Stable Isotopes in Climatic Reconstructions
Case Western Reserve University
The calcium carbonate-water oxygen isotope geothermometer has become the most widely applied quantitative tool for estimating ancient ocean temperatures and has been applied increasingly often to studies of paleoclimate and paleo-oceanography. For many years the greatest impact of isotope paleoclimatology on geologic thinking was in studies of the Quaternary period. More recently, analyses of marine carbonates of Tertiary and late Mesozoic age have permitted refinement of our knowledge of marine temperatures during the past 100 million years (m.y.). This quantification of prePleistocene marine climates has been especially timely, as it evolved when our growing understanding of plate motions and resultant changing oceanic geometry, and of sea levels, has encouraged the development of theories to explain the causes of climatic change. Sufficient data soon will be available to provide boundary conditions for mathematical models of atmospheric circulation, at least during Neogene time.
With progressively older sediments, the occurrence of carbonate material suitably preserved for oxygen isotope paleotemperature studies becomes increasingly scarce. The detailed and quantitative paleotemperature records that have been reconstructed for Tertiary and late Cretaceous time cannot be envisaged for pre-Cretaceous time with samples now available or likely to become available. In addition, paleoclimatic information from nonisotopic sources becomes more difficult to obtain as we proceed backward through the geologic record and knowledge of climatic history becomes correspondingly poorer. Hence, while the kinds of paleoclimatic information obtainable using isotopic techniques becomes increasingly imprecise as we proceed back through time, the important questions about the climate of those earlier times can be meaningfully answered with less precisely interpretable data. For these earlier times, other isotope paleoclimatic techniques, in addition to the calcium carbonate-water paleothermometer, become useful. Most notable of these so far has been the paleoclimatic interpretation of the oxygen (and hydrogen) isotopic compositions of cherts. Some isotopic methods can provide information about terrestrial climates. In this paper the calcium carbonate-water paleothermometer and the climatic record it has yielded are reviewed. Other isotopic techniques that have provided, or that have the potential to provide, useful paleoclimatic data are also discussed.
THE CALCIUM CARBONATE-WATER PALEOTHERMOMETER AND MARINE PALEOCLIMATES
The use of the calcium carbonate-water isotope paleothermometer has been reviewed many times since the technique was proposed by Urey (1947) and developed and applied by Epstein et al. (1951) and Urey et al. (1951). Critical reviews and discussions of various aspects of the method include those by Craig (1965), Bowen (1966), Teis and Naidin (1973), Savin and Stehli (1974), Hecht (1976), Hudson (1977), Savin (1977), and Berger (1979). The basic principles are straightforward. If calcium carbonate is deposited in isotopic equilibrium with seawater the difference between the 18O/16O ratio of the carbonate and that of the seawater is strictly a function of temperature. If the temperature dependence of that difference has been calibrated, if the 18O/16O ratio of the seawater can be estimated, and if the 18O/16O ratio of the carbonate has not been altered since formation, the temperature of carbonate deposition can be calculated. It is this calculated temperature that is referred to as an isotopic temperature.
In practice, uncertainties are encountered when the isotopic paleotemperature method is applied to the study of marine carbonates. These uncertainties, discussed in the reviews mentioned above, lead to ambiguities in relating isotopic temperatures to climatically meaningful temperatures at specific localities and depths within the water column. Somewhat less uncertainty is entailed in estimating the water-temperature changes than in estimating absolute values of water temperature.
The most successful applications of isotope paleoclimatology have been in the study of foraminifera from deep-sea sediments. An assumption in most of these studies has been that planktonic foraminifera deposit their tests in isotopic equilibrium with seawater. Although there is some indication that this is not always completely true (van Donk, 1970; Shackleton et al., 1973; Grazzini, 1976; Williams et al., 1977), it seems a sufficiently close approximation to reality that for most purposes it can be taken as if true. Most taxa of benthic foraminifera clearly show departures from isotopic equilibrium (Duplessy et al., 1970; Woodruff et al., 1980). Departures may be as great as 1 per mil or more. Fortunately, departures from equilibrium may be taken as approximately (but not exactly) constant for a species, and isotopic compositions may therefore be interpreted in terms of temperatures.
