Cover Image

Not for Sale



View/Hide Left Panel

3
Climate Steps in Ocean History—Lessons from the Pleistocene

WOLFGANG H.BERGER

Scripps Institution of Oceanography and Universität Kiel

INTRODUCTION

There is evidence from the study of deep-sea sediments that the climatic (and therefore geochemical and evolutionary) history of the ocean is characterized by a series of transitions from one state to another. The general impression of instability, which arises from contemplating a history full of transitions, has been captured in the phrase “commotion in the ocean” (Berggren and Hollister, 1977). Prime examples (see Table 3.1) of major transitions are the Cretaceous termination (Thierstein, Chapter 8), the Eocene-Oligocene boundary event (Benson, 1975; Kennett and Shackleton, 1976), the mid-Miocene oxygen isotope shift (Savin, 1977), the 6 Ma (million years ago) carbon isotope shift (Keigwin, 1979; Vincent et al., 1980), and the 3 Ma northern glaciation onset (Berggren, 1972; Shackleton and Opdyke, 1977). Also there are many less spectacular events, some of which apparently are of a recurring kind and belong to cyclic or quasi-cyclic phenomena (see Fischer and Arthur, 1977; Haq et al., 1977; van Andel et al., 1977; Berger, 1979; Arthur, 1979).

There are two fundamentally different ways to approach the discussion of climate steps recorded in sediments. One is to take each event as a unique occurrence that calls for a unique explanation. For example, the 3 Ma northern glaciation event might be viewed as a result of closing the Isthmus of Panama, with the deflection of previously westward traveling Caribbean waters into the Gulf Stream resulting in increased moisture supply in high latitudes, and hence increased snowfall. Alternatively, the same event might be seen as an inevitable consequence of a general cooling trend with strong positive feedback setting in, from albedo increase, once a snow-cover lasts For a certain part of the year. The first approach emphasizes a cause that is external to the climate-producing system, the second focuses on positive feedback mechanisms within the system. The second approach can be applied to an entire class of events, as well as to the amplification of cyclic signals within the period in question. It need not, of course, exclude the search for prime causes, both for the general trend onto which the event is grafted and for the exact timing of the event.

This chapter summarizes some concepts in connection with event analysis (Berger et al., 1977, 1981; Thierstein and Berger, 1978; Vincent et al., 1980). The basic proposition is to separate the external causes (e.g., irradiation changes, opening or closing of basin connections) from the internal sources of insta-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 43
Climate in Earth History: Studies in Geophysics 3 Climate Steps in Ocean History—Lessons from the Pleistocene WOLFGANG H.BERGER Scripps Institution of Oceanography and Universität Kiel INTRODUCTION There is evidence from the study of deep-sea sediments that the climatic (and therefore geochemical and evolutionary) history of the ocean is characterized by a series of transitions from one state to another. The general impression of instability, which arises from contemplating a history full of transitions, has been captured in the phrase “commotion in the ocean” (Berggren and Hollister, 1977). Prime examples (see Table 3.1) of major transitions are the Cretaceous termination (Thierstein, Chapter 8), the Eocene-Oligocene boundary event (Benson, 1975; Kennett and Shackleton, 1976), the mid-Miocene oxygen isotope shift (Savin, 1977), the 6 Ma (million years ago) carbon isotope shift (Keigwin, 1979; Vincent et al., 1980), and the 3 Ma northern glaciation onset (Berggren, 1972; Shackleton and Opdyke, 1977). Also there are many less spectacular events, some of which apparently are of a recurring kind and belong to cyclic or quasi-cyclic phenomena (see Fischer and Arthur, 1977; Haq et al., 1977; van Andel et al., 1977; Berger, 1979; Arthur, 1979). There are two fundamentally different ways to approach the discussion of climate steps recorded in sediments. One is to take each event as a unique occurrence that calls for a unique explanation. For example, the 3 Ma northern glaciation event might be viewed as a result of closing the Isthmus of Panama, with the deflection of previously westward traveling Caribbean waters into the Gulf Stream resulting in increased moisture supply in high latitudes, and hence increased snowfall. Alternatively, the same event might be seen as an inevitable consequence of a general cooling trend with strong positive feedback setting in, from albedo increase, once a snow-cover lasts For a certain part of the year. The first approach emphasizes a cause that is external to the climate-producing system, the second focuses on positive feedback mechanisms within the system. The second approach can be applied to an entire class of events, as well as to the amplification of cyclic signals within the period in question. It need not, of course, exclude the search for prime causes, both for the general trend onto which the event is grafted and for the exact timing of the event. This chapter summarizes some concepts in connection with event analysis (Berger et al., 1977, 1981; Thierstein and Berger, 1978; Vincent et al., 1980). The basic proposition is to separate the external causes (e.g., irradiation changes, opening or closing of basin connections) from the internal sources of insta-

