable to the 5–10 C temperature rise during D-O events determined from gas fractionation in air bubbles trapped within Greenland ice (Severinghaus and Brook, 1999; Severinghaus et al., 2003).
The temperature response in scenario I is localized because the prescribed SST model limits tele- connections and lacks tropical atmosphere-ocean feedbacks demonstrated to be important in glacial climates (Chiang et al., 2003). These linkages, in conjunction with potentially enhanced glacial teleconnections (Yin and Battisti, 2001), could both reinforce the warm temperatures in Greenland and produce far-field responses throughout the Northern Hemisphere.
Looking beyond the annual mean picture, results from scenario I and an additional simulation with sea ice retreat to 80 N around Greenland in summer alone (scenario II) point to winter as
Figure 2.2: Comparison of LGM and reduced sea ice scenario (I). (a) Annual mean sea surface temperature boundary conditions (degrees Celsius) for LGM (left) and reduced sea ice scenario (right). Maximum (February) and minimum (August) sea ice extents are indicated with the solid and dotted lines, respectively. Scenario I has a maximum sea ice extent equivalent to the LGM perennial ice cover, and a minimum sea ice extent equivalent to the present day perennial ice cover.
The ice thickness is 2 metres, which is a typical value for the Arctic today. (b) The difference in surface air temperature between the two simulations (degrees Celsius).
important for generating Greenland warmth (see Table 2.1). Less sea ice in winter reduces the areal extent of insulation over the relatively warm ocean, which is a significant source of heat to the atmosphere during the polar night. In scenario I, winter accounts for 2 C of warming in the 7 C annual signal. Summer changes may also play a role as less sea ice provides a darker surface for absorbing incoming solar radiation. However, the summer changes in scenario I contribute only 1 C to the annual signal; even a large reduction in sea ice during summer months such as in scenario II produces only a 2 C warming in the annual mean.
We can delve further into the issue of seasonality by considering constraints provided by snow accumulation and
O in ice cores. These records indicate that warming during D-O events was accompanied by a 50–100% increase in accumulation (Dahl-Jensen et al., 1993; Cuffey and Clow, 1997) and a 3–4 per mil increase in
O (Grootes and Stuiver, 1997). Table 2.1 and Figure 2.3 show temperature, accumulation and accumulation-weighted temperature from our model simula-
tions broken down as monthly or seasonal contributions to the annual signal. Scenario I shows a doubling of total annual accumulation, which is within the limits of the observational range. Al- though the distribution of accumulation shifts slightly towards winter, most of the actual increase occurs in summer. The sea ice changes in scenario II, which occur only in summer, also produce a doubling of accumulation but with little effect on Greenland temperature. Additional experiments (C. Li et al., unpublished data) suggest that further increases in summer SST produce much more snowfall than observed. Thus, if sea ice is involved, D-O events must include neither extreme sea ice retreat nor excessive ( 2.5 C) North Atlantic warming in summer. These results support observation-based evidence that D-O events are primarily a winter phenomenon (Denton et al., 2005, and references therein).
The
O record has traditionally been used as a measure of air temperature (Dansgaard, 1964), but other factors also influence the isotope signal in Greenland (Cuffey et al., 1994; Severinghaus and Brook, 1999). The complications are related in large part to the fact that the isotope signal is recorded by, and hence dependent on, accumulation. Models generally agree that during glacial winters, extensive sea ice cover inhibits evaporation over the North Atlantic and very zonal at-
Table 2.1: Temperature at 2 m reference height, accumulation and accumulation-weighted temper- ature (weighted T) at Greenland Summit in CCM3 simulations. The simulations shown are PD (present day), LGM, I (reduced sea ice) and II (reduced sea ice in summer only, with the ice line retreating to 80 N around Greenland). For temperature and accumulation, winter (DJF) and sum- mer (JJA) breakdowns are included in addition to the annual averages. The accumulation values in parentheses are the fraction of the total annual accumulation contributed by the given season. The columns marked T, acc and wT show the annual mean difference relative to the LGM. These results are an average over 70 N–75 N and 34 W–48 W near the Greenland Summit. Temperature and accumulation changes quoted for the reduced sea ice scenario correspond to at least several standard deviations of its internal variability and are significant at the 95% confidence level.