The 18O/16O ratio of water in which foraminifera grew must always be estimated. To a first approximation the open oceans can be taken to be well mixed and, hence, their isotopic compositions to be constant through at least Phanerozoic time and through space. Whereas this sort of approximation may be sufficient (and unavoidable) for early Cretaceous and older paleotemperature studies, it is woefully inadequate in the investigation of Tertiary and Quaternary climates where the important problems require more accurate paleoclimatic knowledge. The oxygen isotopic composition of the hydrosphere has probably remained constant through much of, at least Phanerozoic, time. However, that of the oceans has varied in response to the formation and disappearance of 16O-rich continental icecaps. The extent to which this has affected seawater 18O/16O ratios during Pleistocene ice advances and retreats has been a matter of controversy for many years (Savin and Yeh, 1981). As this paper does not deal with Pleistocene climates, this controversy can be largely ignored. Uncertainty in the magnitude and isotopic composition of the Antarctic icecap in middle Miocene and later times does create ambiguities when interpreting the middle and late Miocene and Pliocene isotopic record in terms of temperature changes. (As discussed below, most, but not all, investigators involved with the isotopic record would argue that, prior to middle Miocene time, late Mesozoic and Cenozoic continental ice was never so extensive as to introduce ambiguities into the interpretation of the isotopic data.) Isotopic paleotemperature data for preTertiary glaciations are so sparse that discussion of uncertainties resulting from glacially induced variations in the isotopic composition of the oceans is unwarranted.
Locally, the 18O/16O ratio of surface seawater varies in response to evaporation (increased 18O/16O), precipitation (decreased 18O/16O), and freshwater runoff (decreased 18O/16O). These variations in seawater 18O/16O can cause errors in estimated water temperatures of a few degrees if they are not properly taken into account. Until now, this problem has frequently been largely ignored or dealt with in rudimentary fashion, by assuming that local variations in the past have been analogous to those of today. The time appears to be approaching when local variations in Neogene seawater 18O/16O can be estimated from paleo-oceanographic (including isotopic) data, and estimates of water temperatures can be refined.
Well-preserved calcium carbonate suitable for isotopic analysis is common in Neogene deep-sea sediments, but alteration rendering samples unsuitable becomes progressively more common in older sediments. Suitably preserved samples of any age are rare (but do exist) in rocks exposed on the continents. Additional work is needed to develop techniques for the recognition of minor amounts of alteration that affect but do not obliterate the original isotopic record.
When all criteria have been satisfied, and an accurate isotopic temperature has been obtained, it is still not always a straightforward matter to report a climatologically meaningful ocean temperature. The living habits, and especially the site and water depth of carbonate secretion, must be known. For organisms for which modern counterparts are extant, this can be relatively straightforward, as, for example, for Tertiary and late Cretaceous planktonic foraminifera (Douglas and Savin, 1978). When modern counterparts are not extant, establishment of the environment of carbonate deposition can be more difficult and may in some cases depend on isotopic analyses of large numbers of taxa within fossil assemblages. As noted above, because questions about early Mesozoic and older climates do not usually require answers as precise as do questions about Tertiary climates, meaningful information may be obtained in many cases without solution of these ecological problems.
THE ISOTOPIC RECORD
Isotopic studies of Tertiary and late Cretaceous climates have been concentrated on deep-sea sediments. While in some in-
stances useful data have been obtained from deep-sea piston cores and from rocks exposed on the continents, it is the Deep Sea Drilling Project (DSDP) that has provided the largest collection of samples for isotope paleoclimatic study. The availability of these DSDP samples, more than anything else, has been responsible for the enormous increase in the number of isotopic investigations of pre-Pleistocene climates during the past 10 yr. A compilation of much published (and some unpublished) isotopic data on Cretaceous and Tertiary foraminifera is shown in Figure 18.1. The relationship between the isotopic data and other aspects of paleo-oceanography such as oceanic anoxic events, the biotic crisis at the Cretaceous-Tertiary boundary, Eocene-Oligocene extinctions, the Messian “crisis,” and sea-level changes has recently been reviewed by Arthur (1979).
