Cover Image

Not for Sale



View/Hide Left Panel

9
Long-Term Climatic Oscillations Recorded in Stratigraphy

ALFRED G.FISCHER

Princeton University

INTRODUCTION

The marine stratigraphic record reveals cyclic changes of various sorts, including periodic interruptions of deposition, change in the sedimentary constituents supplied, change in faunas and floras, and change in the nature of the depositional environment. Many of these changes are too general in character and in distribution to be attributed to local causes: they seem to reflect global changes in climate and their effects on the marine system.

These phenomena are discussed here in sequence of increasing period. Bedding phenomena visible at the outcrop level appear to correspond to climatic changes induced by the Earth’s orbital perturbations—in the 20,000–500,000-yr range—the same forces that drove the glacial advances and retreats of the Pleistocene. Broader phenomena that must generally be synthesized from regional or global data suggest a possible climatic cycle in the 30–36 million years (m.y.) realm. This in turn appears to ride on an extremely long cycle (not necessarily of fixed period) that brought on alternation of “icehouse” and “greenhouse” climates: in the last 700 m.y., the Earth seems to have completed two and started on a third of these cycles.

CHANGES IN THE 20,000–500,000-YEAR (MILANKOVITCH) RANGE

21,000- and 43,000-Year Cycles

Sediments when viewed at the level of a roadcut or hillside are characterized by stratification. This is generally attributed to random fluctuations in the supply of sediment to, and removal of sediment from, a given depositional site. Such processes might be expected to produce a fairly random aggregation of thicker and thinner strata, yet many sedimentary “formations” show rather striking uniformity of bedding thickness. This is particularly true of many limestones, in which thicker beds of biogenically formed carbonate alternate with thin interbeds of shale, recording a simple oscillation cycle, as recognized by Gilbert (1895, 1900), Schwarzacher (1975), Fischer (1980, 1981), and others.

Gilbert (1895) attributed rhythmic bedding in the Cretaceous of Colorado to climatic influence of the axial precession having a period of about 21,000 yr. These variations in insolation were first worked out quantitatively by Milankovitch (1941) and have been revised by Berger (1980).



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 97
Climate in Earth History: Studies in Geophysics 9 Long-Term Climatic Oscillations Recorded in Stratigraphy ALFRED G.FISCHER Princeton University INTRODUCTION The marine stratigraphic record reveals cyclic changes of various sorts, including periodic interruptions of deposition, change in the sedimentary constituents supplied, change in faunas and floras, and change in the nature of the depositional environment. Many of these changes are too general in character and in distribution to be attributed to local causes: they seem to reflect global changes in climate and their effects on the marine system. These phenomena are discussed here in sequence of increasing period. Bedding phenomena visible at the outcrop level appear to correspond to climatic changes induced by the Earth’s orbital perturbations—in the 20,000–500,000-yr range—the same forces that drove the glacial advances and retreats of the Pleistocene. Broader phenomena that must generally be synthesized from regional or global data suggest a possible climatic cycle in the 30–36 million years (m.y.) realm. This in turn appears to ride on an extremely long cycle (not necessarily of fixed period) that brought on alternation of “icehouse” and “greenhouse” climates: in the last 700 m.y., the Earth seems to have completed two and started on a third of these cycles. CHANGES IN THE 20,000–500,000-YEAR (MILANKOVITCH) RANGE 21,000- and 43,000-Year Cycles Sediments when viewed at the level of a roadcut or hillside are characterized by stratification. This is generally attributed to random fluctuations in the supply of sediment to, and removal of sediment from, a given depositional site. Such processes might be expected to produce a fairly random aggregation of thicker and thinner strata, yet many sedimentary “formations” show rather striking uniformity of bedding thickness. This is particularly true of many limestones, in which thicker beds of biogenically formed carbonate alternate with thin interbeds of shale, recording a simple oscillation cycle, as recognized by Gilbert (1895, 1900), Schwarzacher (1975), Fischer (1980, 1981), and others. Gilbert (1895) attributed rhythmic bedding in the Cretaceous of Colorado to climatic influence of the axial precession having a period of about 21,000 yr. These variations in insolation were first worked out quantitatively by Milankovitch (1941) and have been revised by Berger (1980).