Exp 2m temperature (C) accumulation (cm/y) weighted T (C) DJF JJA ANN T DJF (%) JJA (%) ANN acc (%) ANN wT PD -33 -8 -22 20.5 (22) 34.2 (36) 23.6 -18.6
LGM -63 -24 -45 0.6 (4) 11.0 (70) 4.0 -28.3
I -55 -20 -38 7 1.2 (4) 20.1 (62) 8.0 +100 -25.5 2.5 II -63 -19 -43 2 0.6 (2) 25.5 (79) 8.1 +100 -22.4 5.8
T (C)
−80
−60
−40
−20
acc (cm/y)
0 10 20 30
month
acc (%)
J F M A M J J A S O N D
0 10 20
PD LGM I II
Figure 2.3: Seasonal cycle of temperature (top), monthly accumulation (middle) and monthly accu- mulation as a percentage of total annual accumulation (bottom) at the Greenland Summit. The Greenland Summit is taken to be 70N–75N and 34W–48W; all quantities shown are an area- weighted average over this box.
mospheric circulation prevents cyclones from advecting moisture over Greenland (Fawcett et al., 1997; Krinner and Genthon, 1997; Werner et al., 2000). As a consequence, Greenland sees very little snowfall in winter compared to summer, a result that is supported by evidence from close-off porosity measurements (Raynaud et al., 1997). Hence, there is a pronounced seasonality in accu- mulation rates during glacial times that introduces a warm (summer) bias to the ice core record (Steig et al., 1994). Although some have argued for the importance of other factors in determining the isotopic composition of Greenland snowfall (Charles et al., 1994; Boyle, 1997), Werner et al.
(2000) showed that snow seasonality is the dominant control for the glacial/interglacial transition.
Indeed, the differences in annual mean temperature ( T = 23 C) and accumulation-weighted tem- perature ( T = 9.7 C) between our LGM and control experiments reproduce the disparity in the glacial/interglacial temperature transition from borehole and O measurements.
If D-O events also involve a shift in snow seasonality, then we can calculate an accumulation-
weighted temperature change associated with the warming that should correspond to the muted (5–7 C)
O paleothermometer estimate rather than the more pronounced (5–10 C) gas fractiona- tion estimates (Severinghaus et al., 2003). Whereas the latter records a true temperature response, the former feels two competing effects during a retreat of sea ice: (1) the actual warming enriches the
O signal in snowfall (heavier or warmer values), but (2) the increased contribution from the winter season to the annual snowfall depletes the
O signal (lighter or colder values).
From Table 2.1, we see that the sea ice retreat in scenario I does indeed produce a damped “iso- tope” (accumulation-weighted temperature) signal ( wT = 2.5 C) compared to the actual annual mean temperature signal ( T = 7 C) in the model. In fact, the accumulation-weighted temperature after the retreat is too depleted (cold) in the annual mean. Winter is so dry in the LGM base state that even a small increase in snowfall shifts the seasonality of accumulation a significant amount.
The greater the shift, the greater the depleting effect offsetting the enrichment of the signal by the warming. Thus, despite its tendency to weaken the D-O isotope signal, a change in the seasonality of precipitation alone may not explain entirely the
O record. Ongoing work aims to ascertain the importance of additional effects such as changes in the source regions or transport of Greenland- bound water vapour, both of which conceivably enrich the isotope signal when there is less sea ice (Charles et al., 1994; Werner et al., 2000).
2.4 Discussion
This study has shown that changes in winter sea ice extent in the North Atlantic are consistent with observed signals in temperature, accumulation and
O in Greenland during D-O events. Although rapid displacements of the sea ice edge may be responsible for the abrupt warming signal, they must be driven by other parts of the climate system. The driver could be local (OTC changes in the North Atlantic (Ganopolski and Rahmstorf , 2001; Kaspi et al., 2004; Knutti et al., 2004)) or reside in more distal regions such as the tropics (Yin and Battisti, 2001).