The temperature trend during the past 130 m.y. has been generally downward, but it has been neither smoothly nor monotonically downward. Bottom-water temperatures have decreased to their modern low values from values of more than 15°C, which prevailed during much of the Cretaceous. Late Cretaceous time saw significant cooling of bottom waters to values of 10 to 12°C. This cooling was not especially abrupt or intense compared to subsequent Tertiary events. From late middle Eocene through early Miocene time there was a series of warmings and coolings of bottom waters. Waters seldom warmed as much as they had cooled, and the net drop in bottom temperature between the middle Eocene high and the late Oligocene low was 10 or 11°C. [An alternative interpretation of these data, offered by Matthews and Poore (1980) is that the temperature drop was not so great as this and that a portion of the oxygen isotopic change reflected significant growth of continental ice during Eocene and Oligocene time rather than a temperature drop.] Many of the decreases in Paleogene bottom-water temperatures appear quite abrupt. Best documented of these is that near the Eocene-Oligocene boundary, where Kennett and Shackleton (1976) proposed a cooling of 5°C in 100,000 yr. An aspect of interpretation of these data that remains incompletely resolved is the extent to which the bottom-water temperature fluctuations record the surface-temperature history of a single source region (perhaps at the coast of Antarctica) and the extent to which they record alternations in the source area for bottom-water production (e.g., from high northern to high southern latitudes).
The Miocene benthic isotopic record from DSDP Site 289 (Woodruff et al., 1981) is an especially striking one and is shown in Figure 18.2. Between 15 and 13.5 million years ago (Ma) a large net increase in benthic foraminiferal 18O/16O occurred. This isotopic shift probably reflects both bottom-water cooling and rapid accumulation of ice on Antarctica. There is every reason to believe that a major Antarctic icecap has persisted from that time to the present. However, variations of its size and isotopic composition during Miocene and Pliocene times are not well known. This introduces a degree of uncertainty about the extent to which late Miocene and Pliocene benthic foraminifera isotopic variations should be taken to reflect bottom-water temperature variations as opposed to variations in continental ice volume and isotopic composition.
Through most of Cretaceous and Tertiary time, benthic and planktonic oxygen isotopic compositions fluctuated in roughly parallel manner. This pattern changes during middle Miocene time. Although high-latitude planktonic foraminiferal 18O/16O ratios increase, as do those of the benthics (Shackleton and Kennett, 1975), tropical values decrease (Savin et al., 1975), indicating a warming of tropical surface waters. Hence, middle Miocene time is characterized not only by the rapid growth of ice on Antarctica but by a change in the way heat is distributed on the surface of the Earth and by a marked increase in the equator-to-pole temperature gradient. Meridional heat transfer must have been sharply reduced, causing high latitudes to cool while low latitudes warmed. An understanding of the cause of this major change in the Earth’s thermal regime is fruitful ground for future research.
The earliest application of oxygen isotope measurements to paleoclimates was the study of Cretaceous climates by Urey et al. (1951) followed by that of Lowenstam and Epstein (1954). This study included numerous analyses of belemnites sampled from outcrops. A characteristic of this study, as well as other similar studies, is the large amount of scatter in the isotopic data, which makes interpretation difficult. Among possible reasons for isotopic variation are postdepositional alteration; variations in the temperature and isotopic composition of seawater in the relatively shallow, nearshore sedimentary environments in which most or all of the samples were deposited; and migration of belemnites of different ontological stages into different growth environments. A synthesis by Stevens and Clayton (1971) of paleotemperature trends derived from several studies of Jurassic and Cretaceous megafossils is shown in Figure 18.3. Some of the results that appear discrepant from study to study may reflect real climatic differences from place to place, but in many instances this is unlikely. Further work is needed on the isotope systematics of megafossils in outcrop and the factors (especially diagenetic alteration) that affect their isotopic compositions. Some work on these problems has been done in the past several years, using the scanning electron microscope and cathodoluminescence-equipped microscope, which have become readily available. With these new tools it may well be possible to develop criteria to identify and eliminate samples that have undergone postdepositional isotopic alteration.
There exists in the literature a small number of analyses of pre-Jurassic carbonate shells. All of these are, of necessity, from the continents, owing to the lack of oceanic crust this old. Hence, extreme caution is needed to avoid the effects of post-depositional alteration. Because of the small number of analyses of samples widely scattered in space and time there has been little impetus to synthesize such data. However, the search for suitable samples for study and the synthesis of the existing and new data should be encouraged. Both may yield valuable information about early Mesozoic and Paleozoic climates.