OCR for page 97
Climate in Earth History: Studies in Geophysics The timing of such rhythms may be approached in two different ways, from “below” or from “above.” In varved sequences, in which beds are composed of presumed annual laminations, the duration of a bed in years should equal the number of varves within it. The alternative is to take a radiometrically well-dated interval, such as a stage or an epoch, and to divide its length by the number of beds found within it. The first of these methods was applied to the Green River Oil Shale (Eocene) by Bradley (1929). Bradley did not actually count the varves in a bed—he determined the mean thickness of varves from thin sections and the mean thickness of beds from measurements on outcrop and found the ratio to be 21,000:1. Whereas this work needs independent verification, it seems to have confirmed Gilbert’s hypothesis in principle. The second method has been applied to Cretaceous pelagic and hemipelagic limestones by Arthur (1979a) and by Fischer (1980, 1981). Various problems arise with this approach. One is that few rhythmic bedding sequences and continuous exposures span the length of a stage, so that is becomes necessary in most sequences to extrapolate. Another is that in rival time scales—Obradovich and Cobban (1975) versus Van Hinte (1976)—some stages differ by a factor of 3, so that it becomes necessary to average the results from several stages. A series of 11 Cretaceous sequences from Colorado, France, and Italy (Fischer 1980, 1981) yielded raw averages ranging from 10,000 to 87,000 yr per bed. The eight shorter ones yielded a mean of 17,125 on the Obradovich-Cobban scale and 26,375 on the Van Hinte scale, for a combined mean of 21,750. Of the remaining three, one is poorly dated and the other two seem to lie in the vicinity of 50,000 yr and might be related to the 43,000-yr cycle in obliquity (Milankovitch, 1941; Berger, 1980). Thus the existence of the 21,000-yr rhythm—and therewith of a precessional influence on Cretaceous sedimentary regimes—appears to be moderately well established. The case for a sedimentary record of the cycle in tilt, on the other hand, is not strong except in the deep-sea record (Arthur, 1979b). 100,000-Year Cycle Simple bedding rhythms of the type discussed above tend to occur in sets, and while there is considerable variation in the number per set, statistical averages out of any one sequence usually yield a mean number of about five (Schwarzacher, 1975). This ratio holds for the Precambrian-Cambrian boundary beds in Morocco (Monninger, 1979); for Carboniferous limestones in Ireland, Triassic limestones in the Alps, Triassic lake deposits in New Jersey, and Jurassic limestones in southern Germany (Schwarzacher, 1975); and for five of the eight Cretaceous sequences studied (Fischer, 1980, 1981). If the Cretaceous sequences cited above are of precessional origin, then the “Schwarzacher bundles,” which they compose, would seem to have a timing of about 100,000 yr. Furthermore, by analogy, it appears reasonable to interpret this bedding pattern, characteristic of various parts of the Phanerozoic, as a record of precessional cycles grouped into 100,000-yr sets. This 100,000-yr rhythm is the strongest of the glacial rhythm signals in the Pleistocene marine record (Hays et al., 1976), and is attributed there to the orbital quasi-rhythm in eccentricity (Milankovitch, 1941; Berger, 1980). It is extremely tempting to consider the Schwarzacher bundles as the product of the precession coupled with eccentricity. Indeed, the precession can influence climate only by way of orbital eccentricity, so that a bundling of precessional beds into larger sets is virtually demanded by theory. Cycles at the 500,000-Year Level An example of long rhythms in stratal sequences is provided by the Permo-Carboniferous megacycles of Kansas (Moore et al., 1951; Heckel, 1977), in which terrigenous deposits at base and top separate a marine sequence characterized by a perculiarly patterned alternation of shales and limestones. Some 25 of these megacycles characterize the 10 m.y. of Missourian-Virgilian time, a mean duration of 400,000 yr. In various other sequences, such as the Precambrian-Cambrian boundary beds in Morocco (Monninger, 1979), the Triassic lake deposits of New Jersey (Van Houten, 1964), and the Cretaceous and Eocene in central Italy, Schwarzacher bundles are in turn grouped into sets of four to six, representing about 0.5 m.y. each. This cycle too appears to have a match in orbital perturbations, namely in a longer cycle of eccentricity, which emerges from Berger’s (1980) calculations. This cycle has recently been recognized in the Pleistocene record by Briskin and Berggren (1975). Multiple Pathways of Expression Whereas the sediments in which these bedding patterns have been found are mainly limestone sequences, they include different depositional regimes, in which rhythmicity is induced by different factors. In the alpine Triassic, for example, the cause is a variation in sea level, leading to repeated emergence and submergence of carbonate banks (Fischer, 1964). In the late Cretaceous of central Italy, the setting is one of deep water throughout, and the rhythms reflect a change from carbonate deposition to clay deposition—either because the carbonate supply was reduced or because of carbonate dissolution on the seafloor (Arthur and Fischer, 1977). In the Cretaceous of Kansas, and in the Mid-Cretaceous (Aptian-Albian) of Italy and of the present ocean floor, some of the rhythmicity was produced by changes in oxygen content of bottom waters: the depositional sites oscillated between aerobic and anaerobic conditions (Arthur and Fischer, 1977; Arthur, 1979b; Fischer, 1980). In the Triassic lake deposits of the Newark rift (Van Houten, 1964), the rhythmicity resulted from changes in lake level and in the chemistry of the lake. The only common denominator for all of these changes is climate—climatic fluctuation so severe as to change sea level (presumably by growth of glaciers), the chemistry and behavior of the oceans, and the salinity of lakes.