OCR for page 43
Climate in Earth History: Studies in Geophysics TABLE 3.1 Examples of Fast Climatic Transitions in the Cenozoic 1. Terminations of the late Pleistocene (0.8 Ma)   Effect: Rapid deglaciations, warming of mid-latitudes, pCO2 change. Refs.: Emiliani, 1955, 1966, 1978; Broecker and van Donk, 1970; Dansgaard et al., 1971; Shackleton and Opdyke, 1973; Berger and Killingley, 1977; Berger et al., 1977; Rooth et al., 1978; Delmas et al., 1980. 2. 1-Ma Event   Effect: Onset of large climatic fluctuations after period of quiescence. Refs.: Shackleton and Opdyke, 1976; van Donk, 1976; Vincent and Berger, 1982. 3. 3-Ma Event   Effect: Onset of Pleistocene-type climatic fluctuations. Refs.: Berggren, 1972; Shackleton and Opdyke, 1977; Keigwin, 1978, 1979. 4. Messinian Salinity Crisis (~5.5 Ma)   Effect: Isolation of Mediterranean through regression, strong cooling. Refs.: Hsü et al., 1973, 1977; van Couvering et al., 1976; Adams et al., 1977. 5. 6-Ma Carbon Shift   Effect: Isotope ratios of the ocean’s carbon shift to lighter values, presumably owing to organic carbon input from regression and erosion. pCO2 change(?). Refs.: Bender and Keigwin, 1979; Keigwin, 1979; Vincent et al., 1980; Berger et al., 1981. 6. Mid-Miocene Oxygen Shift(?) (~15 Ma)   Effect: Isotope ratios of the ocean’s oxygen shift to heavier values, presumably owing to Antarctic ice buildup. Refs.: Douglas and Savin, 1975; Savin et al., 1975; Shackleton and Kennett, 1975; Savin, 1977. 7. Mid-Oligocene Oxygen Shift(?) (~30 Ma)   Effect: First occurrence of rather heavy oxygen isotope values in deep-sea benthic foraminifera, presumably owing to polar bottom-water formation. Refs.: Savin, 1977; Arthur, 1979. 8. Eocene Termination Event (~38 Ma)   Effect: Cooling in high and low latitudes, expansion of polar highs. Significant changes in deep-sea benthic foraminifera and of high-latitude planktonic foraminifera. Rapid drop of the carbonate compensation depth. Refs.: van Andel and Moore, 1974; Benson, 1975; Kennett and Shackleton, 1976; Savin, 1977; Burchardt, 1978. bility. These sources of instability are of two kinds. The first are regular amplification mechanisms, such as albedo feedback, whose strength is more or less proportional to the excursion from the steady-state condition. The second are strong interferences from “transient reservoirs” of geochemical disequilibrium, which are rather unpredictable. An important feature of the concept of transient reservoirs is that an exchange of “disequilibrium energy” between transient reservoirs of different kinds can lead to oscillations of the type observed in mechanical and electrical systems. THE TASKS OF PALEOCLIMATOLOGY The need for step analysis as a means to advance the science of paleoclimatology must be framed within the entire scope of this discipline. The task of paleoclimatology is to record and explain the climatic trends and events that have occurred throughout the Earth’s history. In order to develop models for sequences of climatic states, we must study periods of some duration, with adequate sampling sets of states and their transitions. The Pleistocene, especially the late Pleistocene, which includes the largest known climatic fluctuations, is such a set. One central task is to analyze the Pleistocene record in a way that provides analogies for the understanding of more ancient climates. In doing so, it is useful to focus on the systematic aspects of climate (Kominz and Pisias, 1979; Imbrie and Imbrie, 1980) rather than on the physical aspects of climate, which tend to be poorly constrained for this time period. The Pleistocene is itself part of a long-term climatic trend—that of an overall cooling since the early Tertiary. It also contains a trend within it—that of ever-increasing amplitudes of climatic excursions (see e.g., Shackleton and Opdyke, 1976). Both the onset of northern glaciation and the trend within the Pleistocene are reasonable to ascribe to an increase in positive feedback within the system, the obvious candidate being an increase in the role of snow cover in the heat budget of the Earth’s surface. Likewise, instability evidently grows with growing ice caps. Rapid large transgressions become possible through the storage of continental ice masses; such transgressions can cause rapid changes in albedo (water is dark; land is bright; see Table 3.2). TABLE 3.2 Albedo Values of Ocean and Land Surfacesa,b Annual global average: 14 Ocean: low latitudes: 4–7 mid-latitudes: 4–19 high latitudes: 6–50 Great lakes: min. (summer): 6 max. (winter): 55 Land: desert: 20–30 grasslands, coniferous, and deciduous forests: 15 wetlands: 10 tropical rain forests: 7 snow-covered land: 35–82 Antarctic Continental Ice Cap: 85 pack ice in water: 40–55 aReflectivity in percent of incident light, during noon. bSource: Compilation of Hummel and Reck, 1978.

OCR for page 43
Climate in Earth History: Studies in Geophysics The buildup of continental ice causes overall regression and may, itself, be a consequence of regression (Hamilton, 1968)—a prime example of positive feedback. Regression, besides leading to increased albedo, may create other sources of instability in addition to snow and ice, as we shall see. The most striking feature of the Pleistocene record—the climate cycles driven by the Milankovitch mechanism—reveals the activity of strong positive feedback (Kukla, 1975; Suarez and Held, 1979). Parts of the pre-Pleistocene record show cycles also, but generally of a much smaller amplitude. We must assume that before the Ice Age the lack of strong albedo feedback allowed negative feedback to hold sway in dampening climatic fluctuations. Negative feedback, of course, ultimately provides the climatic stability that allowed life to exist on Earth for billions of years. However, negative feedback typically has substantial time lags, which presumably are of the same order as the leads and lags between various climatic indices, as well as the climatic response times. In the Pleistocene such lags are of the order of 5000–10,000 years (Moore et al., 1977; Imbrie and Imbrie, 1980). Thus, when strong positive feedback exists, large climatic excursions can develop before the system brakes itself. To study the various ways in which sea-level fluctuations produce (and interact with) trends, cycles, and events and the roles of positive and negative feedback in amplifying and dampening climatic input functions would appear to be the most challenging tasks of paleoclimatology. We are still at the beginning of meeting this challenge. Correlations of various climatic, geochemical, and evolutionary signals with sea-level fluctuations have been suggested (Fischer and Arthur, 1977), but the linking mechanisms remain obscure. Quantitative analysis of pre-Pleistocene climatic trends has been attempted (Donn and Shaw, 1977) but without consideration of feedback. The most advanced studies are those modeling the climatic conditions of the last glacial (Gates, 1976a, 1976b; Manabe and Hahn, 1977) and those that extract frequencies and phase shifts from Pleistocene deep-sea records (Hays et al., 1976; Pisias, 1976; Moore et al., 1977). The modeling of climatic conditions is not the same as the modeling of climatic change. The extraction of frequencies is of prime importance for finding the input functions, but it leaves open the question of internal feedback. Thus, the most intriguing mystery of the ice ages—how climate can change so fast—remains unresolved. Phase shifts between different climate-related signals are potentially revealing as far as cause-and-effect chains, much as one would expect from the analysis of time segments of rapid change. However, (1) shifts may differ between various types of climate excursions, so that the result of a bulk analysis extending over a long period may be misleading; and (2) shifts between signals may be produced artificially through mixing processes in the record as a result of changes in the concentration of the signal carriers (Hutson, 1980). There can be little doubt that the various tasks of paleoclimatology would be greatly facilitated if we had a detailed record of a number of climatic steps and some idea about the processes associated with them. Before going any further, however, we must ask whether suitable steps exist at all. THE REALITY OF STEPS The reality of rapid climatic change was first demonstrated by Emiliani (1955) through oxygen isotope stratigraphy of long continuous deep-sea records. These records provide the strongest support for the Milankovitch mechanism of Pleistocene climatic fluctuations (Figure 3.1). Geomagnetic dating of an isotope stratigraphy in the western equatorial Pacific (Shackleton and Opdyke, 1973) established the chronology that allowed the correct identification of the periodicities involved (Hays et al., 1976; Pisias, 1976). The isotope fluctuations, of course, are not a direct representation of an irradiation curve but the result of a convolution of radiation input with climatic feedback mechanisms involving ocean, atmosphere, snow cover, ice cover, and vegetation. When studying the curves generated by Emiliani (1955, 1966) and by Shackleton and Opdyke (1973, 1976) we note a striking phenomenon, important especially for the survival of higher organisms. The fluctuations never go beyond a certain maximum value on either side of the range (see Figure 3.1). Obviously, there is a limit to warming: radiation of heat into space increases approximately as the fourth power of the Earth’s surface temperature (the StefanBoltzmann law). The rapid rise of backradiation with increas- FIGURE 3.1 Composite δ18O curve Ocurvc of Emiliani (1978), showing sawtooth pattern and well-defined limits of δ18O maxima and minima. The rapid change after most δ18O maxima suggests attainment of a critical setting (interpreted as a buildup of sufficiently large transient reservoirs) and a “runaway” effect once melting reaches some critical rate (interpreted as reservoir collapse).