SILICA-WATER ISOTOPIC PALEOTEMPERATURES
Biogenic silica is a common constituent of marine sediments and is frequently found where biogenic carbonate is absent (e.g., below the calcium carbonate compensation depth). In recent years there has been a great deal of progress made on the development of a biogenic silica-water oxygen isotope geothermometer, chiefly by Labeyrie and coworkers (Labeyrie, 1974; Mikkelsen et al., 1978; Labeyrie and Juillet, 1980). However, analytical methods for isotopic analysis of opaline silica are difficult and arduous. As yet, applications of this approach to paleoclimatological problems have been limited.
Opaline silica becomes diagenetically altered to opal-cristobalite (opal-CT) and then to microcrystalline quartz. Both opal-CT and quartz can be analyzed by methods that have become routine for isotopic analysis of silicates. Studies of the isotopic effects accompanying the conversion of opaline silica through opal-CT to microcrystalline quartz in DSDP sediments have been done by Knauth and Epstein (1975) and Kolodny and Epstein (1976). These studies have shown that isotopic exchange with pore waters accompanies mineralogical reactions at depths of tens to hundreds of meters below the sediment-water interface. Furthermore, the reactions can occur some tens of millions of years following deposition. Hence, isotopic temperatures obtained from opal-CT and microcrystal-
line quartz should be related to, but somewhat higher than, bottom temperatures at a time a few millions to tens of millions of years more recent than the time of deposition of the silica. Kolodny and Epstein (1976) have presented a comparison between Tertiary and Cretaceous benthic foraminiferal isotopic temperatures and chert isotopic temperatures based on the study of DSDP materials, and the comparison is consistent with the behavior outlined above. The cherts do contain a climatic signal that is sufficiently imprecise to be of little use in the study of Tertiary or Cretaceous climates. However, for older times, for example the Precambrian, for which quantitative climatic data are indeed scanty, this approach can provide climatic information of considerable use. As a cautionary note, in some geologic settings the conversion of diatomaceous silica to chert may occur under conditions that at least partially mask the climatically relevant isotopic signal. Murata et al. (1977) analyzed two sections of opal, opal-CT, and chert from the Miocene Monterey Formation of California and found 18O/16O ratios in the quartz consistent with diagenetic formation at 80°C. Using the quartz-water isotopic fractionation curve given by Knauth and Epstein (1976) that temperature would be reduced to 66°C. Even that is, of course, substantially warmer than any reasonable estimate of Miocene surface or bottom temperatures.
Perry and coworkers (most recently, Perry et al., 1978) and Knauth and coworkers (most recently, Knauth and Lowe, 1978) have published large numbers of isotopic analyses of Precambrian cherts. Both of these series of papers have served to document a striking tendency toward lower 18O/16O ratios with increasing age. A summary of these data is given in Figure 18.4. Although the data of these two research groups are consistent with one another, their interpretations differ. Knauth and Lowe have argued that the data are best explained if the oxygen isotopic composition of the ocean has remained essentially constant, with time, and that the low 18O/16O ratios of early Precambrian charts reflect warm temperatures, perhaps as warm as 80°C. Muehlenbachs and Clayton (1976) have proposed that the 18O/16O ratio of the modern hydrosphere is determined by two seawater-lithosphere reactions: low-temperature weathering reactions, which deplete the oceans in 18O and high-temperature hydrothermal alteration of basalt, which enriches the oceans in 18O. Knauth and Lowe (1978) have suggested that if similar processes occurred throughout Precambrian time the 18O/16O ratio of seawater would have remained constant. Gregory and Taylor (1981) have concluded that 18O/16O ratio of seawater is constrained to a value similar to today’s by the interaction between water and rock associated with the seafloor spreading process. Perry et al. (1978) on the contrary, have argued that the 18O/16O ratio of seawater was substantially lower in early Precambrian time than today and that ocean temperatures need not have been markedly greater than Phanerozoic temperatures. Perry et al. have suggested that in the Archaean intense weathering and low-temperature alteration of volcanic rocks were the dominant processes controlling the 18O/16O ratio of the oceans. This, they argued, could entail a depletion of 18O in seawater by perhaps as much as 12 to 24 per mil relative to today’s values.