OCR for page 97
Climate in Earth History: Studies in Geophysics Conclusions and Problems Regarding Cyclicity at the Milankovitch Level The data summarized above have led me to conclude that climatic oscillations driven by the Earth’s orbital perturbations have not been limited to the Pleistocene but have affected the Earth’s climates through Phanerozoic time—the last 600 m.y. We have barely begun to recognize their record in the sediments and are far from having adequate descriptions, let alone understanding. There are suggestions that the nature of this record has changed with time. In the Late Pleistocene record, for example, the eccentricity signal is strongest, the obliquity signal next, and the precessional signal weakest. In the Late Carboniferous cyclothems of Kansas—another glacial time—eccentricity cycles—in particular, the 400,000-yr cycle—seem again to dominate the picture. At nonglacial times the precessional signal seems the strongest. Are there definite time changes in the kinds of rhythms—signals of changes in orbital character or of changes in the Earth’s response to constant signals? Is there a solar factor in addition? We do not know the answers. What the record tells us is that different parts of the Earth recorded the climatic changes in different ways, and this in turn should serve to develop some understanding of the functioning of the Earth. Puzzling, for example, are the sharp sea-level changes suggested by the record for times generally thought to have been free of polar ice. Was there perhaps mountain glaciation on scales far beyond that of today? Such questions call for further studies. CHANGES AT THE 30-MILLION-YEAR AND 300-MILLION-YEAR LEVELS 30-Million-Year Cycle While historical geologists since Lyell have given lip service to the principle of uniformitarianism, in which the Earth is viewed as having developed in a gradual and steady manner, a majority of stratigraphers and tectonicists, going back to Cuvier and d’Orbigny, including Chamberlin, Grabau, and Umbgrove, have been impressed with the segmentation of geologic history into episodes. Some of these changes appear to be rhythmic, and one of the rhythms represented lies at the 30–36-m.y. level. Dorman (1968) suggested a 30-m.y. cycle in global temperatures, based on oxygen isotope analyses of Cenozoic mollusks. Damon (1971) analyzed the record of marine transgressions and regressions from the continents and concluded that Phanerozoic sea level rose and fell with a periodicity of 36±11 m.y. and that this bore some correspondence to periodicities in global mountain building and in regional plutonism. Fischer and Arthur (1977) suggested that the Mesozoic-Cenozoic part of Earth history is logically subdivided not into four periods as currently practiced but into seven, with a mean duration of 32 m.y., corresponding essentially to Grabau’s (1940) seven pulses: the Triassic, Liassic, “Jurassic,” Comanchean, Gulfian or “Cretaceous,” Paleogene, and Neogene. Each of these corresponds to an expansion of organic diversity in the pelagic marine realm (development of polytaxy) followed by a decline to an “oligotaxic” state. This pattern appears in global counts of coexisting genera and species as well as in the structure of marine communities, in which the polytaxic state is characterized by the development of superpredators, while the crash leading to oligotaxy is accompanied by the spread of opportunistic generalists (Figure 9.1). Fischer and Arthur tentatively recognized some reflections of this cycle in marine temperature regimes (Figure 9.2), in the oxygenation of the oceans, in carbon isotope ratios, in the ups and downs of the calcite compensation depth, in the development of submarine unconformities, and in other factors. Their overall conclusion was that oceanic structure and behavior have changed on a time scale of about 32 m.y., responding to some change in general climate: in polytaxic episodes the high latitudes were warmer and the temperature of the ocean mass as a whole was higher than during oligotaxy. However, these fluctuations ride on a much longer oscillation, which will be discussed below. During the last 100 m.y. the Earth has passed through three polytaxic episodes—that of the Late Cretaceous, that of the Eocene, and that of the Miocene—delimited by three oligotaxic ones—the Maastrichtian-Danian boundary crisis, the Oligocene crash, and the current decline. During this time we have experienced the “climatic deterioration” long recognized by the terrestrial paleobotanists (Dorf, 1970). Each episode of polytaxy has been merely a step back “up” in what has been a general “downward” trend toward colder high latitudes and colder ocean masses—a trend that finally culminated in the glacial episode in which we find ourselves now. The history of pelagic diversity in the Paleozoic offers some support for the existence of the 30-m.y. cycle through the Paleozoic, but the precision of Paleozoic data remains marginal. For that matter, the existence of this cycle in the Mesozoic-Cenozoic is still a matter of debate; Hallam (Chapter 17), for example, finds no convincing evidence for the postulated Mid-Jurassic break, and some of the polytoxic episodes recognized by Fischer and Arthur are split by minor reductions in faunal diversity. The general causes for the postulated 30-m.y. cycle remain uncertain. The pattern suggests that it was a minor modulation of the long (300-m.y.) greenhouse-icehouse cycle discussed below. I am therefore inclined to think that it, too, was engendered by changes (lesser ones) in atmospheric carbon dioxide pressure and that it, too, expresses imbalances between the rates at which carbon dioxide is added to the outer Earth by volcanism and withdrawn from it by weathering and sedimentation (see discussion below). Indeed, just as the long-range cycle is here attributed to first-order changes in volcanism and in sea level, so the 30-m.y. cycle seems to match shorter fluctuations in these factors (Damon, 1971). 300-Million-Year Level In the last 700 m.y. the Earth has undergone three major episodes of glaciation, during which ice caps not only covered one or both of the polar regions but extended at times more than