OCR for page 43
Climate in Earth History: Studies in Geophysics ing temperature constitutes efficient negative feedback. But what negative feedback is preventing continued cooling? Why does not the Earth cover itself with ice? Once ice spreads, albedo increases. Thus, less and less of the incoming radiation is retained for heating the Earth’s surface—ice becomes the stable phase over larger and larger areas. Yet, the Earth does not become white but remains blue and brown. One type of negative feedback that has been invoked to prevent total glaciation is the decrease of moisture transport that accompanies a general lowering of temperatures (cold air cannot hold much water) and a covering up of the sea surface in high latitudes with pack ice (Emiliani, 1978). Also, migration of the polar front toward mid-latitudes prevents moist tropical air from reaching the original centers of glacial growth. The glaciers in high latitudes would “starve” under such conditions. But would they disappear? And would glaciers not continue to grow in mid-latitudes? The question of negative feedback, which prevents the Earth from icing over, is open. One important factor, probably, is the removal of positive albedo feedback from falling sea level. Once the sea level falls to the shelf edge, a further drop does not result in much decrease of ocean surface; thus the albedo stabilizes. We now turn to the third important feature of the oxygen isotope record. Glacial maxima are followed, almost inevitably, by glacial minima, and the transitions are extremely rapid. The phenomenon was emphasized by Broecker and van Donk (1970), who called the transitions “terminations” and gave them numbers. They also coined the term “sawtooth” pattern to characterize the alternations between rapid deglaciations and more gradual (but fluctuating) buildup of ice. The smoothing activity of bioturbation on the seafloor is such that a change of the rapidity suggested by the deep-sea record of Termination I or Termination II (11,000 and 127,000 yr ago, respectively) is extremely difficult to envisage. Both an instantaneous flip-over from one climatic state to the next (Peng et al., 1977) and an overshoot phenomenon (Berger et al., 1977; Berger, 1978) have been suggested to account for this difficulty. If bioturbation worked then as it does today, almost any physically reasonable transition should be more gentle than that observed. One possibility is that we are looking at a hiatus in sedimentation. A gap in the recording would juxtapose different stages in history. If we entertain this notion of a gap, we are then faced with the necessity of providing a short event, an impulse, that produces nondeposition or erosion at the correct time. Thus the question of rapid change within the system would return, having merely been shifted from paleoclimatology to geochemistry. It is true that cessation (or great reduction) of bioturbation also would help; the problem might then be shifted to deep-sea biology. However, such shifting of responsibility does not come to grips with the central problem: that the system is changing rapidly and that this has consequences for the circulation of the ocean and atmosphere as well as for their chemical composition, and hence for climate and evolution. What is the importance of the deglaciation event, other than demonstrating the existence of rapid climatic change, or climate steps? Can we learn something about climate steps in general, even though physical mechanisms may vary widely? We may assume that a system in rapid transition cannot in any way be thought of as being intermediate between the previous and the subsequent state. The best support for this proposition again comes from the study of Pleistocene deep-sea sediments, this time from a region where high sedimentation rates allow a detailed look at Termination I, namely, the Gulf of Mexico (Kennett and Shackleton, 1975; Emiliani et al., 1975). The oxygen isotope records of planktonic foraminifera show a marked excursion toward light values at the end of a rapid (but apparently pulsating) rise from the glacial maximum (Figure 3.2). The curves, in fact, look much like the standard amplitude-versus-time response plots in the textbooks of systems analysis, familiar to students of mechanical and electrical engineering. In principle, such plots describe the response of a system to a step input. The steepness of the transition from one state to the next is a measure for the sensitivity of the system, as is the amount of overshoot (Figure 3.3). Essentially we see here the result of the competition between negative feedback, which slows the transition, and positive feedback, which accelerates it and builds up the overshoot. The overshoot is characterized by being short-lived and by preceding another period that mimics the original state, i.e., a kind of undershoot. To produce this rebound, positive feedback is active in reverse. More oscillations can then follow, depending on the strength (or weakness) of the dampening processes in the system. Let us, for the sake of argument, accept the proposed analogy between the oxygen isotope record of the Gulf of Mexico and a two-phase step response (Figure 3.4). Have we thereby done any more than introduce some terms from systems analysis to the description of a set of phenomena, without a net increase in knowledge or understanding? FIGURE 3.2 Oxygen isotope record of the last glaciation in the Gulf of Mexico by Kennett and Shackleton (1975), as dated by correlation with a similar (14C-controlled) record of Emiliani et al. (1975).

OCR for page 43
Climate in Earth History: Studies in Geophysics FIGURE 3.3 Sketch of three types of response to a step input in a two-dimensional system. On the contrary, I suggest that we have now changed the mode of attack on the problem. Recall that the Gulf of Mexico isotope anomalies were originally interpreted by their discoverers as indicating an unusual influx of meltwater from the disintegrating Laurentian Ice Sheet (Kennett and Shackleton, 1975; Emiliani et al., 1975). According to these authors the low 18O/16O ratio in the meltwater produced the anomaly, after mixing with the seawater of the Gulf. Ensuing discussions accepted the explanation (which appeared reasonable) but questioned Emiliani’s 14C date of 11,500 yr ago because it did not agree with evidence on land, regarding the course and the timing of meltwater flow. In fact, these discussions may have missed the point. The meltwater influx could have been at a maximum well before the isotope anomaly. The rapid change in the isotope signal that precedes the anomaly is as much witness to the rapid introduction of meltwater as is the anomaly itself. We cannot dismiss the possibility that the difference in δ18O between the upper waters of the Gulf and the global ocean was just as great during this period, which corresponds to the “ramp” of the signal, as during the anomaly itself. Thus, an exclusive focus on the anomaly is not the right approach: the anomaly and the ramp belong together, just as the step-function analogy would suggest. Evidently, the anomaly might be expected at the end of the rapid change, whether or not the flux of the Mississippi River increased at anomaly time. Thus, a purely formal consideration—seeing the record as a step response—changes the argument considerably, THE SEARCH FOR POSITIVE FEEDBACK The questions that arise within the step-function analogy are quite general: What physical processes limit the rate of transition from glacial to postglacial values? Is the gradient of δ18O versus time steeper in the Gulf than outside of it? Is there an overshoot phenomenon outside the Gulf also? If so, can we identify positive feedback mechanisms that could produce it? What is the nature of the “rebound” following the anomaly? If it has the structure of a pendulum swing, why is the period near 2000 yr? And which are the reservoirs of disequilibrium that pass the equivalent of kinetic and potential energy between them? We cannot answer these questions at present. However, we might usefully consider where to begin the search. In the present case, naturally, the most obvious candidates for physical processes providing positive feedback are those having to do with the melting of ice. The melting takes heat, and there is a limit to the rate at which heat can be transported to the site of melting. Incidentally, if much of the heat comes in the form of rain, the runoff will be a mixture of rain and melt-water and its oxygen isotope composition will be somewhere between that of rain and glacial ice. To get positive feedback, we must ask that the melting, once started at some minimum rate, enhance further melting. For example, increased local absorption of radiation by exhumed debris on top of glacial ice could help. The effect would seem insufficient, however, because it can be removed quickly by snow cover. Continuing vigorous heat transfer from the tropics would appear necessary. How can strong initial melting change the heat budget of the entire system in favor of its continuation? Meltwater does mainly two things: it raises sea level, and it decreases the salinity of the ocean. A sea-level rise decreases the albedo: as the ocean surface expands, absorption increases and more of the Sun’s radiation is used to heat the Earth’s surface. A rise in sea level can also, presumably, destabilize those parts of the glacial ice that rest on shelf and can be floated (J.T.Andrews, University of Colorado, personal communication, 1979). Conceivably such floating could favor the occurrence of ice surges of the type envisaged by Wilson and others (Wilson, 1964, 1969; Hollin, 1972; Flohn, 1974), which would accelerate the rise of sea level. (In the ice-surge hypothesis, the focus is on the attendant increase in albedo and cooling, however.) FIGURE 3.4 The Gulf of Mexico record (Figure 3.2) interpreted as a response of a two-dimensional system to a couplet of step inputs.