If ocean-driven, a successful hypothesis must explain a strengthening of the ocean thermohaline circulation (OTC) in the North Atlantic that initiates abruptly and persists for several centuries be- fore returning to a weaker mode. Modelling studies have identified such OTC oscillations (Winton, 1993; Ganopolski and Rahmstorf , 2001), but the use of idealized boundary conditions (bathymetry, basin geometry) and coarse grids are known to be problematic for accurate simulation of ocean cir-
−30 0 30 60 90
−1 0 1
Atlantic OHT (PW)
latitude
modern LGM I
Figure 2.4: Implied ocean heat transport (OHT) in the Atlantic Ocean for the present day, LGM control and reduced sea ice (scenario I) simulations. OHT was calculated by integrating the surface heat imbalance southwards from the North Pole.
culation. Furthermore, model experiments indicate that in the present day climate, changes in OTC lead to short-lived (on the order of decades) changes in sea ice extent (Vellinga et al., 2002) which induce a restoring atmospheric response (Deser et al., 2004). This ocean-driven hypothesis instead requires positive feedbacks to sea ice changes from the overlying atmosphere in the LGM to account for the observed duration of warming after a D-O event.
Although our model does not include an interactive ocean, the surface heat budget can provide indirect evidence of how the ocean is affected by a poleward displacement of the sea ice edge. By making the assumption that any imbalance in the budget must be accounted for by ocean circulation, and then integrating the imbalance from the North Pole to a given latitude, we estimate the amount of heat transported by the ocean to sustain the specified sea ice and SST distributions (Fig. 2.4). This calculation shows a very small change in implied ocean heat transport (OHT) between the LGM and reduced sea ice (I) scenarios, with a difference equivalent to less than several Sverdrups of overturning using present day scaling (Ganachaud and Wunsch, 2000). We estimate a 15% increase in OHT at 40 N to sustain the reduced sea ice distribution.
Because sea ice is sensitive to forcing from the atmosphere as well as the ocean (Bitz et al., 2005), an alternative driver for sea ice retreat is a change in surface wind stress in the North Atlantic (Fang and Wallace, 1994). The surface wind anomalies could themselves have non-local origins such as the interaction of atmospheric circulation with land-based ice sheets or reorganizations of the tropical atmosphere-ocean system (Yin and Battisti, 2001). There are no indications that such
phenomena occur or can occur in today’s climate, but we are interested in whether they occurred in the LGM climate. We aim to address this issue by examining the large-scale atmospheric flow during glacial times to better understand its mean state and how this mean state may be perturbed.
In the following chapters, we use model simulations to probe the workings of LGM climate in an effort to gain insight into the plausibility of abrupt wind shifts.
CHAPTER3
A SIMULATION OF NORTHERN HEMISPHERE ATMOSPHERIC CIRCULATION DURING LAST GLACIAL MAXIMUM
The Last Glacial Maximum, 21 thousand years before present, was a time of maximum land ice extent in the last ice age. Previous studies using uncoupled atmospheric model simulations suggest that, during LGM, atmospheric circulation in the Atlantic sector was similar to the present day in terms of jets and eddy activity, with the main difference being a northeastward extension in the geographical location of the maxima in these fields. A recent climate simulation of the Last Glacial Maximum performed using a fully coupled atmosphere-ocean-sea ice-land surface climate model produces a very different Atlantic circulation. In this simulation, we observe a strong, steady Atlantic jet with weak eddy activity at low levels and aloft.
We investigate the nature and origin of the differences between the uncoupled and coupled climate simulations of LGM climate. The uncoupled models examined here were forced with a prescribed set of land ice and SST/sea ice boundary conditions; the coupled model was forced with improved land ice boundary conditions, and has interactive ocean and sea ice components that create their own SST/sea ice fields. A series of uncoupled perturbation experiments identifies the improved land ice as the important factor in producing the strong, steady jet in the coupled model simulation, while the combination of improving the land ice boundary condition and prescribing non-interactive sea surface conditions is responsible for suppressing eddy activity. The existence of an altered at- mospheric circulation state during LGM has implications for our understanding of the stability of glacial climates, for the possibility of multiple atmospheric regimes, and for the interpretation of the observational record of abrupt climate change during glacial times.