OCR for page 97
Climate in Earth History: Studies in Geophysics FIGURE 9.1 Pelagic diversity, superpredators, and blooms of opportunists over last 220 m.y. On left, changes in global diversity. Genera of ammonities (A) and species of planktic (globigerinacean) foraminifera (G), plotted logarithmically. Episodes of increasing diversity are separated by biotic crises of varying magnitude. Crises of moderate and high intensity recur at intervals of approximately 32 m.y. (shaded bands, defining seven cyclic episodes or pulses of diversification: polytaxy). These essentially coincide with transgressive pulses of Grabau (right). Each polytaxic pulse brought superpredators exceeding 10 m in length, a role that has been successively filled by ichthyosaurs, pliosaurs, mosasaurs, whales, and sharks, as shown in middle. Superpredators are known only from stages opposite the names. Mid-Triassic ichthyosaurs: Cymbospondylus and Shastasaurus; Toarcian ichthyosaur: Stenopterygius; Oxfordian pliosaur: Stretosaurus; Albian pliosaur: Kronosaurus; Campanian-Maastrichtian mosasaurs: Hainosaurus and others; Eocene whale: Basilosaurus; Mio-Pliocene shark: Carcharodon megalodon. Biotic crises are accompanied by local mass-occurrences of single pelagic species, rare in normal biotas. These are interpreted as blooms of opportunists and have been plotted in black circles. B, Braarudosphaera, a coccolithophorid; P, Pithonelta, a problematicum; E, Ethmodiscus rex, a giant diatom. From Fischer and Arthur (1977). halfway to the equator (see Chapter 6 for summary and literature). The times of these first-order glaciations (Figure 9.3) are Late Precambrian [about 750–650 m.y. ago (Ma)], Late Carboniferous-early Permian (340–255 Ma), and Pleistocene-Recent, having commenced about 2 Ma and stretching on into the unknown future. Another glacial episode, of lesser vigor and short duration, was associated with the end of the Ordovician, about 435 Ma. In between, the world seems to have lacked polar ice caps. The Paleogene of the Arctic region, for example, contains a warm temperate to subtropical forest assemblage including large trees, remains of amphibians, a wide range of reptiles, and, among the mammals, horses and monkeys (Koch, 1963; West and Dawson, 1978). Also, the marine molluscan fauna of western Greenland is distinctly subtropical (Kollmann, 1979). This paleobotanical evidence for a once very different world is corroborated by the marine record: The present ocean’s warm waters are confined to a thin surficial layer in the lower FIGURE 9.2 Paleotemperatures derived from oxygen isotope ratios in calcitic fossil skeletons, assuming constant oxygen isotope ratios in seawater. 1, belemnites, northwestern Europe, uncorrected lat. 45–55°; 2, planktonic foraminifera, South Atlantic, uncorrected lat, 30–32°; 3, planktonic foraminifera, South Pacific, uncorrected lat. 47–52°; 4, planktonic foraminifera, tropical Pacific, uncorrected lat. 7–19°. From Fischer and Arthur (1977).

OCR for page 97
Climate in Earth History: Studies in Geophysics FIGURE 9.3 Relation of inferred climates to secular patterns in volcanism, sea level, and organic diversity. Volcanism: emplacement of plutons in North America, after Engel and Engel (1964). Sea level: A, first-order eustatic curve of Vail et al. (1977); B, compromise between North American and Russian records, constructed from Hallam (1977); the scale at left refers to this curve. Biotic record; N, Stehli et al.’s (1969) curve of disappearance of animal families; C, net gain-and-loss curve of Cutbill and Funnel (1967), overlap shaded. Inferred climatic states from Fischer (1981); minor oscillations (which may bring about growth of ice sheets, shaded) after Fischer and Arthur (1977). Diagram modified from Fischer (1981). and middle latitudes. Bottom waters are close to the freezing point throughout the oceans, and the mean temperature of the oceanic water masses lies at about 3°C. While we do not have reliable measurements of paleotemperatures—oceanic or otherwise—for the ice ages of the Paleozoic and Precambrian, it seems likely that their oceans had a temperature and structure