OCR for page 43
Climate in Earth History: Studies in Geophysics Assimilation of the freshwater into the ocean takes time. The time constant of mixing is nonnegligible, being of the order of 1000 yr in the present ocean. The rate depends on two factors: the strength of the mixing drive and the vertical stability of the water column. Mixing is both wind- and density-driven and is therefore closely tied to temperature gradients and salinity patterns. Stability depends on the density profile, and is likewise tied to temperature and salinity distributions. The relationships are such that an increase in the forces responsible for mixing, which are derived largely from the planetary temperature gradient, also leads to an increase in the opposing stability—mixing rate is resistant to change. However, the introduction of meltwater can change this situation, as it is much lighter than seawater and tends to float. A low-density surface layer does not prevent mixing by current shear and eddies through wind, but it can potentially interfere with deep- and bottom-water production. In theory, there is some range of the rate of introduction of meltwater where there is essentially no effect on mixing. At the upper end of this range there must be a critical point, where the rate of influx begins to exceed the ability of the ocean to assimilate the added freshwater at the given rate of mixing. At this point the mixing rate must slow. The questions are, what is the critical input rate, and was it ever reached during deglaciation? The first question can be attacked by modeling; the second must be read from the record, for example, through comparison of stable isotope stratigraphies of deep-sea benthic and planktonic foraminifera. Let us assume that the critical influx is indeed reached, and mixing slows. At this point the system changes its mode of operating. Because mixing has now slowed, the value for the critical rate of influx will begin to fall. The system develops a strong positive feedback by building up stable stratification, with a halocline at the bottom of the wind-mixed layer in the broad sense, say, between 500 and 1000 m. Furthermore, the ocean will now “remember” this condition for some time, in fact, for about 1000 yr or so. The value for the critical rate of influx will be lowered during this entire period, rising but slowly to its original level. The point of the discussion is this: if early in deglaciation there is a strong pulse of meltwater influx, stable stratification can be maintained through a series of lesser pulses later. There is in fact evidence that sea level rose in pulses (Fairbridge, 1961; Mörner, 1975), but the matter is still under discussion. Meltwater influx, then, can conceivably set into motion a strong nonlinear positive feedback mechanism, which goes beyond albedo decrease from ocean-area expansion and which involves the development of a low-salinity layer of the type suggested by Worthington (1968). But how can this potential for feedback be translated into an energy budget favorable for melting? These questions call for some rigorous modeling; at present, we can only guess what a low-salinity lid on the ocean would do to the climate. Presumably, a lack of communication with cold deep waters would allow low-latitude waters to heat up considerably, enhancing the meridional temperature gradient and hence the heat transport to higher latitudes. Also, the temporary decoupling of three fourths of the ocean mass from the heat budget should increase climatic instability: the inertia of the system is decreased. If true, this would favor delivery of meltwater in pulses, thus maintaining instability. While seeking strong positive feedback, we discovered a source of instability that can, once activitated, develop feedback for instability itself. Within the unstable system, small changes in input (e.g., from the Sun’s radiation) can be translated into larger climatic fluctuations. We have here one way to produce the rapid changes in climate that characterize the transition from glacial to postglacial time. Is the development of instability typical for fast climate transitions? Are there nonglacial climate-changing mechanisms analogous to the melting of ice? If yes, analogous in which sense? THE PHENOMENON OF RESERVOIR COLLAPSE Glacial ice may be seen as a transient reservoir of water, outside the main ocean basin, which, when reunited with the ocean, suddenly raises sea level with all the attendant effects on climate through a decrease in albedo and an increase of supply of moisture to the atmosphere. We can view the melting of ice caps as a “reservoir collapse” that feeds on itself once destruction proceeds at a minimum rate. Glacial ice is a reservoir of freshwater, hence the potential for additional complications. Can we envisage other types of transient reservoirs? Perhaps so. Several other possibilities are shown in Figure 3.5. The most obvious transient reservoirs are adjunct ocean basins and marginal seas. A reservoir of glacial ice is analogous to a reservoir of water in an isolated basin. If the basin can run dry, as the Mediterranean did at 5 Ma (Hsü et al., 1977), the analogy is almost perfect. We can produce, through alternating emptying and filling of such a basin, rapid transgressions and regressions. The volume of the Mediterranean Sea, for example, would have allowed almost instantaneous changes in sea level of near 10 m in the Messinian, whereas that of the South Atlantic might have allowed changes of about 50 m in the Aptian (Berger and Winterer, 1974). If such changes occurred, we should see them both in the deep-sea record and on the shelves. We also should expect substantial evidence for climatic instability. On the one hand, emptying a marginal sea can affect albedo over a large area—including not only the basin itself but also the hinterland around it, which depends on moisture from the basin to maintain its vegetation. On the other hand, rapid transgressions and regressions have their own global effects on albedo and moisture distribution. The potential for the existence of isolated basins is quite large since the breakup of Pangaea. Salt deposits at ocean margins, e.g., around the Atlantic (Emery, 1977), suggest that large isolated basins existed at various times in the Mesozoic and in the late Paleozoic. If such basins did indeed exist, we should see the evidence in the global correlation of fast sealevel fluctuations during certain periods. Isolated basins can provide separate transient reservoirs of water, but they can also provide reservoirs of water of low or

