The past 70 million years

Figure 1.8 shows the paleogeography at the end of the Cretaceous, 65 million years ago. The continent of Antarctica has approached the South Pole, and will continue to drift over the next 40 million years or so until it is more nearly centered on the pole. There is open water at the North Pole in the late Cretaceous, and the open Arctic Ocean continues throughout the subsequent time through the present. The modern continents of North and South America, Eurasia, and Africa are still early in their separation, leaving a narrow Atlantic ocean and a very broad Pacific. The continents will continue to drift apart as they approach their modern configuration; the Atlantic widens steadily throughout the span of time under discussion.

The record of benthic foram 618O in Fig. 1.9 provides a good overview of the climate evolution for the past 70 million years. Towards the beginning of the period, 618O is considerably lower than it is in the modern ocean. It reaches a minimum value of -0.1 around 51 million


l^fc. ]



f0 A

m mm

Figure 1.8: Position of the continents at the end of the Cretaceous, 65 million years ago. Continents are shown on a Mollweide map projection, with the North Pole at the top and the South Pole at the bottom. Light areas are continents while dark grey areas are ocean.

years ago. Independent geological evidence shows that there was no significant amount of ice sequestered in land glaciers until 36 million years ago, so the variations in ¿18O before this time can be interpreted as polar ocean temperature changes. Melting the present ice in Antarctica and Greenland would leave the ocean with a £18O of around -.7%orelative to VSMOW. Plugging this into the paleotemperature equation, we estimate a high-latitude ocean temperature of 15.5C. This estimate applies to the coldest seasonal temperature attained either in the Arctic ocean or in the waters surrounding Antarctica. This is well above the freezing point of sea water, and so we conclude that the oceans were ice-free year round, even in the Arctic and Antarctic regions which are very cold in the modern climate. We'll refer to this kind of climate state as a hothouse climate. The late Cretaceous polar temperatures were about 4C cooler than those at the peak warmth, but still warm enough to guarantee ice-free conditions.

Other lines of evidence also support warm Northern high-latitude conditions and above-freezing winter conditions. In the early twenty-first century, the first useful deep-time Arctic marine cores were recovered, and TeX-86 proxies applied to these cores indicated Arctic ocean up to 22C during the time of the spike marked PETM in Fig. 1.9, with temperatures in the range of 17-18C before and after. Fossil vegetation from Arctic land also supports temperatures in this range. Moreover, the abundant evidence that lemurs and crocodiles were able to survive in high Northern latitudes points toward mild winters, since these creatures cannot survive sub-freezing temperatures for any significant length of time.

While evidence for warm, ice-free polar conditions in the Eocene and late Cretaceous is unambigous, the nature of the tropical climate is somewhat problematic. Up until the year 2001, most paleoceanographers would have said, based on planktonic foraminiferal 18 O that the Cretaceous tropical sea surface was no more than two or three degrees warmer than present, and the Eocene tropical sea surface was no warmer than today, and might even have been cooler. This posed the paradox of the "low gradient" climate - how to warm up the planet enough to prevent polar ice, without frying the tropics. In 2001, it was discovered that most of the evidence for a cool tropics was spurious, having been affected by diagenetic alteration of sediments. The surviving non-altered data indicated a warmer tropics, but there was precious little data left after the

1 Composite Marine Core Benthic Foram 6 O

1 Composite Marine Core Benthic Foram 6 O

Antarctic Ice

Marine Sediment Core Record

Figure 1.9: A composite record of benthic foram ¿18O (vs. PDB) over the past 70 million years, based on the average of several marine sediment cores. Data is from Zachos et al. 2001.

Antarctic Ice mNH Ice

Figure 1.9: A composite record of benthic foram ¿18O (vs. PDB) over the past 70 million years, based on the average of several marine sediment cores. Data is from Zachos et al. 2001.

diagenetically contaminated data was discarded. Gradually, new sediments and new proxies have come to the fore, and the story continues to develop. Tex-86 proxies and Mg/Ca proxies now indicate tropical temperatures of up to 34C or even 37C in places, in the Eocene warm interval. Tropical surface temperatures were two or three degrees cooler in the late Cretaceous. This still gives a considerable reduction in the pole-equator temperature gradient as compared to the modern climate, but the problem is not as severe as it was. It is quite certain that the tropical sea surface temperatures could not have been as high as 40C, since the planktonic forams that are seen in the sedimentary record could not have survived at such high temperatures. A striking feature of the data is that tropical surface temperatures seem to remain fairly constant throughout the Eocene, though polar temperatures (as indicated by the benthic forams) decrease towards the Oligocene.