rather similar to that of today. In contrast, studies of oxygen isotopes from Cretaceous oceanic deposits (e.g., Douglas and Savin, 1975) show that the temperatures of the bulk of the Cretaceous ocean masses lay in the vicinity of 15°C, i.e., that the mean temperature of the oceans averaged some 10° above that of the present ones. This suggests that bottom waters were not formed as today, by chilled polar surface waters mixed with a strong dash of meltwater from the polar ice caps. Several alternatives appear possible. Either lightly cooled surface waters descended in the high latitudes, to form a bottom water of simple origin, or, alternatively, bottom waters generated in the paraequatorial dry belts (horse latitudes) to form a deep warm layer, as yet not sampled. A likely compromise is that bottom waters arose from a mixture of both of these processes. In short, it appears that the Cretaceous and Paleogene periods had a markedly different climate and ocean: tropical temperatures were much as they are today, while the temperature gradients toward the poles (and, in the ocean, toward the bottom) were very much lower. This implies a more uniform distribution of energy received from the Sun. The most appealing mechanism for this is that of a climatic “greenhouse” (Budyko, 1977; Manabe and Wetherald, 1975), in which an enrichment of atmospheric carbon dioxide inhibits radiation losses of energy into space. Temperatures in the tropical seas would not rise appreciably, because of increased evaporation, but the water content of the atmosphere would increase, and the transport of heat to the higher latitudes would occur largely as latent heat of evaporation, released in the high latitudes by heavy rainfall. Fischer (1981) has contrasted these climatic states of the Earth as the “icehouse state” and the “greenhouse state” (Figure 9.3). I view the history of the last 700 m.y. as a passage through two great icehouse-greenhouse cycles and into the beginning of a third. Associated phenomena include first-order changes in sea level, mean ocean temperatures, and oceanic aeration, possibly linked to changes in volcanicity and plate motions, as explained below. The transitions from one state to the other appear to be punctuated by the four major biotic crises. A special explanation, in this scheme, has to be invented for the glaciation and biotic crisis at the end of Ordovician time. Associated Phenomena Volcanism Long-term secular variations in global volcanism are not easily apprehended. Of the various volcanic processes, those associated with the generation of the oceanic lithosphere rank largest, but their pre-Jurassic record is lost by the recycling of the oceanic crust back into the mantle. On the continents, the great andesitic volcanic piles formed at convergent plate boundaries are largely lost to erosion. The best record of long-term volcanism is probably that of the granitic plutons that form the substructure of these belts (Fischer, 1981). If we may take the emplacement of such plutons as indicative of volcanic activity in general, and if North America is representative, then a plot of the rate of “granite” emplacement in North America, through geologic time, should serve as an index to worldwide volcanicity. Such a plot, by Engel and Engel (1964) is shown at the bottom of Figure 9.3. It is essentially bimodal, showing one broad (and bifid) peak that matches the inferred greenhouse state of the Ordovician-Devonian and another, sharper peak that matches the Mesozoic greenhouse interval. Sea-Level Change The eustatic curves by Vail et al. (1977) show the fluctuations of sea level relative to the continents. While this demonstrates a lively history of sea-level oscillations, its most generalized version—Vail’s first-order curve or Hallam’s (1977) curve (here modified by Fischer)—coincide with the inferred succession of icehouse and greenhouse states: the three major glaciations occur within the three lows of the curve, while the greenhouse states correspond to the highs. While glaciation itself drives sea level through oscillations because of withdrawal of water from the hydrosphere into ice