OCR for page 43
Climate in Earth History: Studies in Geophysics FIGURE 3.5 Diagrammatic representation of sources of instability. External: Sun and mantle processes (1, 2). (Feedback from ice growth and decay on mantle processes cannot be excluded, however.) Internal: transient reservoirs of geochemical energy; 3 to 5, marginal seas with unusually saline waters or freshwaters; 6, continental ice masses; 7, adjunct ocean basins with salinity deviations; 8, easily erodable shelf carbonates, carbon deposits, and phosphatic sediments; 9, “soil” carbon accessible to fast erosion; 10, man’s activity (industrial CO2, deforestation, accelerated erosion). The transient reservoirs 3 to 7 both influence and respond to sea level variation, the reservoirs 8 and 9 collapse during regression. Source 10 may eventually respond to negative feedback from climatic change. high salinity. Injection of such waters, presumably occurring repeatedly during periods of critical isolation, could have had a profound influence on the history of evolution of climate and life (Gartner and Keany, 1978; Thierstein and Berger, 1978; Berger and Thierstein, 1979). More generally, the existence of semi-isolated reservoirs of brackish or supersaline water is a source of instability whose scale is tied to the size of the reservoir, the degree of deviation of salinity from the global average, and the potential flux exchange. There are, even today, two large reservoirs that are almost isolated and that could become much more so with a drop of sea level of between 100 and 200 m: the Mediterranean Sea and the Arctic Ocean. The Mediterranean is anomalously salty and plays an important role in the deep circulation of the Atlantic Ocean. The history of the Mediterranean outflow may be closely tied to that of North Atlantic bottom-water production (Reid, 1979). Deep circulation in the Mediterranean apparently reversed its direction in the earliest Holocene because of an increased supply of freshwater (Kullenberg, 1952; Williams et al., 1978), which removed one source of heavy deep water in the North Atlantic. The effects (if any) on the deep circulation have not been modeled; I suspect they were substantial. The Arctic Ocean has unusually low salinities in its surface waters. Its connection with the world ocean is somewhat tenuous; at least one geologist suggests that it was severed entirely during glaciation (M.Vigdorchik, INSTAAR, personal communication, 1978). If, as expected, the connection between the Arctic and Pacific Oceans was cut off during glacials, and that between the Arctic and Atlantic Oceans was greatly reduced, the Arctic Ocean might have collected brackish water throughout. Thus, when sea level first rose, low-salinity water from the Arctic Ocean could have helped to start off the deglaciation feedback chain postulated earlier. The development of transient-water reservoirs and hence of a potential for strong climatic instability is not necessarily restricted to the time since the breakup of Pangaea. In earlier times back-arc basins might have provided transient reservoirs under favorable conditions. Again there is a recent analog: the Pleistocene record of the Japan Sea contains layers with brackish-water diatoms, suggesting substantial isolation (Burckle and Akiba, 1978). Although the occurrence of transient-water reservoirs is a fact, the effect of such reservoirs on climate is virtually unstudied. The reservoirs are part of the hydrological cycle; they build up energy within this cycle, which might be released all at once, with quasi-catastrophic consequences. Are there other geochemical cycles for which this might also be true? TRANSIENT CARBON RESERVOIRS The carbon cycle is another obvious candidate for the presence of transient reservoirs. The phosphorus cycle also is a likely choice (Arthur and Jenkyns, 1980), as is the sulfur cycle (Holser, 1977). That the carbon cycle is intimately connected to climatic change is obvious from an overall parallelism of carbon isotope fluctuations with oxygen isotope fluctuations in the deep-sea record on various scales (Broecker, 1973; Berger, 1977b; Fischer and Arthur, 1977; Shackleton, 1977). Various mechanisms have been proposed through which the linkage could be achieved. Long-term fluctuations in the ratio of 13C to 12C in carbonates were ascribed by Tappan (1968) to variations in the accumulation rate of organic carbon. This idea has since been elaborated on in several guises. Rapid fluctuations in δ13C in the Pleistocene were related by Shackleton (1977) to the buildup and destruction of tropical rain forests,

OCR for page 43
Climate in Earth History: Studies in Geophysics FIGURE 3.6 Upper Miocene isotopic stratigraphies, DSDP Site 238, tropical central Indian Ocean, from Vincent et al. (1980). Note the carbon isotope “shift” of approximately 0.8 ‰ toward lighter values upward in the section between samples 19–6 and 20–6. Note the possibility of an “overshoot” and a “rebound,” and the (phase-shifted?) fluctuations in the δ18O signal.