It is of course possible that higher temperatures and lower oceanic 18O/16O ratios have both contributed to the oxygen isotope record of Precambrian cherts. Resolution of the relative importance of these two variables at various times during Precambrian time is of the utmost importance to Precambrian paleoclimatology. The approach of Knauth and Epstein (1976) in which both 18O/16O and D/H ratios of cherts are used, may provide the best solution. However, even without resolution of this question, useful paleoclimatic conclusions can be drawn. As an example, Oskvarek and Perry (1976) concluded from the analyses of cherts from the 3800 Ma Isua Series (Greenland) that the highest possible surface temperature during Isua time was 150°C. This is based on chert precipitation from a hypothetical ocean of +6 per mil, the approximate value of water degassed from the mantle, and the highest 18O/16O ratio for seawater given by any reasonable model for ocean formation. Lower estimates of the 18O/16O ratio in seawater would give lower isotopic temperatures. A minimum temperature of 0°C can be estimated from the fact that the Isua series is made up of apparently water-laid sediments. For any time in the Phanerozoic, an ocean temperature estimate of 0–150°C would be trivial. For a 3800-m.y.-old ocean it is not. (Keep in mind that 150°C is the temperature of water in equilibrium with an atmospheric of 4.65 atm and that only approximately 46 m of seawater would have to be evaporated in order to achieve that .) Even crude estimates of early Archaean ocean temperatures can be useful in constraining models of the atmosphere and of the Sun during the Earth’s earliest history.
THE ISOTOPIC COMPOSITION OF PRECIPITATION ON THE CONTINENTS
Ocean temperature is the parameter of climate most frequently and readily determined from the isotopic study of ocean sediments. Isotopic studies of terrestrial materials for paleoclimate purposes are far less common. The most readily determined climatic parameter obtained from isotopic studies of terrestrial materials is not temperature but the isotopic composition of precipitation. This is in turn dependent largely but not exclusively on temperature of precipitation. Moreover, because precipitation in few areas is uniformly distributed throughout the year, mean temperature of precipitation can be significantly different than mean annual temperature. Still, the 18O/16O (or D/H) ratio of precipitation can be a useful climatic variable.
Most studies that have aimed at estimating isotopic composition of precipitation have had their greatest utility in investigations of Quaternary climate and will be mentioned here only briefly. Hanshaw and Hallet (1978) analyzed the 18O/16O ratio of subglacial calcite precipitated as a crust on rock surfaces over which glaciers flowed. Because the meltwater from which the calcite precipitated must have a temperature of almost exactly 0°C, the isotopic composition of the meltwater can be obtained from the 18O/16O ratio of the calcite. This must be the isotopic composition of precipitation in the glacier’s zone of accumulation. There have been no published attempts to apply this approach to pre-Pleistocene subglacial calcites.
Hendy (1971) and Schwarcz, Harmon, and coworkers (e.g., Schwarcz et al., 1976) have estimated both temperature and isotopic compositions from cave deposits. In many instances, the D/H ratio of fluid inclusions trapped in calcite of speleothems can be shown to be representative of that of the seepage waters from which the calcite was precipitated. Because the 18O/16O and D/H ratios of precipitation, worldwide, are linearly related (Craig, 1961), the 18O/16O ratio of seepage water can be estimated from its D/H ratio. Analysis of the 18O/16O ratio of the calcite permits calculation of an isotopic temperature for speleothem formation. This temperature, and the 18O/16O ratio of precipitation, should approximate mean annual temperature and isotopic composition of precipitation in the groundwater recharge zone.
Attempts to derive paleoclimatic information from isotopic studies of tree rings (especially of cellulose) appear promising (Epstein and Yapp, 1976; Libby et al., 1976; Epstein et al., 1977; Yapp and Epstein, 1977; Long et al., 1978). There are, however, still problems in the interpretation of these data that need to be resolved. The scarcity of suitable old material for analysis and uncertainties about preservation of original isotope ratios over long periods of time indicate that for the foreseeable future most applications of this method will be to Quaternary samples.