OCR for page 97
Climate in Earth History: Studies in Geophysics caps, the longer persistence of these first-order lows shows them to have been a condition for the growth of ice sheets. The latter merely contributed feedback. For the reasons why high sea levels should coincide with greenhouse states, see below. Oceanic Aeration The ocean of our day is moderately well oxygenated throughout, so that animal life, dependent on aerobic respiration, is possible on almost all bottoms and throughout almost all of the water mass. Exceptions are encountered only in localized areas such as the Black Sea, the Cariaco Trench, and the Santa Barbara Basin and, seasonally, in certain tropical belts of upwelling. Because of this, bottom sediments are almost everywhere stripped of organic matter by scavengers and bacteria and are plowed and mixed in the process (Fischer and Arthur, 1977). Accumulation of petroleum source beds is at a minimum. In contrast, during much of the past, black, organic-rich, finely laminated sediment was widely deposited on marine bottoms deprived of free oxygen. In my experience, widespread anaerobism of this sort peaked in the Ordovician to Devonian interval and again in the Jurassic and Cretaceous (Fischer and Arthur, 1977; Jenkyns, 1980). These times correspond to the greenhouse states. Berry and Wilde’s (1979) alternative explanation, that the Paleozoic black shales are holdovers from the poorly oxygenated atmosphere of the Precambrian, fails to explain the earlier (Lower Cambrian) spread of highly oxygenated (red) fossiliferous marine sediments, as well as the recurrence of anoxia in the Mesozoic. Within these episodes, there was a waxing and waning of anoxia, on what Fischer and Arthur (1977) interpreted as a 30-m.y. cycle, as well as on the yet smaller scales of the Milankovitch cycles. The causes for these variations in oceanic aeration are not resolved. We may think of the ocean in analogy to an organism that digests food: In anaerobic periods, the sea has indigestion. This may be brought on in one of two ways: either by a surfeit of organic matter supplied to it, which overwhelms its digestive capacity, or by a breakdown in the digestive system itself. Fischer and Arthur (1977) ascribed mainly to the latter cause and linked aeration to the cooling of oceans, which (a) permits a given volume of water to absorb more oxygen while in contact with the atmosphere and (b) favors a vigorous marine circulation, shortening the residence time of water in the depths, between times of recharge at the surface. This explanation thus offers indirect evidence for a warming of seas between times of massive glaciation. On the other hand, the greenhouse state is likely to have increased organic production on the lands, owing to more plant growth in high latitudes, to more widespread rainfall, and to the greater availability of carbon dioxide. It seems likely that during greenhouse states the supply of organic matter from the lands to the oceans may have been greater than it is today. I have therefore come to believe that both factors worked together to promote oceanic anaerobism. Punctuation by Biotic Crises Two curves at the top of Figure 9.3 illustrate changes in faunal diversity revealed by the fossil record. N is Stehli et al.’s (1969) compilation of the disappearance of animal families; C is Cutbill and Funnel’s (1967) analysis, depicting net gain and loss in invertebrate diversity. Six times of large-scale disappearance of taxa—first-order biotic crises, numbered 1–6—stand out in both curves. Of these, numbers 1, 3, and 5 coincide rather well with the boundaries between the climatic states suggested in the middle of the diagram. Number 2 coincides with the brief plunge into glaciation that occurred in the middle of the Early Paleozoic greenhouse. In a previous paper (Fischer, 1981) I sought to relate crises 4 and 6 to the climatic transitions as well, but this does not appear feasible. Both occur too soon, 4 before the breakup of Pangea, 6 before the greenhouse came to an end, as evidenced by the warm nature of the Arctic in Paleogene times (see above). Also, there is now strong evidence for an extraterrestrial origin of crisis 6 (Alvarez et al., 1980). The transition to the Late Cenozoic icehouse state is marked, instead, by the Late Eocene-Oligocene biotic crisis, which does not show on Stehli et al.’s curve, is a second-order crisis in Cutbill’s and Funnell’s compilation, and was strongly developed in the marine realm (Fischer and Arthur, 1977). Possible Causes Elsewhere (Fischer, 1981), I have suggested that the fluctuations in atmospheric carbon dioxide content result from fluctuations in the rate of supply (from volcanism) and in the rate of withdrawal (by the linked processes of weathering and sedimentation) (Budyko, 1977; Holland, 1978). Both are linked to a major cyclic process—mantle convection. The hypothesis can only be outlined here. While there is much uncertainty about the manner in which the lithospheric plates are driven, thermal convection of the mantle is generally taken to be the ultimate cause (e.g., Morgan, 1972). There is no reason to suppose that this process runs at a constant rate, and Fischer (1981) has proposed that the historical pattern in which episodes of continental dispersion and episodes of continental aggregation succeed each other results from a cyclicity in mantle convection in the following manner. In one phase, the mantle is in a quiescent state, having few and slowly turning cells, which tend to sweep the continental masses together into one pangea. As a result of few cells, the total length of midoceanic ridge is relatively short; because of slow convection and slow spreading rates, the ridge is relatively narrow. It follows (Russell, 1968; Hays and Pitman, 1973) that the displacement of waters from the ocean basins by the ridge is small and that the continents stand high and dry. Contributions of carbon dioxide from the interior to the ocean-atmosphere system are at a low, because both basaltic volcanism in rift zones (which brings carbon dioxide from the mantle) and andesitic volcanism from subduction zones (which recycle lithospheric carbon back to the atmosphere) are minimal. At the same time, the large area of the lands implies that uptake of carbon dioxide by weathering is at a maximum. As a result, a high atmospheric content of carbon dioxide, inherited from a former state, cannot be maintained. Carbon dioxide pressure will drop until the rate of weathering slows and the rate of carbon dioxide withdrawal approaches the rate of volcanic addition. This low balance results in the development of the ice-house state. In this, the growth of a glacial armor over the lands

