Sedimentary Geology v. 101, p. 43-53 (1996) received 26 May 1994; revised version accepted 21 February 1995The whole-rock cerium anomaly, tested for outer shelf-upper slope stratigraphic sections from the middle Ordovician through the lower Silurian of Scotland, is proposed as an empirical technique to develop a eustatic 3rd-order or finer-scale sea-level curve. This interval was chosen as it straddles the well-documented Late Ordovician glaciation and can be defined by graptolite zones. The anomaly is calculated from neutron activation analysis of low-carbonate, phosphate-free, fossil-free field-identified shales of the graptolite facies by comparison of the normalized cerium content with the linearized trend of the normalized composition of other rare earth elements in order of atomic number. For sections originally deposited in the main pycnocline below the surface mixed layer, values of the anomaly for a given sample would indicate its position on the redox curves developed for the early Paleozoic by Wilde (1987). Changes in the anomaly that are positive with time would indicate a lowering of sea level as the apparent depth on the redox curve would reflect more oxic conditions. Relative changes negative with time would indicate a rise in sea level as the apparent depth reflects more anoxic conditions. Depending on the vertical sample spacing and the time interval sampled, resolution of the order of 1 m.y. might be achieved. Thus the Vail et al. (1977) curves of the 3rd order (1 to 10 m.y.) or of finer scale could be obtained by this technique with the proper choice of section. Accordingly, for the early through middle Paleozoic when the main pycnocline was anoxic, this geochemical technique could be used to develop eustatic sea-level curves and additionally offer an independent calibration for seismic stratigraphy as well as an indicator of glacial-interglacial climatic sequences or eustatic changes due to fluctuations in global ridge crest volumes.
ABSTRACT
1. Introduction
Vail et al. (1977) produced a series of sea-level curves through time based on interpretation of seismic records from the continental shelf. Although such curves have not been accepted without skepticism (Watts, 1982; Hallam, 1992), the Vail curves must be considered one of the major geologic advances in the past 15 years. Independent verification of these curves has proven difficult as previous geological methods for determining sea level have been qualitative, being the sedimentological and stratigraphic record of transgressions and regressions (see, for example, Summerhayes, 1986). A possible independent method for determining relative sea level, for specific geologic situations, is suggested by the use of the cerium anomaly as part of a chemostratigraphic study of anoxic sediments from outer shelf-upper slope sections in the Paleozoic.
2. Cerium anomaly
The use of the cerium anomaly was first proposed by Elderfield and Greaves (1982) as a consequence of the change in the ionic state of Ce as a function of oxidation state. Of the rare earths only Ce and possibly Eu show potential variations as a function of oxidation-reduction conditions found in natural sedimentary/oceanic environments. Calculation of the anomaly is based on the assumption of a "linear" decline in rare earth concentrations with an increase of atomic number when the elements are normalized to some standard, for example, chondrites. This assumption is based on an empirical observation of such a pattern in certain sedimentary rocks. Only Ce and Eu deviate significantly from a line that declines toward the heavier rare earths. We normalized to chondrite abundance of La, Ce and Sm (Haskin et al., 1966), which means dividing the NAA measurement in parts per million by 0.3 (La), 0.84 (Ce), and 0.21 (Sm). An anomaly occurs if the normalized concentration plots above or below a value calculated by the assumption of the straight line variation.As the order of rare earth elements is La, Ce, Pr, Nd, Pm, Sm. ideally the anomaly would be calculated as normalized Ce values that plot above or below the straight line extrapolation between La and Pr. This results in an anomaly of log[2Ce*/(La* + Pr*)]. The superscript '*' implies the concentration normalized to chondrite. For analytical reasons Pr and Pm rarely are reported, but Nd and Sm are. Thus, the Ce anomaly with respect to Nd is log[3Ce*/(2La* + Nd*)] and with respect to Sm is log[5Ce*/(4La* + Sm*)]. Fig. 1 shows the anomaly calculated using Sm. because more samples contained detectable concentrations of Sm than Nd. Nonetheless, when the anomaly was calculated using either Nd or Sm. the observed pattern was essentially the same.