Clay minerals formed during weathering acquire D/H and 18O/16O ratios that reflect the temperature and isotopic composition of the weathering environment. Once formed, clay minerals are highly resistant to subsequent alteration of isotopic compositions except when mineralogic reactions also occur (Lawrence, 1970; Savin and Epstein, 1970; Lawrence
and Taylor, 1972). Hence, isotopic compositions of ancient soils should yield climatic information. The climatic signal can be obscured in part because each clay mineral has its own isotopic systematics. However, careful mineralogic as well as isotopic study may be sufficient in many cases to resolve the isotopic systematics of different clay minerals within mineralogically complex soils. Lawrence and Taylor (1971) analyzed a large number of Quaternary age soils from the western United States. They found isotopic compositions basically consistent with the modern climatic regime. Lawrence (1970) analyzed D/H and 18O/16O ratios of kaolinite-rich soils of Tertiary age from the western United States and found a distribution pattern of D/H generally similar to those of today. However, he found a smaller difference between Tertiary D/H ratios of coastal regions and those of inland regions, suggesting less extreme climatic differences between the two areas than at present,
Studies of soil isotopic composition cannot be expected to provide climatic information of the same precision as can a number of other isotopic paleoclimate tools. However, the method should provide information of value where other approaches may not be applicable. Until now the method has found extremely little application to paleoclimatic problems. It warrants further consideration and, perhaps, development.
There are a number of techniques whereby paleoclimatic information can be obtained from stable isotope data. Most precise and widely used is the carbonate-water paleotemperature scale. However, its applicability is restricted, for the most part, to the Cenozoic and late Mesozoic record because of lack of preservation of most older samples. The silica-water system has the potential for yielding useful climatic information for Precambrian times, but further developmental work is required before the results can be uniquely interpreted in terms of climate. However, for early Archaen time, even the most approximate estimates of surface temperatures can be useful and the silica-water system has provided these.
Isotopic methods are less useful for yielding information about pre-Pleistocene climates on the continents than they are about marine climates. However, some approaches such as the analysis of paleosols, speleothems, or subglacially precipitated carbonates may be useful in some cases.
Financial support was provided by the National Science Foundation, Grant OCE 79–17017 (CENOP). Alan Hecht and Paul Knauth provided helpful reviews of the original manuscript. Contribution No. 133, Department of Geological Sciences, Case Western Reserve University.
Arthur, M. (1979). Paleoceanographic events-recognition resolution and reconsideration, Rev. Geophys. Space Phys. 17, 1474–1494.
Berger, W.H. (1979). Stable isotopes in foraminifera, Soc. Econ. Paleontol. Mineral. Short Course 6, pp. 156–198.
Bowen, R. (1966). Paleotemperature Analysis, Elsevier, Amsterdam, 265 pp.
Craig, H. (1961). Isotopic variations in meteoric waters, Science 133, 1702–1703.
Craig, H. (1965). The measurement of oxygen isotope paleotemperatures, Proc. of the Spoleto Conf. on Stable Isotopes in Oceanographic Studies and Paleotemperatures 3, Cons. Naz. Richerche, Lab. Geol. Nucleare, Pisa, pp. 1–24.
Douglas, R.G., and S.M.Savin (1978). Oxygen isotopic evidence for the depth stratification of Tertiary and Cretaceous planktic foraminifera, Mar. Micropaleontol. 3, 175–196.
Douglas, R.G., and F.Woodruff (1981). Deep sea benthic foraminifera, in The Sea, Vol. 7, C.Emiliani, ed., Wiley-Interscience, New York.
Duplessy, J.C., C.LaLou, and A.C.Vinot (1970). Differential isotopic fractionation in benthic foraminifera and paleotemperatures reassessed, Science 168, 250–251.
Epstein, S., and C.J.Yapp (1976). Climatic implications of the D/H ratio of hydrogen in C-H groups in tree cellulose, Earth Planet. Sci. Lett. 30, 252–261.
Epstein, S., R.Buchsbaum, H.A.Lowenstam, and H.G.Urey (1951). Carbonate-water isotopic temperature scale, Geol. Soc. Am. Bull. 62, 417–426.