OCR for page 43
Climate in Earth History: Studies in Geophysics which represent a substantial part of the biosphere. We can extend this idea to include the carbon of extratropical forests, marshes and swamps, peat and “soil carbon” in shelf and coastal deposits (all readily available for decay or erosion) in a pool of “transient carbon”; short-term variations in the size of this reservoir should have effects on the chemistry of the ocean and on the CO2 content of the atmosphere (Figure 3.5). Direct evidence for changes in pCO2, from glacial to postglacial time, was recently reported from ice cores (Delmas et al., 1980). There is little question that the size of the transient carbon reservoir must have varied considerably during the Pleistocene. Apparently, the last glacial maximum was dry, deserts being widespread and the tropical rain forests being greatly reduced (Sarnthein, 1978). In contrast, the early Holocene was a wet period over large areas; the reversal of the deep circulation in the Mediterranean is part of this phenomenon. During the transition from the dry glacial to the wet Climatic Optimum, the transient carbon reservoir was affected both by the buildup of the biosphere (essentially forests) and the erosion of soil carbon. The buildup extracts carbon from the ocean-atmosphere system; the erosion delivers carbon to it. If these processes fluctuate, with a phase shift near 180°, there may be a remarkable potential for introducing climatic instability via variation of the CO2 content of the atmosphere. That the carbon chemistry of the ocean underwent major changes during deglaciation is clear from large vertical excursions of carbonate preservation levels on the seafloor, during a short period (Berger, 1977a). If stable stratification developed during deglaciation, the exchange of CO2 between ocean and atmosphere must have been severely affected (Worthington, 1968). How do these observations and speculations bear on prePleistocene climates? Evidently, the waxing and waning of ice caps is a sufficient but not a necessary condition for producing fluctuations in the transient carbon reservoir. Any mechanism producing fluctuations in sea-level and dry-wet cycles will do. Of special interest is the possibility that the size of the transient carbon reservoir can be greatly increased if the deep ocean provides temporary carbon storage through changes in mixing time and oxygenation of the deep sea. Reactive organic carbon can accumulate on a poorly oxygenated seafloor and can be redelivered to the system on improvement of aeration. Opportunities for instability and for rapid climatic change might arise from the presence of such a marine transient carbon reservoir. For example, small radiation input cycles, leading to slight fluctuations in the oxygenation of various parts of the ocean via pulsating production of deep and bottom waters, could then translate into a pulsating supply of CO2 from the ocean to the atmosphere. The existence of oxygenation cycles in Mesozoic and early Tertiary deep-sea sediments is of interest in this connection (Dean et al., 1978). OVERSHOOT AND REBOUND Earlier, when discussing the nature of steps, we have seen that a step input can produce an overshoot and a rebound toward the original condition. In the case of deglaciation, the period known as “Alleröd” apparently was an overshoot and the “Younger Dryas” was a rebound. Having identified two components of the climate system that can store disequilibrium— the hydrosphere and the active carbon sphere—it is now in principle possible to construct a two-dimensional oscillating system. In such a system the disequilibrium is passed back and forth from one compartment to the other; the rate of transfer determines the periodicity of the oscillation. In our deglaciation example, stable stratification (a “meltwater lid”) might lead to CO2 buildup in the deep sea (Worthington, 1968), which leads to CO2 loss in the atmosphere and hence cooling. Release of the deep-stored CO2 after mixing could then increase the CO2 content of the atmosphere and produce warming. The CO2-induced cooling and warming, of course, would feed back into meltwater pulsing. An oscillation period of 2000 to 3000 yr would seem reasonable, in view of a 1000-yr mixing time for the normal ocean. Incidentally, the CO2 fluctuation hypothesis agrees well with the CO2-concentration record reported from Antarctic ice (Delmas et al., 1980). In a recent study of oxygen and carbon isotope variations in the late Miocene focusing on the Magnetic Epoch-6 Carbon Shift, Vincent et al. (1980) presented evidence for a step followed by increased fluctuations in climatic signals (Figure 3.6). One is tempted to identify an overshoot and a rebound in the carbon isotope stratigraphy, following the step at 6.2 Ma. The wavelength of such an oscillation would be several hundred thousand years. However, it is difficult in this instance to separate possible oscillations from a general increase in instability that was presumably introduced by the factor causing the step. If regression and the isolation of the Mediterranean was a crucial factor in producing the signal, this creation of a transient reservoir would likewise be expected to increase climatic instability. In any case, one would like to see a closer spacing of samples, because the typical frequencies of climatic fluctuations—whether they be oscillations or not—should contain clues to the responsiveness of the transient reservoirs involved. Unfortunately, the quality of the cores traditionally recovered by the Glomar Challenger sets severe limits for stratigraphic resolution. The new method of hydraulic piston coring removes this obstacle and offers a better definition of climatic fluctuations; this tool should be used to full advantage. SUMMARY AND CONCLUSIONS The deep-sea record provides a number of stratigraphic intervals showing a rapid transition from one climatic-geochemical state to another. These intervals provide an opportunity to study the dynamics of the ocean-atmosphere system on a scale from 103 to 106 yr. Step-function analysis provides useful concepts for the study of such transitions, as can be readily demonstrated using the last “termination” event in the Pleistocene record. There is an intriguing possibility that a system in transition oscillates, because of the passing of geochemical “disequilibrium energy” from one transient reservoir to another, analogous to mechanical systems (potential versus kinetic energy) and electric circuits (magnetic versus electric fields). In a glaciated world, the likely candidates for transient reservoirs are ice caps and temporary carbon pools in forests and