OCR for page 97
Climate in Earth History: Studies in Geophysics further reduces weathering, providing a negative feedback that may help to explain why the Earth has never turned into a complete iceball. Mantle convection increases, by development of more cells and of more vigorous convection. New rifts develop, and old ones spread more rapidly. As marine ridges grow in length as well as in width, and as continents spread out by rifting, seawater is displaced, flooding the continents. The increased volcanism in rift belts and in belts of plate convergence raises the output of carbon dioxide to the atmosphere-hydrosphere system. At the same time, flooding of the continents cuts carbon dioxide losses to the lithosphere by weathering. The net result is that atmospheric CO2 pressure must rise, until the weathering rates, increased thereby, once again withdraw carbon dioxide at rates matching the volcanic addition. This high balance results in the greenhouse state. The late Ordovician ice age, coming in the middle of a greenhouse episode, at a time of sea-level highs, and associated with a major biotic crisis, seems altogether exceptional and asks for special explanation. One possibility that comes to mind is that of a greenhouse that overshot, producing a cloud cover so dense as to reflect enough of the solar radiation to cool the Earth. The alternative is to find a geologically transient sink for carbon dioxide. Conclusions and Problems Global climates have alternated between states susceptible to widespread glaciation (icehouse states) and greenhouse conditions. Two such cycles have been completed in the 600 m.y. of Phanerozoic time. The reasons for them are not firmly established. While several authors (cf., Pearson, 1978) have suggested a tie to changes in insolation, related to the cycle, of galactic rotation, an apparent correlation with sea-level changes and volcanicity suggests an internal cause. This is here sought in hypothetical cycles of mantle convection, which drive sea levels and atmospheric carbon dioxide content by independent pathways linked by a feedback mechanism (weathering). Whereas the Phanerozoic record suggests a length of about 300 m.y. per cycle, a rigorous periodicity throughout Earth history is not implied, inasmuch as mantle behavior must change in a cooling Earth. Nevertheless, it seems likely that we are in the early part of an icehouse state. Riding on this long cycle are a family of smaller climatic fluctuations, of which one seems to have a periodicity of perhaps slightly more than 30 m.y. In this one, too, the climatic effect may depend on changes in carbon dioxide, but the mechanisms remain obscure. CONCLUSIONS In summary, I suggest that the stratigraphic record holds evidence of a wide range of global changes in climatic state. Largest among these are the 150-m.y.(?) alternations between the major greenhouse and icehouse states. We know only the latter, and the traditional attempts to reconstruct the Mesozoic or the mid-Paleozoic world along strictly actualistic lines are grossly inadequate. Upon this great cycle rides a smaller one, having a period somewhere around 30 m.y. to 36 m.y., which mimics the large one on a smaller scale and which is recorded in its effects on life. The oligotaxic times that happen to coincide with the turnover in the large cycle, from its greenhouse phase to its ice-house phase and vice versa, are particularly pronounced as some of the world’s great biotic crises. Smaller pre-Pleistocene climatic oscillations are seen at the 500,000-yr level, at the 100,000-yr level, at the 50,000-yr level, and at the 20,000-yr level, in round numbers. These appear to match similar periods in ice flux within the Pleistocene, which have been attributed to climatic effects of the Earth’s orbital perturbations. These rhythmic events are recorded in a wide variety of sedimentary settings and record a multitude of pathways by means of which climatic change became expressed in sediments. We have only begun to recognize them and are far from any understanding of them. In the normal course of development, we could expect to slide more deeply into the icehouse state for some millions of years to come, with continued gradual loss of species. The burning of fossil fuels may instead provide a brief brush with the greenhouse state within a generation. That would be a brief passage only, limited by the amounts of fossil fuels available. The effects on climate, I must leave to others more qualified. However, I believe that a full greenhouse of the kinds that existed in the mid-Paleozoic and Mesozoic would require the complete melting of the ice caps and the warming of the oceans as a whole. That would produce a major biotic crisis of the sort that brought about the partial or complete collapse of some of the world’s organic communities in Devonian, Triassic, and Oligocene time. That event seems unlikely to occur for another 70 m.y., but even a brief brush with greenhouse conditions may upset the accustomed structure and behavior of atmosphere and oceans. It might thus have marked effects on the biosphere and on human life and history. REFERENCES Alvarez, L.W., W.Alvarez, F.Asaro, and W.V.Michel (1980). Extraterrestrial cause of the Cretaceous-Tertiary extinctions, Science 208, 1095–1108. Arthur, M.A. (1979a). Sedimentologic and geochemical studies of Cretaceous and Paleogene pelagic sedimentary rocks; The Gubbio sequence, Dissertation, Princeton U., Part I, 174 pp. Arthur, M.A. (1979b). North Atlantic Cretaceous black shales: The record at Site 398 and a brief comparison with other occurrences, in Initial Reports of the Deep Sea Drilling Project Vol. 47, pt. 2, U.S. Government Printing Office, Washington, D.C. Arthur, M.A., and A.G.Fischer (1977). Upper Cretaceous-Paleocene magnetic stratigraphy at Gubbio, Italy. I. Lithostratigraphy and sedimentology, Geol. Soc. Am. Bull. 88, 367–371. Berger, A.L. (1980). The Milankovitch astronomical theory of paleoclimates: A modern review, Vistas in Astronomy 24, 103–122. Berry, W.B.N., and P.Wilde (1979). Progressive ventilation of the oceans—an explanation for the distribution of the Lower Paleozoic black shales, Am. J. Sci. 278, 257–275. Bradley, W.H. (1929). The varves and climate of the Green River epoch, U.S. Geol. Surv. Prof. Pap. 645, 108 pp. Briskin, M., and W.A.Berggren (1975). Pleistocene stratigraphy and