Similar (that is varying in absolute value, but not in qualitative understanding) results are obtained if rare earth element concentrations are normalized using the North American Shale Composite (NASC) (Gromet et al., 1984). We prefer to use chondritic values, as our examination of general shale compositions suggests that the NASC is not representative of anoxic shales (Quinby-Hunt et al., 1991). In any case, the qualitative results using either normalization method are equivalent and, as noted below, the relevance of absolute values is questionable when examining rocks from such an ancient period. In order to convert the Ce anomaly values given here to results normalized by the NASC, one should use: La = 31 ppm, Ce = 67 ppm, Nd = 34 ppm, Sm = 6.1 ppm, Eu = 1.18 ppm, Gd = 5.6 ppm, Tb = 0.85 ppm, and Yb = 3.1 ppm (R. Schmitt, pers. commun., 1989).
In oxic conditions, Ce is less readily dissolved in seawater, so that oxic seawater is more depleted with respect to Ce, whereas oxic sediments are more enhanced with respect to Ce. Accordingly, organisms extracting phosphate from oxic seawater show a negative Ce anomaly (Wright et al., 1987), whereas Fe oxide-rich oxic sediments, such as red clay, have a positive Ce anomaly (Thomson et al., 1984). Conversely, in suboxic seawater Ce-containing sediments are mobilized so that Ce is released into the water column resulting in a less negative to a positive anomaly in seawater (DeBaar et al., 1985, 1988; DeBaar, 1991; Sholkovitz and Schneider, 1991). Therefore in anoxic sediments, Ce is depleted and the sediments show a negative anomaly. Most of the black shales discussed in this paper have the very low Mn concentrations and/or low Fe concentrations associated with anoxic sediments (Quinby-Hunt and Wilde, 1994). Only five samples have a positive anomaly; these and the samples with the least negative cerium anomalies are associated with conditions generally considered more oxic. See below, discussion of the example from the Ordovician-Silurian boundary.
As the remains of organisms sequestering phosphate from dominantly anoxic waters would have positive anomalies (Wright et al., 1987; Grandjean-Lecuyer et al., 1993) and anoxic sediments would show negative anomalies, mixtures of such phosphatic fossils with anoxic sediments could give some average, non-indicative value.
Accordingly, only sediments which do not contain conodonts, or other phosphatic fossils can be used for determination of whole-rock Ce anomalies. With some care, the graptolite facies of the Paleozoic, particularly low-calcium black shales without conodonts, can be used to compute a Ce anomaly indicative of anoxic conditions.
Originally, the cerium anomaly was used in chemostratigraphy for the determination of redox conditions in the ocean (Wang et al., 1986; Wright et al., 1987) from analysis of the rare earth distribution of phosphates in conodonts and fish bones. Such studies have indicated that in the early and middle Paleozoic and occasionally in later times (such as the Early Triassic and Cretaceous Anoxic Events), significant volumes of seawater have been anoxic. Recently, the use of the cerium anomaly has been extended to whole-rock analyses (Berry et al., 1987) in sections which only rarely contain conodonts or inarticulate brachiopods. Conceptually, the Ce anomaly in whole rock should be reversed in sign from that of fossils, which follow the Ce in the water column. The whole-rock Ce anomaly through the Ordovician-Silurian boundary section at Dob's Linn, Scotland, corresponds with the qualitative sea-level curve compiled by Fortey (1984, p. 39; Brenchley, 1984, p. 310; Hallam, 1992) for the Ordovician based on sedimentologic and biologic grounds. Sedimentologic evidence indicating that the bottom was shallowing independent of eustatic changes permitted the extrapolation of the sea-level curve into the Silurian.
As the bottom shallowed, the anomaly became more positive (Fig. 1). Changes in the value of the anomaly could be related to the redox conditions predicted by the ventilation model of Wilde (1987) with more negative values found during warmer climates and transgressive conditions and more positive values found during cooler to glacial climates and regressive conditions. Thus the wholerock cerium anomaly, in such facies, potentially is a geochemical parameter that characterizes chemical paleo-oceanographic conditions related to relative eustatic sea-level changes independent of sedimentological or seismic considerations. Furthermore, the use of a chemical parameter offers the possibility of the extension of its interpretation to quantitative paleo-environmental studies beyond that of sea-level curves.