3.1 Introduction
The Last Glacial Maximum was a cold period approximately 21 thousand years before present (21 ka BP), when massive ice sheets covered much of the Northern Hemisphere continents. Extreme
climates such as this are useful case studies that allow us to test ideas and intuition about how the climate system might react when subjected to large perturbations. The Last Glacial Maximum (LGM) is especially suitable for these purposes because we have relatively good constraints on the forcings (land ice, insolation and atmospheric greenhouse gases), and, thanks to a host of geological and geochemical proxy data, on surface climate variables such as temperature and precipitation.
There are, however, limits to what proxy records can reveal about the actual workings of the climate system. Many factors can obfuscate the translation of a proxy variable into a physically meaningful variable. For example, while it is possible to deduce ocean circulation from ice rafted debris (Lassen et al., 1999) or the ratios of radioactive tracers (McManus et al., 2004) in marine sediment cores, one must be aware that a whole series of assumptions goes into that deduction.
And even for proxy variables that are well understood or easily translatable into climate signals, the spatial sampling is usually highly irregular and prohibitively sparse.
Climate models have proved to be increasingly useful as a tool for tackling some questions that cannot be answered using proxy data alone. In other words, “climate stories” needn’t be based ex- clusively on proxy data; a story might be tested using numerical simulations, and then corroborated by comparing the simulation results with proxy data. The question of particular interest here per- tains to the large-scale circulation of the atmosphere during glacial times, and how it compares to what is observed today. The existence of a different atmospheric circulation regime during LGM has implications for our understanding of global heat transport and the stability of glacial climates, as well as for the interpretation of the proxy records themselves.
In the 1990s, the Paleoclimate Model Intercomparison Project was undertaken to evaluate past climates using a collection of climate models. The first phase of the project (PMIP-1) comprised, for the most part, uncoupled atmospheric general circulation models, and furnishes a suite of simu- lations of climate during Last Glacial Maximum. There have since been efforts to examine changes in atmospheric circulation, transient activity and storm tracks in these PMIP-1 simulations. For ex- ample, the U.K. Universities’ Global Atmospheric Modelling Project (UGAMP) model’s simulation of the Last Glacial Maximum exhibits a poleward-shifted Atlantic jet with a pronounced SW-NE tilt; and enhanced low level baroclinicity accompanied by an increase in eddy activity at low lev- els but a decrease in eddy activity aloft (Hall et al., 1996). Further analysis of the storm tracks in this simulation reveals stronger but shallower synoptic waves over the Atlantic, i.e., increased low-
modes largely account for changes in the position and dominant wavenumber of the storms, but not the amplitude of the actual storm tracks (Kageyama and Valdes, 2000).
Kageyama et al. (1999) examined all the European models in PMIP-1 for systematic changes in storm tracks during glacial times. They report a northeastward extension of both Northern Hemi- sphere storm tracks and an elongation of the Atlantic storm track in six of seven cases. They note
“no systematic increase or decrease in the storminess from the present climate to the last glacial maximum one”, but the figures in their paper seem to indicate an increase in peak low level tran- sient eddy activity for most of the models (see their Figures 1–2, 6–7).
Subsequent to PMIP-1, LGM simulations have been performed using atmosphere models cou- pled to slab oceans (Dong and Valdes, 1998) or using intermediate complexity models (Justino et al., 2005). In both cases, the same ICE-4G (Peltier, 1994) land ice configuration used in the PMIP-1 simulations (discussed above) was employed. These coupled simulations have also shown evidence for increased storminess at low to middle levels of the atmosphere in the Atlantic sector, but there is no general consensus on the changes in the Pacific sector. More recently, models with dynamical oceans have been used to address the importance of ocean circulation changes (mainly Pacific and tropical) in determining the state of the atmosphere during glacial times (Bush and Philander, 1998;
Timmermann et al., 2004). However, most of the literature documenting the fully coupled LGM simulations, such as those involved in the second phase of PMIP, does not comment on the charac- ter of storms or storm tracks (Kitoh and Murakami, 2001; Hewitt et al., 2003; Kim et al., 2003; Shin et al., 2003; Otto-Bliesner et al., 2006).