OCR for page 97
Climate in Earth History: Studies in Geophysics quantitative oceanography of tropical core V-16 205, Micropaleontol. Spec. Publ. 1, 167. Budyko, M.I. (1977). Climatic Changes, American Geophysical Union, Washington, D.C., 261 pp. Cutbill, J.L., and B.M.Funnel (1967). Computer analysis of the fossil record, in The Fossil Record, W.B.Harlan et al., eds., Geol. Soc. London, pp. 791–820. Damon, P.E. (1971). The relationship between late Cenozoic volcanism and tectonism and orogenic-epeirogenic periodicity, in Late Cenozoic Glacial Ages, K.K.Turekian, ed., Yale U. Press, pp. 15–36. Dorf, E. (1970). Paleobotanical evidence of Mesozoic and Cenozoic climatic changes, in Proceeding of the North American Paleontological Convention, Vol. 2, E.I.Yochelson, ed., Allen and Unwin, London, pp. 323–346. Dorman, F.H. (1968). Some Australian oxygen isotope temperatures and a theory for a 30-million-year world temperature cycle, J. Geol. 76, 297–313. Douglas, R.G., and S.M.Savin (1975). Oxygen and carbon isotope analyses of Cretaceous and Tertiary microfossils from Shatsky Rise and other sites in the North Pacific Ocean, in Initial Reports of Deep Sea Drilling Project, Vol. 32, U.S. Government Printing Office, Washington, D.C., pp. 509–521. Engel, A.E.J., and C.G.Engel (1964). Continental accretion and the evolution of North America, in Advancing Frontiers in Geology and Geophysics, A.P.Subramaniam and S.Balakrishna, eds., Indian Geophysical Union, pp. 17–37e. Fischer, A.G. (1964). The Lofer cyclothems of the alpine Triassic, Kansas State Geol. Surv. Bull. 169, Vol. 1, pp. 107–150. Fischer, A.G. (1980). Gilbert—bedding rhythms and geochronology, in The Scientific Ideas of G.K.Gilbert, E.I.Yochelson, ed., Geol. Soc. Am. Spec. Pap. 183, pp. 93–104. Fischer, A.G. (1981). Climatic oscillations in the biosphere, in Biotic Crises in Ecological and Evolutionary Time, M.Nitecki, ed., Academic, New York, pp. 103–131. Fischer, A.G., and M.A.Arthur (1977). Secular variations in the pelagic realm, in Deep Water Carbonate Environments, H.E. Cook and P.Enos, eds., Soc. Econ. Paleontol. Mineral. Spec. Publ. 25, pp. 18–50. Gilbert, G.K. (1895). Sedimentary measurement of geologic time, J. Geol. 3, 121–127. Gilbert, G.K. (1900). Rhythms and geologic time, Am. Assoc. Adv. Sci. Proc. 49, 1–19. Grabau, A.W. (1940). The Rhythm of the Ages, Henry Vetch Publ. Co., Peking, 561 pp. Hallam, A. (1977). Secular changes in marine inundation of USSR and North America through the Phanerozoic, Nature 269, 769–772. Hays, J.D., and W.C.Pitman III (1973). Lithospheric motion, sea level changes and climatic and ecological consequences, Nature 246, 18–22. 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. Heckel P.H. (1977). Origin of phosphatic black shale facies in Pennsylvanian cyclothems of midcontinent North America, Am. Assoc. Petrol. Geol. Bull. 61, 1045–1068. Holland, H.D. (1978). The Chemistry of Oceans and Atmospheres, Wiley, New York, 351 pp. Jenkyns, H.C. (1980). Cretaceous anoxic events: From continents to oceans, J. Geol. Soc. 137, 171–188. Koch, D.E. (1963). Fossil plants from the lower Paleocene of Agatdalen (Angmartusuk) area, central Nuqssuaq Peninsula, northwest Greenland, Medd. Groenl. 172, 1–120. Kollmann, H. (1979). Distribution patterns and evolution of gastropods around the Cretaceous-Tertiary boundary, in Cretaceous-Tertiary Boundary Events, Vol. II Proceedings, W.K.Christensen and T.Birkelund, eds., U. of Copenhagen, Copenhagen, Denmark. Manabe, S., and R.T.Wetherald (1975). The effect of doubling the CO2 concentration on the climate of a general circulation model, J. Atmos. Sci. 32, 3–15. Milankovitch, M. (1941). Kanon der Erdbestrahlung und seine Anwendung auf das Eiszeitenproblem, Ed. Spec. Acad. Roy. Serbe, Belgrade, 633 pp., English translation, U.S. Department of Commerce. Monninger, W. (1979). The section of Tiout (Precambrian/Cambrian boundary beds, Anti-Atlas, Morocco); An environmental model, Arb. Palaeontol. Inst. Wuerzburg (Germany), 289 pp. Moore, H.C., J.C.Frye, J.M.Jewett, W.Lee, and H.O’Connor (1951). The Kansas rock column, Kansas Geol. Surv. Bull. 89, 132 pp. Morgan, W.J. (1972). Plate motions and deep mantle convection, in Studies in Earth and Space Sciences, R.Shagam, ed., Geol. Soc. Am. Mem. 132, 7–22. Obradovich, J.D., and W.A.Cobban (1975). A time scale for the Late Cretaceous of the Western Interior of North America, in The Cretaceous System in the Western Interior of North America, W.G. E.Caldwell, ed., Geol. Assoc. Canada Spec. Pap. 13, pp. 31–54. Pearson, R. (1978). Climate and Evolution, Academic, New York, 274 pp. Russell, K.L. (1968). Oceanic ridges and eustatic changes in sea level, Nature 218, 861–862. Schwarzacher. W. (1975). Sedimentation Models and Quantitative Stratigraphy, Developments in Stratigraphy, Vol. 19, Elsevier, New York, 377 pp. Stehli, F.G., R.G.Douglas, and N.D.Newell (1969). Generation and maintenance of gradients in toxonomic diversity, Science 164, 947–949. Vail, P.R., R.M.Mitchum Jr., and S.Thompson (1977). Seismic stratigraphy and global changes in sea level, part 4, in Seismic Stratigraphy, C.E.Peyton, ed., Am. Assoc. Petrol. Geol. Mem. 26, pp. 83–97. Van Hinte, J.E. (1976). A Cretaceous time scale, Am. Assoc. Petrol. Geol. Bull. 60, 498–516. Van Houten, F.B. (1964). Cyclic lacustrine sedimentation, Upper Triassic Lockatong Formation, central New Jersey and adjacent Pennsylvania, Kansas Geol. Surv. Bull. 169, 497–531. West, R.M., and M.R.Dawson (1978). Vertebrate paleontology and Cenozoic history of the North Atlantic region, Polarforschung 48, 103–119.