There is no general agreement as to the mechanism for Ce-depletion in seawater and enhancement in sediments. Some researchers attribute this phenomenon to oxidation of Ce+3 to Ce+4 and incorporation into Mn oxyhydroxides as CeO24 preferentially to Ce(OH)4. At pH > 7.5, Ce(OH)4 precipitates associating with Fe-Mn-AI-Ti-oxyhydroxide coatings on carbonate minerals. More recently, researchers have observed that Ce appears to be preferentially mobilized with Mn in nitratic (suboxic, Mn-reducing) waters (German and Elderfield, 1989, 1990; DeBaar, 1991; German et al., 1991). The expression derived for the Ce concentration in seawater is more strongly dependent on pH than on PO2 (Eh). The mechanism postulated by Liu et al. (1988, 1989) implies that only marine carbonates, conodonts and ichthyoliths deposited in pelagic, slowly accumulating oxic sediments are suitable candidates for obtaining reliable Ce anomalies for indicating marine environmental conditions. This places extreme limitations in examining ancient marine environments, as many of them are anoxic and unfossiliferous. This limitation on the use of the Ce anomaly has not been accepted by Murray and coworkers who have applied the Ce anomaly to elucidate source terms of REE's in cherts and silicious rocks (Murray et al., 1990, 1991a, b, 1992a, b).
The observed Ce anomaly is not limited to marine carbonates, conodonts and ichthyoliths (Elderfield et al., 1981a). If the right constraints (discussed below) are applied, the whole-rock Ce anomaly can be used as an indicator of intensity of anoxia and therefore of eustatic sea level. In marine waters, pH is dependent on many variables, including atmospheric CO2, dissolved oxygen concentration and intensity of anoxia. While the pH of seawater is relatively constant in modern aerobic ocean water, such is not the case for anoxic waters and the more anoxic the waters, the lower the pH (Baas-Backing et al., 1960; Skirrow, 1975; Grasshoff, 1975; Jacobs et al., 1985). Thus, in marine waters pH and oxygenation/ anoxia are inextricably linked. Furthermore, the chemistry of anoxic waters differs significantly from that of oxic seawater (Bacon et al., 1980; Jacobs et al., 1985, 1987; Elderfield et al., 1981a, b; DeBaar et al., 1985, 1988; German and Elderfield, 1989, 1990). Thus as Liu et al. (1988, 1989) note, the stability relationships determined for pH 7.8-8.1, alter considerably under anoxic conditions: under pH 7.5 Ce+4 can be expected to be reduced to Ce+3. At pH < 7.5 and, certainly at pH < 7.0, the stability of Ce phosphates and carbonates is reduced significantly. Thus, although Liu et al.'s observations are important for researchers examining waters that are primarily oxic and range from fresh to marine conditions, the conditions for formation of the low-calcic marine black shales under consideration here differ. They are solely marine, slowly deposited, experiencing highly variable Eh-pH (redox) regimes. The association of the Ce anomaly with varying redox conditions, first reported by Berry et al. (1987) is based on empirical observation. Thus, it is reasonable to suggest that there are at least two mechanisms to produce a cerium anomaly in marine sediments. One is applicable in near-shore oxic settings following the reasoning of Liu and his colleagues, with the other applicable in deeper-marine pelagic conditions with varying redox conditions discussed here.
3. Interpretation of the whole-rock cerium anomaly
It is premature or potentially impossible to assign quantitative meaning to the absolute values of measured Ce in the whole rock. That is, as yet, a scalar relationship between the Ce anomaly and Eh or the fugacity of oxygen, and, secondarily, a direct relationship between depth of water and the absolute value of the cerium anomaly, is still to be defined. This is due in part to the artificial nature of the calculation of the anomaly and our rudimentary understanding of the geochemistry of Ce in both sediments and seawater (Piper, 1974; Elderfield et al., 1981a, b; DeBaar et al., 1985, 1988; Elderfield and Sholkovitz, 1987; Sholkovitz, 1988; Sholkovitz and Elderfield, 1988; Liu et al., 1988, 1989). However, for uniform depositional and source conditions, it should be possible to interpret relative change with a range from anoxic (negative) to oxic (positive) values. Murray et al., 1990, 1991a) have shown that the Ce anomaly depends on depositional setting. Thus, increasing values indicate more oxic conditions, whereas decreasing values indicate more reducing or anoxic conditions. Wilde's (1987) ventilation model of oceanic redox conditions shows three potential redox profiles for different climatic conditions for a given oxygen concentration in the atmosphere (Fig. 2). If the depth of water during sedimentation is shallower than the anoxic maximum in the pycnocline for any climatic condition, then the redox conditions and presumably the Ce anomaly in sediment would vary with depth of water. Thus for eustatic changes, a transgression would effectively deepen any fixed point on the seafloor (assuming the rate of deposition is much smaller than the eustatic sea-level change). During a transgression the bottom waters would become more anoxic and the Ce whole-rock anomaly more negative. A regression would shoal any fixed point on the bottom. Thus during a regression, the bottom waters would become more oxic and the Ce whole-rock anomaly would be more positive.Eustatic changes would be masked if: (A) the seafloor moved vertically at a rate faster than sea-level fluctuations; or (B) the rate of sedimentation/ erosion exceeded the rate of sea-level fluctuations. For case A, the whole-rock Ce anomaly would increase during uplift as the seafloor would experience more oxic conditions of shallower water, and would decrease during sinking or down warping as the real depth would decrease toward the anoxic maximum. Case B could only be observed concurrent with increased sedimentation and thus would always increase the Ce anomaly. Such changes in sedimentation would be seen in the Ethology and increased coarsening of the grain size. In the near shore, Cases A and B are not independent, for example, as shallowing would tend to place the sampling point nearer the coast line and produce increased sedimentation and coarser sediments.