Epstein, S., P.Thompson, and C.J.Yapp (1977). Oxygen and hydrogen isotopic ratios in plant cellulose, Science 198, 1209–1215.
Grazzini, C.V. (1976). Non-equilibrium isotopic compositions of shells of planktonic foraminifera in the Mediterranean Sea, Paleogeogr. Paleoclimatol. Paleoecol. 20, 263–276.
Gregory, R.T., and H.P.Taylor, Jr. (1981). Oxygen isotope profile in a section of Cretaceous oceanic crust, Samail Ophiolite, Oman: Evidence for δ18O buffering of the oceans by deep (greater than 5 km) seawater hydrothermal circulation at mid-ocean ridges, J. Geophys. Res. 86, 2737–2755.
Hanshaw, B.B., and B.Hallet (1978). Oxygen isotope composition of subglacially precipitated calcite: Possible paleoclimatic implications, Science 200, 1267–1270.
Hecht, A.D. (1976). The oxygen isotopic record of foraminifera in deep-sea sediment, in Foraminifera, Vol. 2, R.H.Hedley and C.G. Adams, eds., Academic Press, London, pp. 1–43.
Hendy, C. (1971). The isotopic geochemistry of speleothems—I. The calculation of the effects of different modes of formation of the isotopic composition of speleothems and their applicability as paleoclimatic indicators, Geochim. Cosmochim. Acta 35, 801–824.
Hudson, J.D. (1977). Oxygen isotope studies on Cenozoic temperatures, oceans, and ice accumulation, Scottish J. Geol. 13, 313–325.
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.
Knauth, L.P., and S.Epstein (1975). Hydrogen and oxygen isotope ratios in silica from the JOIDES Deep Sea Drilling Project, Earth Planet. Sci. Lett. 25, 1–10.
Knauth, L.P., and S.Epstein (1976). Hydrogen and oxygen isotope ratios in nodular and bedded cherts, Geochim. Cosmochim. Acta 40, 1095–1108.
Knauth, L.P., and D.R.Lowe (1978). Oxygen isotope geochemistry of cherts from the Onverwacht Group (3.4 billion years), Transvaal, S. Africa, with implications for secular variations in the isotopic composition of cherts, Earth Planet. Sci. Lett. 41, 209–222.
Kolodny, Y., and S.Epstein (1976). Stable isotope geochemistry of deep sea cherts, Geochim. Cosmochim. Acta 40, 1195–1209.
Labeyrie, L.D. (1974). New approach to surface seawater paleotemperatures using 18O/16O ratios in silica of diatom frustules, Nature 248, 40–42.
Labeyrie, L.D., and A.M.Juillet (1980). Oxygen isotopic exchangeability of biogenic silica, EOS: Trans. Am. Geophys. Union 61, 259.
Lawrence, J.R. (1970). 18O/16O and D/H ratios of soils, weathering zones, and clay deposits, Unpublished Ph.D. thesis, California Institute of Technology.
Lawrence, J.R., and H.P.Taylor, Jr. (1971). Deuterium and oxygen-18 correlation: Clay minerals and hydroxides in Quaternary soils compared to meteoric waters, Geochim. Cosmochim. Acta 35, 993–1003.
Lawrence, J.R., and H.P.Taylor, Jr. (1972). Hydrogen and oxygen isotope systematics in weathering profiles, Geochim. Cosmochim. Acta 36, 1377–1393.
Libby, L.M., L.J.Pandolfi, P.H.Payton, J.Marshall, III, B. Becker, and V.Giertz-Sienbenlist (1976). Isotopic tree thermometers, Nature 261, 284–290.
Long, A., J.C.Lerman, and A.Ferhi (1978). Oxygen-18 in tree rings: Paleothermometers or paleohygrometers, in Short Papers of the Fourth International Conference, Geochronology, Cosmochronology, Isotope Geology, 1978, R.E.Zartman, ed., U.S. Geol. Surv. Open-File Rep. 78–701, pp. 253–254.
Lowenstam, H.A., and S.Epstein (1954). Paleotemperatures of the post-Aptian Cretaceous as determined by the oxygen isotope method, J. Geol. 62, 207–248.
Matthews, R.K., and R.Z.Poore (1980). The Tertiary δ18O record: An alternative view concerning glacio-eustatic sea level fluctuations, Geology 8, 501–504.