OCR for page 43
Climate in Earth History: Studies in Geophysics soils and on shallow seafloors. Geologic periods associated with strong positive feedback in the system (variable snow-cover) and with a large transient reservoir potential (ice caps, semi-isolated basins, reactive terrestrial and marine soil carbon and biocarbon) are characterized by climatic instability. The breakup of Pangaea and the subsequent dispersion of continents and creation of semi-isolated ocean basins must have produced constellations of instability at various stages in the evolution of present-day geography. One type of evidence for such constellations is in the salt deposits of continental margins. From an operational point of view, system analysis suggests an approach that considers the following questions when studying climate steps: (1) Climate step or hiatus? (2) Strongly or weakly damped? (3) Overshoot and rebound present? (4) External cause (astronomy, tectonics)? (5) Nature of internal feedback? (6) Likelihood of transient reservoir collapse? (7) Nature of disequilibrium oscillations? Even tentative answers to such questions should lead to fruitful working hypotheses regarding climatic change over geologic time spans. The concept of interacting transient reservoirs suggests that climatic systems are not strictly deterministic, that is, they are “almost-intransitive” (Lorenz, 1968) even over long periods of time. ACKNOWLEDGMENTS My interest in climate steps developed in the course of studies done in collaboration with R.F.Johnson, J.S.Killingley, H.R.Thierstein, and E.S.Vincent. Hannes Vogler (on hearing the “meltwater spike” story several years ago) first suggested to me to apply impulse-function concepts from systems analysis. W.L.Gates kindly criticized an earlier draft of the manuscript. The work was funded by the National Science Foundation (Oceanography Section) and by the Office of Naval Research (Marine Geology and Geophysics). REFERENCES Adams, C.G., R.H.Benson, R.B.Kidd, W.B.F.Ryan, and R.C. Wright (1977). The Messinian salinity crisis and evidence of late Miocene eustatic changes in the world ocean, Nature 269, 383–386. Arthur, M.A. (1979). Paleoceanographic events—recognition, resolution, and reconsideration, Rev. Geophys. Space Phys. 17. 1474–1494. Arthur, M.A., and H.Jenkyns (1980). Significance of rock phosphate and other geochemical sediments in the Cretaceous and Miocene oceans (abs.), 26e Congrès Géologique International Résumés, 1357. Bender, M.L., and L.D.Keigwin, Jr. (1979). Speculations about the upper Miocene change in abyssal Pacific dissolved bicarbonate 13C, Earth Planet. Sci. Lett. 45, 383–393. Benson, R.H. (1975). The origin of the psychrosphere as recorded in changes of deep-sea ostracod assemblages, Lethaia 8, 69–83. Berger, W.H. (1977a). Deep-sea carbonate and the deglaciation preservation spike in pteropods and foraminifera, Nature 269, 301–304. Berger, W.H. (1977b). Carbon dioxide excursions and the deep-sea record: Aspects of the problem, in The Fate of Fossil Fuel CO2 in the Oceans, N.R.Anderson and A.Malahoff, eds., Plenum, New York, pp. 505–542. Berger, W.H. (1978). Oxygen-18 stratigraphy in deep-sea sediments: Additional evidence for the deglacial meltwater effect. Deep Sea Res. 25, 473–480. Berger, W.H. (1979). Impact of deep-sea drilling on paleoceanography, in Deep Drilling Results in the Atlantic Ocean: Continental Margins and Paleoenvironment, M.Talwani, W.Hay, and W.F.B. Ryan, eds., Maurice Ewing Series 3, American Geophysical Union, Washington, D.C., pp. 297–314. Berger, W.H., and J.S.Killingley (1977). Glacial-Holocene transition in deep-sea carbonates: Selective dissolution and the stable isotope signal, Science 197, 563–566. Berger, W.H., and H.R.Thierstein (1979). On Phanerozoic mass extinctions, Naturwissenschaften 66, 46–47. Berger, W.H., and E.L.Winterer (1974). Plate stratigraphy and the fluctuating carbonate line, in Pelagic Sediments on Land and under the Sea. K.J.Hsü and H.Jenkyns, eds., Spec. Publ. Int. Assoc. Sedimentol. 1, pp. 11–48. Berger, W.H., R.F.Johnson, and J.S.Killingley (1977). “Unmixing” of the deep-sea record and the deglacial meltwater spike, Nature 269, 661–663. Berger, W.H., E.Vincent, and H.R.Thierstein (1981). The deep-sea record; Major steps in Cenozoic ocean evolution, in Symposium and Results of Deep-Sea Drilling, R.G.Douglas, J.Warme, and E.L. Winterer, eds., Soc. Econ. Paleontol. Mineral. Spec. Publ. 30. Berggren, W.A. (1972). Late Pliocene-Pleistocene glaciation, in Initial Reports of the Deep Sea Drilling Project 12, U.S. Government Printing Office, Washington, D.C., pp. 953–963. Berggren, W.A., and C.D.Hollister (1977). Plate tectonics and paleo-circulation—commotion in the ocean, Tectonophysics 38, 11–48. Broecker, W.S. (1973). Factors controlling CO2 content in the oceans and atmosphere, in Carbon and the Biosphere, G.M.Woodwell and E.V.Pecan, eds., AEC Symposium 30, pp. 32–50. Broecker, W.S., and J.van Donk (1970). Insolation changes, ice volumes, and the O18 record in deep-sea cores, Rev. Geophys. Space Phys. 8, 169–198. Burchardt, B. (1978). Oxygen isotope paleotemperatures from the Tertiary period in the North Sea area, Nature 275, 121–123. Burckle, L.H., and F.Akiba (1978). Implications of Late Neogene fresh water sediment in the Sea of Japan, Geology 6, 123–127. Dansgaard, W., S.J.Johnson, H.B.Clausen, and C.C.Langway (1971). Climatic record revealed by the Camp Century ice core, in Late. Cenozoic Glacial Ages, K.K.Turekian, ed., Yale U. Press, New Haven, Conn., pp. 37–56. Dean, W.E., J.V.Gardner, L.F.Jansa, P.Cepek, and E.Seibold (1978). Cyclic sedimentation along the continental margin of Northwest Africa, in Initial Reports of the Deep Sea Drilling Project 41, U.S. Government Printing Office, Washington, D.C., pp. 965–989. Delmas, R.J., J.M.Asccncio, and M.Legrand (1980). Polar ice evidence that atmospheric CO2 20,000 yr BP was 50% of present, Nature 284, 155–157. Donn, W.L., and D.M.Shaw (1977). Model of climate evolution based on continental drift and polar wandering, Geol. Soc. Am. Bull. 88, 390–396. Douglas, R.G., and S.M.Savin (1975). Oxygen and carbon isotope analyses of Tertiary and Cretaceous microfossils from the Shatsky Rise and other sites in the North Pacific Ocean, in Initial Reports of the Deep Sea Drilling Project 32, U.S. Government Printing Office, Washington, D.C., pp. 509–520. Emery, K.O. (1977). Structure and stratigraphy of divergent continental margins, in Continuing Education Course Note Series No. 5, Am. Assoc. Petrol. Geol., pp. B1–B20. Emiliani, C. (1955). Pleistocene temperatures, J. Geol. 63, 538–578. Emiliani, C. (1966). Paleotemperature analysis of Caribbean cores P6304–8 and P6304–9 and a generalized temperature curve for the past 425,000 years, J. Geol. 74, 109–126.