R.A. Schmitt (pers. commun., 1989) suggested that the Ce anomaly in the Case B situation could reflect transgression/ regression changes on a near-shore sedimentary source using the oxic cerium model of Liu et al. (1988, 1989). In this manner, during a regressive phase, eroded nearshore sediments will be enriched in Ce(OH)4 increasing the Ce anomaly in marine sediments. During a transgressive phase, there will be less deposition of Ce(OH)4-enriched sediments transported into deeper waters, giving a relative "negative" anomaly. Schmitt, using the NASC as a standard, felt that the regressive phase might give a positive anomaly, whereas the transgressive phase would produce a "normal" NASC Ce value in marine sediments. Such a model is not applicable to the example here: the pelagic "starved" basin at Dob's Linn. However, the Schmitt model might be tested in an obvious turbidite regime such as the time synchronous basin at Girvan, where there is stratigraphy.
To avoid or reduce such tectonic and sedimentological complications as mentioned above, sections from the outer shelf and upper slope, characteristic of the graptolitic lithologic facies should be chosen for analysis. Deep ocean and continental slope and margin sediments deposited during mild non-glacial climates should reflect the poor ventilation of the overlying waters below the surface mixed layer resulting in widespread dark, organic-rich hemi-pelagic to pelagic deposits. In contrast, during glacial intervals, slope and margin deposits would only be partially reduced to grey and green silica elastic sediments reflecting the higher oxygen content of overlying bottom waters.
Another problem in assigning absolute depth changes to changing Ce anomaly values is that the three climatically based redox curves from (Fig. 2) are displaced at a given depth. During a regression, the Ce anomaly would show a maximum increase if the climate changed from nonglacial to glacial and a minimum increase if the climate was already glacial. During a transgression the maximum decrease in the Ce anomaly would be if the climate changed from glacial to non-glacial and the minimum decrease would be if the climate was already non-glacial. Thus the Ce anomaly is more sensitive to sea-level changes a similar detailed graptolite during changes in climate than during intervals of climate stability. However, within intervals of climate stability, the anomaly shows greater change with sea-level fluctuation in non-glacial climates than during glacial climates due to the greater expansion of anoxic conditions during non-glacial climates. This assumes variations in climate at the same atmospheric level of oxygen. Over longer times, the %PAL (present atmosphric level) might change, which would shift the redox curves to more anoxic with decreasing %PAL and more oxic with increasing %PAL.
Accordingly, the best test of the whole-rock Ce anomaly method would be in an interval of climatic fluctuation, but with a relatively constant level of oxygen in the atmosphere. The Ce anomaly would track non-climatic eustatic changes such as those caused by increased ridge volume (MacKenzie and Piggot, 1982; Hallam, 1992), but, as noted above, with less resolution.
4. Choice of the Ordovician-Silurian boundary section
Wilde and Berry (1982) and Wilde (1987) have discussed early Paleozoic deep-ocean ventilation pointing out that the deep oceans were ventilated then, as now, during intervals of glaciation and sea-ice formation at high latitudes. One such interval of undoubted glacial conditions was during the Late Ordovician. The Late Ordovician was preceded and followed by global mild climates (Frakes, 1979; Frakes et al., 1992) with minimal or no sea-ice formation. During these prolonged mild intervals, pycnoclinal and deep oceanic waters would become depleted with oxygen due to both longer recycling time and warmer temperatures at the sites of water mass formation (Brass et al., 1982; Wilde and Berry, 1982).Budyko et al. (1987) and Berner and Canfield (1989) agree that the concentration of atmospheric oxygen in the early Paleozoic was less than today. Fig. 2 shows the potential variations in the redoxocline as a function of climate assuming deep ventilation (Wilde and Berry, 1982; Wilde, 1987) and a value of 70% PAL. Paleogeographic (Scotese, 1986) and paleo-oceanographic (Wilde, 1991) reconstructions indicate a tropical latitude for the Dob's Linn area from the middle Ordovician through the Early Silurian. Dob's Linn appears to have been equatorward of the Austral Sub-Tropical Convergence (Wilde, 1991), except possibly during the glacial intervals in the Late Ordovician. Thus, tropical redox profiles with depth (Fig. 2) are a valid choice.