Mikkelsen, N., L.Labeyrie, and W.H.Berger (1978). Silica oxygen isotopes in diatoms: A 20,000 year record in deep-sea sediments, Nature 271, 536–538.
Muehlenbachs, K., and R.N.Clayton (1976). Oxygen isotope composition of the oceanic crust and its bearing on seawater, J. Geophys. Res. 81, 4365–4369.
Murata, K.J., I.Friedman, and J.D.Gleason (1977). Oxygen isotope relations between diagenetic silica minerals in Monteray Shale, Temblor Range, California, Am. J. Sci. 277, 259–272.
Oskvarek, J.D., and E.C.Perry, Jr. (1976). Temperature limits on the early Archaean ocean from oxygen isotope variations in the Isua supracrustal sequence, West Greenland, Nature 259, 192–194.
Perry, E.C., Jr., S.N.Ahmad, and T.M.Swulius (1978). The oxygen isotope composition of 3800 m.y. old metamorphosed chert and iron formation from Isukasia, West Greenland, J. Geol. 86, 223–239.
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., and S.Epstein (1970). The oxygen and hydrogen isotope geochemistry of clay minerals, Geochim. Cosmochim. Acta 34, 25–42.
Savin, S.M., and F.G.Stehli (1974). Interpretation of oxygen isotope paleotemperature measurements; Effect of 18O/16O ratio of sea water, depth stratification of foraminifera, and selective solution, in Colloq. Int. CNRS No. 219, Les Methodes Quantitative d’Etudes des Variations au cours du Pleistocene, pp. 183–191.
Savin, S.M., and H.W.Yeh (1981). Stable isotopes in ocean sediments, in The Sea, Vol. 7, C.Emiliani, ed., Wiley-Interscience, New York.
Savin, S.M., R.G.Douglas, and F.G.Stehli (1975). Tertiary marine paleotemperatures, Geol. Soc. Am. Bull. 86, 1499–1510.
Schwarcz, H.P., R.S.Harmon, P.Thompson, and D.C.Ford (1976). Stable isotope studies of fluid inclusions in speleothems and their paleoclimatic significance, Geochim. Cosmochim. Acta 40, 657–665.
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., J.D.H.Wiseman, and H.A.Buckley (1973). Nonequilibrium isotopic fractionation between seawater and planktonic foraminiferal tests, Nature 242, 177–179.
Stevens, G.R., and R.N.Clayton (1971). Oxygen isotope studies on Jurassic and Cretaceous belemnites from New Zealand and their biogeographic significance, N.Z. J. Geol. Geophys. 14, 829–897.
Teis, R.V., and D.P.Naidin (1973). Paleothermometry and Isotopic Composition of Oxygen in Organic Carbonates, Moscow, 256 pp. (in Russian).
Urey, H.C. (1947). The thermodynamic properties of isotopic substances, J. Chem. Soc., 562.
Urey, H.C., H.A.Lowenstam, S.Epstein, and C.R.McKinney (1951). Measurement of paleotemperatures and temperatures of the upper Cretaceous of England, Denmark, and the southeastern United States, Geol. Soc. Am. Bull. 62, 399–416.
van Donk, J. (1970). The oxygen isotope record in deep sea sediments, Ph.D. thesis, Columbia U., New York, 228 pp.
Williams, D.F., M.A.Sommer, and M.L.Bender (1977). Carbon isotopic compositions of recent planktonic foraminifera of the Indian Ocean, Earth Planet. Sci. Lett. 36, 391–403.
Woodruff, F., S.M.Savin, and R.G.Douglas (1980). Biological fractionation of oxygen and carbon isotopes by Recent benthic foraminifera, Mar. Micropaleontol. 5, 3–11.
Woodruff, F., S.M.Savin, and R.G.Douglas (1981). A high resolution oxygen isotope study of Pacific Miocene bottom temperatures, Science 212, 665–668.
Yapp, C.J., and S.Epstein (1977). Climatic implications of D/H ratios of meteoric waters over North America (9500–22,000 yr b.p.) as inferred from ancient wood cellulose C-H hydrogen, Earth Planet. Sci. Lett. 34, 333–350.