In this study, we examine the atmospheric circulation at Last Glacial Maximum using a simula- tion from a state-of-the-art, fully coupled climate model, the Community Climate System Model (CCSM3) developed at the National Center for Atmospheric Research (NCAR). We observe a strong, stable Atlantic jet and enhanced low level baroclinicity, but diminished wintertime eddy activity at all levels of the atmosphere compared to today’s climate. In other words, results from the CCSM3 simulation appear to be at odds with the atmosphere-only simulations discussed above.
We begin with a brief description of the model and methods in section 3.2. Section 3.3 presents the model results which indicate that, during Last Glacial Maximum, the atmosphere exhibited a strong, steady mean circulation with decreased eddy activity. Section 3.4 compares the CCSM3
simulation with simulations from PMIP-1, most of which show no weakening of low level eddy activity. In section 3.5, we analyze results from a series of uncoupled atmosphere model experiments designed to identify why the atmospheric circulation during glacial times is different in CCSM3 than in the PMIP-1 models, with a focus on determining the relative importance of land ice and sea surface conditions. Finally, the main results of this chapter are summarized in section 3.6.
3.2 Model description and methods
In this study, we investigate changes between present day (PD) and Last Glacial Maximum (LGM) climates as simulated by the Community Climate System Model 3 (CCSM3; Collins et al., 2006b), a global coupled atmosphere-ocean-sea ice-land surface climate model developed at the National Center for Atmospheric Research (NCAR). The setup of and results from these model simulations are documented in detail in Collins et al. (2006b) and Otto-Bliesner et al. (2006). Briefly, the CCSM3 comprises the primitive equation Community Atmosphere Model version 3 (CAM3) at T42 horizontal resolution with 26 hybrid coordinate vertical levels (Collins et al., 2006a); a land model with land cover and plant functional types, prognostic soil and snow temperature and a river routing scheme (Dickinson et al., 2006); the NCAR implementation of the Parallel Ocean Program (POP) on a 320 384 dipole grid (nominal horizontal resolution of 1 ) with 40 vertical levels (Smith and Gent, 2002); and a dynamic-thermodynamic sea ice model on the same grid as the ocean model (Briegleb et al., 2004).
The forcings for the LGM simulation are in accordance with the protocols established by PMIP-2 (http://www-lsce.cea.fr/pmip2): 21 ka BP insolation; atmospheric greenhouse gas concentrations based on ice core measurements (Fl ¨uckiger et al., 1999; D ¨allenbach et al., 2000; Monnin et al., 2001); atmospheric aerosols at preindustrial values; land ice and coastlines, corresponding to 120 m sea level depression, from the ICE-5G reconstruction (Peltier, 2004).
We use monthly mean output from 50 years of the simulations. Daily data for transient eddy analyses are taken from a 45 year branch run for the PD and a 25 year branch run for the LGM.
Eddy fields were filtered with a sixth order high-pass Butterworth filter to emphasize variability on 1–5 day timescales. The procedure used to determine that 25 years of data were adequate for generating stable eddy statistics is described in Appendix A.
Of interest here is the large scale atmospheric circulation in the model’s simulations of the Last Glacial Maximum (LGM) and present day (PD) climates. We will focus on the Northern Hemi- sphere Atlantic sector, where differences in forcings between the LGM and PD, and consequently circulation features, are most dramatic. All differences discussed are significant at the 95% level unless otherwise noted.
3.3.1 Circulation and heat transport
Upper level zonal wind and geopotential height provide a useful broad-brush picture of large scale flow characteristics of the atmosphere. Figure 3.1 shows wintertime maps of these two fields for the LGM and PD simulations. Under the ice age forcings described in the previous section, we observe an enhanced stationary wave associated with the Laurentide ice sheet covering most of North America, and a stronger, more zonal jet in the Atlantic sector downstream of the ice sheet.