Graptolite biozones are used to define time intervals across the Ordovician-Silurian boundary stratotype at Dob's Linn, Scotland (Fig. 1). Previous sedimentological studies have shown climatic change in this interval (Berry and Boucot, 1973; Brenchley and Newall, 1984). An examination of this section enables us to study the possible use of whole-rock Ce anomaly curves as a tool to predict sea-level change. As each biozone represents approximately 1 to 3 m.y., the calculation of an average value gives a composite climatic result which could not discriminate glacial-interglacial or sea-level fluctuations occurring within a biozone. Based on Pleistocene studies, glacial-interglacial and thus sea-level changes can occur on at least a 50,000 year cycle (Ruddiman et al., 1986). This procedure, at a minimum, would approximate the 3rd-order sea-level curves of Vail et al. (1977) that are of 1 to 10 m.y. in duration.
Accordingly, the Dob's Linn section meets both the climatic requirements and the chemical requirements to test the Ce anomaly ability to track eustatic sea-level change.
5. Results and interpretation
110 Field-identified shales were analyzed at Los Alamos National Laboratory using neutron activation analysis (Quinby-Hunt et al., 1989; data available on request). Selective Dob's Linn samples have been analyzed for phosphate using inductively coupled plasma analysis (at Acme Analytical Laboratories, provided by Santa Fe Mining). This was done to confirm our observations that conodont apatite was not a factor in the interpretation of the anomaly. If significant amounts of P were from organic phosphates from conodonts or shell material they would correlate with more negative Ce anomalies (Wright et al., 1986; Elderfield and Pagett, 1986).We have calculated the cerium anomaly in these stratigraphically ordered shales from fourteen graptolite biozones from late-middle Ordovician (N. gracilis) into the Early Silurian (R. maximus) (Fig. 1). The curve shown in Fig. 1 is calculated by using a Spencer 21-term filter (Davis and Sampson, 1973, p. 226; Wilde et al., 1988) rather than averages to enhance any cyclic climatic signal. Both the individual sample values and a plot of filtered values (Wilde et al., 1988) are shown. Due to the assumptions of the filtering technique (Davis and Sampson, 1973), values from the zones at either end of the section should be ignored, that is post-conuolutus and pre-bicornis /gracilis. The argument developed above shows that: (1) a positive-trending whole-rock cerium anomaly indicates more oxic conditions or a sea-level fall; and (2) a negative-trending anomaly indicates more reducing conditions or a sea-level rise. Thus at Dob's Linn during the Ordovician, cerium anomaly maxima are shown for extraordinarius and complanatus zones, and at the cligani/bicornis boundary. These would correspond to glacial eustatic lowest stands of sea level. Minima occur in anceps and the linearis zones, reflecting high stands of sea level during interglacial times. In the latest Ordovician ( persculptus), relatively stable non-glacial climates and a return to high stands of sea level are reflected in the more negative values. The gradual increase in the anomaly at the top of the section reflects tectonic uplift due to closing of Iapetus (Scotese, 1986) which would appear as a pseudo-lowering of sea level as described in Case A above. The division of the Late Ordovician glaciation into separate pulses confirms the sedimentological interpretations of Brenchley (1984; Brenchley and Newell, 1984) from the Oslo region.