Returning to Fig. 1.9 we see that following the peak Eocene warmth of 51 million years ago, the climate commenced a long slide toward the icehouse climate characterizing the latter part of the record. Between the peak warmth and the beginning of the Oligocene 34 million years ago, the minimum polar temperature dropped by 8C as indicated by benthic ¿18O. At this point, small ephemeral ice sheets began to form on Antarctica, culminating in a more substantial glaciation of Antarctica that lasted until 26 million years ago, somewhat before the beginning of the Miocene. The Oligocene glaciation is visible as a pronounced ditch in the ¿18O. This first attempt at glaciating Antarctica didn't last however,since the climate recovered and returned to a period of generally cold Antarctic conditions with sea ice but with land ice sheets having volume below 50% of the present volume. This situation lasted until 15 million years ago, when the slide towards icehouse conditions resumed. Antarctic ice sheets grew again, but the Northern hemisphere land glaciation had not yet been initiated by this time. The first abundant evidence of sea ice, based on ice-rafted debris in polar sediment cores, appears at 14 million years ago. The increase in <518O in the next several million years is due to a combination of continued cooling and Antarctic ice sheet growth, culminating in the initiation of the major Northern Hemisphere ice sheets around 6.5 million years ago, as we enter the Pliocene period. The oxygen isotopes begin to show substantial fluctuations at this time, which grow in amplitude as one enters the Pleistocene. These fluctuations are due to waxing and waning of ice sheets, predominantly in the Northern Hemisphere - in other words, the coming and going of "ice ages." The nature of the fluctuations will be examined more closely in Section 1.10.

What accounts for the nature of the hothouse climate state, and for the subsequent hothouse/icehouse transition? This is another of the Big Questions. There is no support in astrophysics for solar variability of a magnitude and type that could explain the transition, so attention has settled primarily on long term fluctuations in the greenhouse gas content of the atmosphere. In an oxygenated atmosphere like that of the Phanerozoic, CO2 is the only known long-lived greenhouse gas that can build up to concentrations sufficient to cause climate changes of a magnitude comparable to those seen over the Phanerozoic; to get fluctuations of the requisite magnitude requires amplification of the direct CO2 effect by water vapor feedbacks, and cloud feedbacks can also substantially modify the response. Another thing that makes CO2 a good suspect for the role of primary agent in Phanerozoic climate evolution is the fact that it is a central participant in all aspects of the inorganic and organic carbon cycle, which offers many possible mechanisms whereby CO2 could evolve over the long term. There are a great many unresolved issues regarding the CO2 theory of Phanerozoic climate evolution, but a central part of testing the theory is to understand the way CO2 and water vapor act in concert to determine the temperature of a planet; the necessary estimates will be given in Chapter 4. One needs to understand how much CO2 it would take to account for Eocene warmth, before one can decide whether there are plausible geochemical mechanisms that could lead to the required concentration.

The greatest impediment to testing the CO2 theory of the hothouse/icehouse dichotomy is the difficulty of estimating past CO2 levels. There are various geochemical and fossil proxies that can be brought to bear on the problem. For example, algae preferentially take up 13C at a rate that depends somewhat on the CO2 concentration in the water in which they grow. Carbon isotopes in fossil soil carbonate also preserves information about past CO2, as does the density of pores (stomata) in fossil leaves. All estimates known to date are subject to considerable uncertainty. Nonetheless there is support for the idea that CO2 concentrations around 70 million years ago could have been 6-10 times modern pre-industrial values. The evidence points to a general decline of CO2 since that time,but there are also some indications that CO2 may have already attained quite low levels during some periods well before the Pliocene icehouse climate set in. This is a rapidly evolving subject, however, and nothing definitive can be said at present. What is certain is that there are known geochemical mechanisms associated with the Urey reaction and silicate weathering, which have the potential for causing changes in atmospheric CO2 of the required magnitude, and on a time scale consistent with the observations. These mechanisms will be discussed in Chapter 8. As with any climate problem, on Earth or elsewhere, uncertainties regarding cloud feedbacks complicate the problem of testing theories of climate response. It is not out of the question that part of the answer lies in modulation of cloud albedo by, say, biologically produced sulfur compounds that seed cloud formation. Further, if data should ultimately support the low-gradient picture of the Cretaceous and Eocene hothouse climates, some mechanism will be needed to keep the tropics from overheating while the poles are warmed by elevation of CO2. This, too, may involve clouds, or it may involve changes in ocean circulation. It has even been suggested, with considerable physical support, that increases in hurricane intensity in a warmer world could provoke ocean circulation changes of the sort required to provoke a low-gradient climate.