Changes in the Pacific are more subtle, with evidence of a slight equatorward shift of the jet and the development of a split flow over Siberia.
The change in atmospheric circulation over the Atlantic sector during LGM is particularly strik- ing. As shown in Figure 3.2, there is an inverse relationship between maximum jet strength and jet width, where jet strength is the maximum zonal wind in the sector, and jet width is the latitude range over which decreases to half its maximum value. The relationship holds for both the PD (red circles) and LGM (blue triangles) climates, but each climate inhabits its own sector of width- strength space. The LGM jet is 30% stronger on average, with 35% less variability in maximum speeds than the PD jet; it is also 30% narrower, with almost six times less variability in width. The results are robust to the choice of month(s) used to define winter, and to the exact longitude range used to define the Atlantic jet.
The stronger winds during LGM are consistent with the stronger equator-to-pole surface tem- perature gradient. We note that, from the point of view of global energy balance, it is not obvious whether this stronger gradient should cause an increase in poleward heat flux by the climate system, or whether it is a symptom of inefficient poleward heat flux by the climate system. However, this topic of discussion is beyond the scope of this thesis work.
DJFM
JJA a
PRESENT DAY
DJFM
JJA b
LGM
DJFM
JJA c
DJFM
JJA d
Figure 3.1: Wintertime atmospheric circulation in CCSM3 simulations. Zonal wind at 250 mb from the (a) PD and (b) LGM simulations (10 ms
contours). The Atlantic jet is stronger and more zonal while the Pacific jet is slightly more zonal but largely unchanged. Geopotential height at 500 mb from the (c) PD and (d) LGM simulations (120 m contours with an offset of -5400 m). The Laurentide ice sheet over North America forces a strong stationary wave that intensifies the flow downstream.
jet width (latitude)
COUPLED
20 30 40 50 60
jet strength (m/s)
35 45 55 65
2 4 6 8
LGM PD
UNCOUPLED
20 30 40 50 60
jet strength (m/s)
vT850 (Kms−1 )
35 45 55 65
2 4 6 8
LGMa PDa
Figure 3.2: Atlantic jet and eddy characteristics in CCSM3 (coupled) and CAM3 (uncoupled) sim- ulations of PD and LGM climate. These plots show the relationship between the width and strength of zonal mean zonal winds and northward eddy heat flux (Kms
) at 850 mb over the At- lantic sector (90 W–0 ). CCSM3 (left): The top panel shows jet width versus jet strength for 50 Januarys in the PD (red) and LGM (blue) simulations; the bottom panels show the strength of the eddy flux versus jet strength (45 Januarys for PD, 25 Januarys for LGM). The horizontal line is the mean for each simulation, with the vertical dash marking the 95% confidence limits. CAM3 (right): These panels show results from the corresponding uncoupled simulations. The jet strength is the maximum zonal wind speed in the sector; the jet width is the latitude range over which the jet speed decreases to half its maximum value. The results are robust to the choice of month(s) used to define winter, and to whether the jet and eddy strengths are taken as the maximum in the zonal mean of different longitude ranges straddling the jet/storm track core, or as the area-weighted mean of the largest 30–50 values at the jet/storm track core.
Figure 3.3 shows implied annual meridional energy transports calculated according to the method of Trenberth and Caron (2001). First, the total required heat transport RT by the climate system is determined by integrating the top-of-atmosphere (TOA) radiation imbalance over all longi- tudes from the North Pole to each latitude
"!#%$&$ (3.1) where R is the radius of the Earth. Next, we sum the shortwave, longwave, latent heat and sensible heat fluxes from the model to get the annual mean surface heat flux('*),+ . Integrating -'*),+ over all ocean points gives the implied ocean heat transport OT
.