OCR for page 43
Climate in Earth History: Studies in Geophysics Emiliani, C. (1978). The cause of the ice ages, Earth Planet. Sci. Lett. 37, 349–352. Emiliani, C., S.Gartner, B.Lidz, K.Eldridge, D.K.Elvey, T.C. Huang, J.J.Stipp, and M.F.Swanson (1975). Paleoclimatological analysis of Late Quaternary cores from the Northeastern Gulf of Mexico, Science 189, 1083–1088. Fairbridge, R.W. (1961). Eustatic changes in sea-level, in Physics and Chemistry of the Earth, Vol. 4., L.H.Ahrens et al., eds., Pergamon, New York, pp. 99–185. Fischer, A.G., and M.A.Arthur (1977). Secular variations in the pelagic realm, Soc. Econ. Paleontol. Mineral. Spec. Publ. 25, 19–50. Flohn, H. (1974) Background of a geophysical model of the initiation of the next glaciation, Quat. Res. 4, 385–404. Gartner, S., and J.Keany (1978). The terminal Cretaceous event: A geologic problem with an oceanographic solution, Geology 6, 708– 712. Gates, W.L. (1976a). Modeling the ice-age climate, Science 191, 1138–1144. Gates, W.L. (1976b). The numerical simulation of ice-age climate with a global general circulation model, J. Atmos. Sci. 33, 1844–1873. Hamilton, W. (1968). Cenozoic climatic change and its cause, Meteorol. Monogr. 8, 128–133. Haq, B.U., I.Premoli-Silva, and G.P.Lohmann (1977). Calcareous plankton paleobiogeographic evidence for major climatic fluctuation in the early Cenozoic Atlantic Ocean, J. Geophys. Res. 82, 3861–3876. Hays, J.D., J.Imbrie, and N.J.Shackleton (1976). Variations in the Earth’s orbit, pacemaker of the ice ages. Science 194, 1121–1132. Hollin, J.T. (1972). Interglacial climate and Antarctic ice surges, Quat. Res. 2, 401–408. Holser, W.T. (1977). Castastrophic chemical events in the history of the ocean, Nature 267, 403–408. Hsü, K.J., M.B.Cita, and W.B.F.Ryan (1973). The origin of the Mediterranean evaporites, in Initial Reports of the Deep Sea Drilling Project 13, U.S. Government Printing Office, Washington, D.C., pp. 1203–1231. Hsü, K.J., L.Montadert, D.Bernoulli, M.B.Cita, A.Erickson, R.E. Garrison, R.B.Kidd, F.Melières, C.Müeller, and R.Wright (1977). History of the Mediterrranean salinity crisis, Nature 267, 399–403. Hummel, J., and R.A.Reck (1978). A global surface albedo model, Res. Publ., General Motors, GMR 2607, 47 pp. Hutson, W.H. (1980). Bioturbation of deep-sea sediments: Oxygen isotopes and stratigraphic uncertainty, Geology 8, 127–130. Imbrie, J., and J.Z.Imbrie (1980). Modeling the climatic response to orbital variations, Science 207, 943–953. Keigwin, L.D., Jr. (1978). Pliocene closing of the Isthmus of Panama based on biostratigraphic evidence from nearby Pacific Ocean and Caribbean Sea cores, Geology 6, 630–634. Keigwin, L.D., Jr. (1979). Late Cenozoic stable isotope stratigraphy and paleoceanography of DSDP sites from the east equatorial and north central Pacific Ocean, Earth Planet. Sci. Lett. 45, 361–382. Kennett, J.P., and N.J.Shackleton (1975). Laurentide ice sheet meltwater recorded in Gulf of Mexico deep-sea cores, Science 188, 147–150. Kennett, J.P., and N.J.Shackleton (1976). Oxygen isotopic evidence for the development of the psychrosphere 38 m.y. ago, Nature 260, 513–515. Kominz, M.A., and N.G.Pisias (1979). Pleistocene climate: Deterministic or stochastic? Science 204, 171–173. Kukla, G.J. (1975). Missing link between Milankovitch and climate, Nature 253, 600–603. Kullenberg, B. (1952). On the salinity of the water contained in marine sediments, Medd. Oceanogr. Inst. Göteborg 21, 1–38. Lorenz, E.N. (1968). Climatic determinism, Meteorol. Monogr. 8, 1–3. Manabe, S., and D.G.Hahn (1977). Simulation of the tropical climate of an Ice Age, J. Geophys. Res. 82, 3889–3911. Moore, T.C., N.G.Pisias, and G.R.Heath (1977). Climate changes and lags in Pacific carbonate preservation, sea surface temperature, and global ice volume, in The Fate of Fossil Fuel CO2 in the Oceans, N.R.Anderson and A.Malahoff, eds., Plenum, New York, pp. 145–165. Mörner. N.A. (1975). Eustatic amplitude variations and world glacial changes, Geology 3, 109–110. Peng, T.M., W.S.Broecker, G.Kipphut, and N.Shackleton (1977). Benthic mixing in deep sea cores as determined by 14C dating and its implications regarding climate stratigraphy and the fate of fossil fuel CO2, in The Fate of Fossil Fuel CO2 in the Oceans, N.R.Anderson and A.Malahoff, eds., Plenum, New York. pp. 355–373. Pisias, N.G. (1976). Late Quaternary variations in sedimentation rate in the Panama Basin and the identification of orbital frequencies in carbonate and opal deposition rates, Geol. Soc. Am. Mem. 145, 375–391. Reid, J.L. (1979). On the contribution of the Mediterrranean Sea outflow to the Norwegian-Greenland Sea, Deep Sea Res. 26A, 1199–1223. Rooth, C.G.H., C.Emiliani, and H.W.Poor (1978). Climate response to astronomical forcing, Earth Planet. Sci. Lett. 41, 387–394. Sarnthein, M. (1978). Sand deserts during glacial maximum and climatic optimums, Nature 171, 43–46. Savin, S.M. (1977). The history of the Earth’s surface temperature during the past 100 million years, Ann. Rev. Earth Planet. Sci. 5, 319–355. Savin, S.M., R.G.Douglas, and F.G.Stehli (1975). Tertiary marine paleotemperatures, Geol. Soc. Am. Bull. 86, 1499–1510. Shackleton, N.J. (1977). Carbon-13 in Uvigerina: Tropical rainforest history and the equatorial Pacific carbonate dissolution cycles, The Fate of Fossil Fuel CO2 in the Oceans, N.R.Anderson and A.Malahoff, eds., Plenum, New York, pp. 401–427. Shackleton, N.J., and J.P.Kennett (1975). Paleotemperature history of the Cenozoic and the initiation of Antarctic glaciation: Oxygen and carbon isotope analyses in DSDP Sites 277, 279, and 281, in Initial Reports of the Deep Sea Drilling Project 29, U.S. Government Printing Office, Washington, D.C., pp. 743–755. Shackleton, N.J., and N.D.Opdyke (1973). Oxygen isotope and paleomagnetic stratigraphy of equatorial Pacific core V28–233: Oxygen isotope temperatures and ice volumes on a 105 year and 106 year scale, Quat. Res. 3, 39–55. Shackleton, N.J., and N.D.Opdyke (1976). Oxygen-isotope and paleomagnetic stratigraphy of Pacific core V28–239, late Pliocene to latest Pleistocene, Geol. Soc. Am. Mem. 145, 449–464. Shackleton. N.J., and N.D.Opdyke (1977). Oxygen isotope and paleomagnetic evidence for early northern hemisphere glaciation, Nature 270, 216–219. Suarez, M.J., and I.M.Held (1979). The sensitivity of an energy balance climatic model to variations in the orbital parameters, J. Geophys. Res. 84, 4825–4836. Tappan, H. (1968). Primary production, isotopes, extinctions, and the atmosphere, Palaeogeogr. Palaeoclimatol. Palaeoecol. 4, 187–210. Thierstein, H.R., and W.H.Berger (1978). Injection events in ocean history, Nature 176, 461–466, van Andel, T.H., and T.C.Moore (1974). Cenozoic calcium carbonate distribution and calcite compensation depth in the central equatorial Pacific, Geology 2, 87–92. van Andel, T.H., J.Thiede, J.G.Sclater, and W.W.Hay (1977).

OCR for page 43
Climate in Earth History: Studies in Geophysics Depositional history of the South Atlantic ocean during the last 125 million years, J. Geol. 85, 651–698. van Couvering, J.A., W.A.Berggren, R.E.Drake, E.Aquirre, and G.H.Curtis (1976). The terminal Miocene event, Mar. Micropaleontol. 1, 263–286. van Donk, J. (1976). O18 record of the Atlantic Ocean for the entire Pleistocene Epoch, Geol. Soc. Am. Mem. 145, 147–163. Vincent, E., and W.H.Berger (1982). Planktonic foraminifera and their use in paleoceanography, in The Sea, Vol. 7, C.Emiliani, ed., Wiley-Interscience, New York. Vincent, E., J.S.Killingley, and W.H.Berger (1980). The Magnetic Epoch-6 Carbon Shift: A change in the ocean’s 13C/12C ratio 6.2 million years ago, Mar. Micropaleontol. 5, 185–203. Williams, D.F., R.C.Thunell, and J.P.Kennett (1978). Periodic fresh-water flooding and stagnation of the eastern Mediterranean Sea during the late Quaternary, Science 201, 147–149. Wilson, A.T. (1964). Origin of ice ages: An ice shelf theory for Pleistocene glaciation, Nature 201, 147–149. Wilson, A.T. (1969). The climatic effects of large-scale surges of ice sheets, Can, J. Earth Sci. 6, 911–918. Worthington, L.V. (1968). Genesis and evolution of water masses, Meteorol. Monogr. 8, 63–67.