Figure 1.9 exhibits a dramatic climate event of considerable importance. The spike in <518O at the Paleocene/Eocene boundary (marked "PETM" in the figure, for Paleocene/Eocene Thermal Maximum) is not a glitch in the data. It represents a real, abrupt and massive transient warm event. The spike looks small in comparison to the range of isotopic variation over the past 70 million years, but in fact it represents the planet accomplishing two million years worth of warming in a warm spike that (on closer examination) sets in within under 10,000 years and has a duration of around 200,000 years. This isotopic excursion corresponds to a global warming of about 4C; other proxy records show that the warming had similar magnitude in the Arctic and at the Equator, and that it extended to the deep ocean. This climate event triggered a mass extinction of benthic species, probably due to a combination of warming, oxygen depletion, and ocean acidification. An important clue as to the cause of the warming is that the record of 613C from the same core (not shown) exhibits a major negative excursion at the same time, going from values of about +1.2%odown to about zero at the bottom of the excursion. This indicates a catastrophic release of large quantities of isotopically light carbon into the climate system, which presumably increased the atmospheric greenhouse effect and led to warming. One possibility is that the release came in the form of methane from destabilized clathrate ices in the ocean sediments; another is that the isotopically light carbon came from oxidation of suddenly exposed organic carbon pools on land, releasing large quantities of CO2. Based on analysis of the carbon isotope record, it has been estimated that 4000-6000 gigatonnes of carbon were released into the ocean-atmosphere system, if the release were from organic matter. This compares with 700 gigatonnes of carbon in the form of CO2 in the modern atmosphere. This would considerably enhance the atmospheric greenhouse effect, though the effect would largely wear off after a thousand years or so, over which time about 80% of the released carbon works its way into the ocean. It is far from clear that one can account for the observed magnitude and duration of PETM warming with the amount of carbon one has at ones' disposal. This is one of the Big Questions. We shall not answer it in this book, but the reader will be provided with the tools needed to assess the warming caused by various amounts of CO2 or methane, and also (in Chapter 8) a bit of insight about the partitioning of carbon between atmosphere and ocean. These are tools one must have at hand in order to evaluate any theory of the PETM.

The Cretaceous is closed by the impact of a large asteroid or comet (known generically as a bolide). This is known as the KT impact event (for "Cretaceous/Tertiary," Tertiary being an obsolete term for the period following the Cretaceous). This event has little or no expression in the isotope record shown in Fig. 1.9, and is instead identified by global presence of a layer of the element iridium. The impact crater has also been identified, which allows an estimate of the energy of the impactor. The KT impactor had effects of extreme consequence despite the lack of an expression in the oxygen isotope record. Notably, it was the dinosaur killer - though many other species went extinct at the same time. Examination of the carbon isotope record shows also that the ocean carbon cycle remained highly perturbed for millions of years. There are many Big Questions associated with the consequences of a bolide impact. What is the mechanism by which the impact causes extinction? Is it direct blast and heat, or some longer-lasting change in the climate? It has been estimated that the impactor released about 5 • 1023 Joules of energy. How does this energy compare to other energy sources in the climate system, and what effects should it have on the atmosphere and ocean? What are the broader effects of a bolide impact on climate, and how long do they last? Does the impact cause a warming (perhaps through release of greenhouse gases) or a cooling (through lofting of a dust and soot cloud)? Some of the climate questions will be taken up in Chapter 4.

The KT impact event is the archetype of impact events, which have been episodically important throughout Earth history. Similarly the Earth has experienced many other mass extinctions besides that at the KT boundary, not all of which are clearly associated with a bolide impact. All mass extinctions lead to Big Questions.




Cambrian. i i i i [


Sil. J—l_L

Devon-1 ian. 1


Jurassic .

Cretaceous i i i i 1 i i i

■ + ■

Figure 1.10: Known occurrences of land glaciation during the Phanerozoic. Tall shaded bars indicate periods of major glaciation in which ice reaches to latitudes of 50N or 50S, while short shaded bars indicate periods of more minor glaciation.

Was this article helpful?

0 0


  • oliviero
    What was the average temperature 70 million years ago?
    1 year ago
  • saku suutari
    What was the climate like 70 million years ago what?
    12 months ago
  • jolly
    How was the tempeture 75 million years ago?
    7 months ago

Post a comment