/
10
+32
45'6)&+
"!7 $ $8 (3.2) where
0
+32
9:<;=
>=? @
if (9 ) is ocean
A
if (9 ) is land B Finally, the atmospheric heat transport AT is calculated as a residual:
C
ED
. B (3.3)
We find that the amount of heat transported towards the poles is remarkably similar in the PD and LGM simulations (compare black curves and filled grey curves in lower panel of Figure 3.3).
The discrepancy between the model’s PD total heat transport and the satellite-derived radiatively required total heat transport from Trenberth and Caron (2001) is less than 0.5 PW, or 10% of the maximum heat transport (top panel of Figure 3.3). Between the two model simulations, the LGM does indeed have slightly more transport, with the atmosphere helping to increase the peak Northern Hemisphere (NH) value to 6.28 F 0.09 PW, about 0.3 PW greater than in PD. The increase is, however, relatively modest. All else being equal, a back-of-the-envelope calculation tells us that a 0.3 PW boost in heat transport at 35 N translates to a 3 Wm
boost in heating rate north of this latitude circle. Assuming a mid-range climate sensitivity of 0.5 C per Wm
, this is equivalent to a 1.5 C warming of the mid-to-high latitude regions. Clearly, all else is not equal, and the
−6
−3 0 3 6
PW
ERBE total NCEP ocean ECMWF ocean
−90 −60 −30 0 30 60 90
−6
−3 0 3 6
latitude
PW
LGM total LGM ocean
Figure 3.3: Meridional heat transport from observations and simulations. The filled curves show the annual heat transport by the atmosphere (dark grey) and ocean (light grey) in the PD simula- tion and are the same in both panels. The top panel compares the PD results to recent estimates (Trenberth and Caron, 2001) using data from the Earth Radiation Budget Experiment (ERBE) and reanalyses from the National Centers for Environmental Prediction (NCEP) and the European Cen- tre for Medium-Range Weather Forecasts (ECMWF). The bottom panel compares the PD to the LGM simulation. Maximum uncertainties are 0.15 PW, which is less than half the value of the largest differences between the model simulations.
ï ï
ï ï
total
stationary wave component
Figure 3.4: Total atmospheric heat transport and heat transport by stationary waves in CCSM3 simulations. The stationary wave heat transport was estimated by taking the vertical integral of [ ]. The LGM has much stronger stationary wave heat transports in the Northern Hemisphere due to ridging caused by the presence of the Laurentide ice sheet.
actual surface temperature difference between the two climates poleward of 35 N is closer to 10 C.1 The robustness of the total heat transport curve is corroborated by other Last Glacial Maximum simulations from coupled climate models (Shin et al., 2003) and from uncoupled general circulation models using a prescribed sea surface temperature (SST) forcing (Hall et al., 1996).
Upon closer inspection, there are interesting differences in the partitioning of these heat fluxes in the PD and LGM simulations. During LGM, the ocean accounts for slightly less of the transport in the NH and the atmosphere transports slightly more through the midlatitudes. Within the atmosphere itself, the pronounced ridging forced by the Laurentide ice sheet boosts the dry stationary wave heat transport contribution in the LGM compared to PD (Figure 3.4). Between 40–65 N, the LGM shows total atmospheric heat transport comparable to that in the PD, but greater stationary wave heat transport. By inference, the transient heat transport term must be diminished. In a global view, these results are in fact in agreement with the energy budget analysis of the UGAMP LGM simulation by Hall et al. (1996). Although they observe an increase in transient heat transport at low levels, there is a compensating decrease aloft such that the column-integrated eddy transport is lower during LGM.
3.3.2 Transient eddy activity
We can diagnose the eddy activity responsible for this transient heat transport by calculating high- pass filtered quantities such as low level temperature flux ( ) and upper level momentum fluxes ( , ). In addition to presenting maps of these eddy fluxes, we will use several metrics to quantify the steadiness of the atmospheric circulation at low levels and aloft during the winter DJFM season. The first metric is the maximum zonal wind at 200 mb in the Atlantic sector (15 N–65 N, 90 W–0 ). The second is a kinetic energy index of monthly departures of the 200 mb flow from the climatological mean averaged over the Atlantic sector,
@
D
D $ (3.4)
1Technically, the 0.5 C per Wm value is a global climate sensitivity, and should not be used to estimate the response of the polar cap to increased heat flux from the lower latitudes. However, the polar cap as defined in our calculation (35 N to the North Pole) is quite a large region, and one that sees a net TOA radiation loss to space. Furthermore, our use of this global climate sensitivity is supported by an experiment in Seager et al. (2002) in which ocean heat transport was turned off, leading to a 1.3 PW reduction in heat moved across 35 N, a 13 Wm decrease in heating rate north of this latitude circle, and a 6 C cooling of the mid-to-high latitude regions.
means. The final metric is the northward eddy heat flux at 850 mb averaged over the Atlantic sector,
@
$ (3.5)
where primes indicate daily fields that have been high-pass filtered to retain variability at periods less than 8 days.
Concentrating on boreal winter in the NH, the eddy fields reveal an LGM climate that appears more quiescent than the PD (Figure 3.5). The reduction in eddy activity from PD to LGM is espe- cially striking in the Atlantic sector, decreasing by over 40% at low levels according to (Ta- ble 3.1). Compared to PD, the LGM jets are strong and tight (Figure 3.6), and the eddy fluxes occupy a narrow latitudinal band hugging the axis of the jet core rather than a broader band perched on the poleward flank of the jet (Figure 3.5a–d). The differences between the two climates can also be seen in Figure 3.2, in which January poleward heat fluxes in the LGM simulation (blue triangles) span a tight range of smaller values than in the PD simulation (red circles). Although we will not be discussing the Pacific sector in this work, it too exhibits changes in jet structure and eddy fluxes (see Appendix B).
The measures of atmospheric flow in Table 3.1 offer additional information about the differences between the Atlantic sector in today’s climate and in glacial climates. The weak eddy activity in the LGM simulation coexists with a stronger, narrower Atlantic jet, as we saw in Figure 3.2. From this plot, we inferred a less variable upper level flow field during LGM compared to PD. The winter season flow metrics now provide additional and more direct ways to evaluate the steadiness of the LGM jet. Looking at Table 3.1, we note a 30% decrease in jet speed variability ( in column two) as well as a 25% decrease in monthly departures of kinetic energy from its climatological mean state ( in column three). Together, these results are consistent with the picture of global heat transport in Figure 3.4, in which a slight overall increase in meridional heat flux during LGM is achieved by a greatly enhanced stationary wave, with the implication that the contribution from transient eddies must be weaker.
As a final comment on this topic, we note that the gross structure of the Atlantic jet and eddies described here for the PD simulation is supported by reanalysis data. The NCEP reanalysis (1948–
Table 3.1: Jet and eddy characteristics in the Atlantic sector (15 N–65 N, 90 W–0 ) for DJFM winter. is the maximum zonal wind at 200 mb in the sector; is the standard deviation of
; is the sector-mean monthly departure from the climatological mean of the kinetic energy of the horizontal wind at 200 mb; and is the sector-mean northward eddy heat flux at 850 mb. The 95% confidence intervals were determined using a Student’s t-test for the means, and a chi-squared test for the standard deviations. The rows below the double line are for uncoupled experiments forced with a variety of boundary conditions; these are discussed in section 3.4 and section 3.5.
m s m s m s K m s
ECMWF 38.5 1.0 6 2 37 2 — —
NCEP 38.2 0.9 6 2 36 3 3.93 0.05
PD 47.3 0.9 6 2 32 3 6.24 0.09
LGM 58.7 0.5 4 1 20 2 3.51 0.09
PDa 45.6 1.1 6 2 34 2 4.17 0.08
LGMa 57.7 0.7 4 1 18 2 2.85 0.07
PMIPa 39.1 1.0 6 2 25 2 3.75 0.09
CLIMAP5G 47.5 0.8 5 2 20 2 3.52 0.09 LGM4G 50.1 0.9 5 2 25 3 3.55 0.08
NHsst5G 46.7 0.8 5 2 21 2 3.10 0.07 NHsi5G 58.4 0.6 4 1 17 2 3.07 0.07