6. Summary
The whole-rock Ce anomaly in anoxic sediments is shown to be a potentially useful geochemical tool in interpreting relative 3rd-order (1 to 10 m.y. resolution) eustatic sea-level changes based on climatic changes under the proper constraints. The method is valid only during interval of tectonic stability, in areas of low rates of sedimentation in relatively deep waters of the outer shelf and continental margins where the pycnocline and lower mixed layer are anoxic and organically mineralized phosphate from conodonts or shell is not significant. In the test section at Dob's Linn, Scotland, the whole-rock Ce anomaly distinguished during the Late Ordovician discrete glacial pulses (regressions) separated by inter-glacial to non-glacial (transgressive) intervals. Because of the constraints mentioned above, the whole-rock Ce anomaly should be used only in well-studied detailed sections where other sedimentological, paleontological, tectonic and chemical information is available to ensure the validity of the method.The requirement of anoxic pelagic to hemipelagic sediments with anoxic overlying waters restricts the use of this method to the preCarboniferous and to subsequent intervals when anoxic waters returned to the lower mixed layer and the pycnocline (Wilde and Berry, 1982). For appropriate older sections in the Paleozoic, the whole-rock Ce anomaly could provide 3rd-order sea-level curves where the Vail et al. (1977) approach is not valid. In younger sections, this technique might be used to calibrate 3rd-order Vail curves.
Fluctuations in the cerium anomaly might be used to decipher transgressive-regressive phases in terrestrially derived near-shore marine sediments as suggested by the mechanisms outlined above by R.A. Schmitt (pers. commun. 1989). An empirical geochemical test in a sedimentologically well-documented section, like the one described here at Dob's Linn for pelagic sediments or one with a well-understood seismic stratigraphy, is indicated.
Acknowledgements
We thank the Institute of Geophysics and Planetary Physics of the University of California for financial support, the Los Alamos Research Reactor Group (INC-5), and, particularly, Dr. Carl Orth for the analyses of elemental abundance. Most of the samples were collected by Prof. W.B.N. Berry, whom we thank for this contribution. Other samples were cuts from the Bulman Collection of the Sedgwick Museum, Cambridge University. Their support is gratefully acknowledged. E. Bloomsteinof Santa Fe Pacific Mining thoughtfully provided the phosphate analyses which helped to calm our doubts about the whole-rock method. We wish to thank Judith Wright whose work on the cerium anomaly in apatitic fossils led to our development of the whole-rock methods for non-carbonate, non-phosphatic rocks. Prof. R.A. Schmitt was most kind in reviewing an early draft of this paper and providing us with his wonderful insights and possible alternative explanations. Profs. A. Hallam and J. Veizer were generous in giving their time for the latest review. An unnamed reviewer also provided critical comments, which were carefully considered for the final draft. M. Krup again did her usual outstanding job in preparing the illustrations. This paper was originally given orally by P. Wilde at the 12th International Geological Congress, Kyoto, Japan in August 1992 and at a poster session version at the Anoxic Conference given by the Geological Society of London in June 1989. B.-D. Erdtmann presented some of this material at several lectures in China in 19891990. This is contribution MSG-93-001 of the Marine Sciences Group.
References
Anders, E. and Ebihara, M., 1982. Solar-system abundances of the elements. Geochim. Cosmochim. Acta, 46: 23632380.Anderson, J.J., 1982. The nitrite-oxygen interface at the top of the oxygen minimum zone in the eastern tropical Pacific. Deep-Sea Res., 29: 1193-1201.
Andersson, A., Dahlman, B. and Gee, D.G., 1982. Kerogen and uranium resources in the Cambrian alum shales of the Billingen-Falbygden and Norske areas, Sweden. Geol. Foren. Stockholm Forh., 104: 197-209.
Baas-Becking, L.G.M., Kaplan, I.R. and Moore, D., 1960. Limits of the natural environment in terms of pH and oxidation potential. J. Geol., 68: 243-284.
Bacon, M.P., Brewer, P.G., Spencer, D.W., Murray, J.W. and Goddard, J., 1980. Lead-210, polonium-210, manganese and iron in the Cariaco Trench. Deep-Sea Res.,27: 119135.
Berner, R.A. and Canfield, D.E., 1989. A new model for atmospheric oxygen over Phanerozoic time. Am. J. Sci., 89: 333-361.
Berry, W.B.N. and Boucot, A.J., 1973. Glacio-eustatic control of late Ordovician-Early Silurian platform sedimentation and faunal changes. Bull. Geol. Soc. Am, 84: 275-283.
Berry, W.B.N., Wilde, P., Quinby-Hunt, M. S. and Orth, C.J . , 1986. Trace elements signatures in dictyonema shales and their geochemical and stratigraphic significance. Nor. Geol. Tidsskr., 66: 45-51.
Berry, W.B.N. , Quinby-Hunt, M. S. , Wilde, P. and Orth, C.J ., 1987. Use of the cerium anomaly in black shales—climatic interpretation in the Ordovician-Silurian boundary interval, Dob's Linn, Scotland. Geol. Soc. Am. Annul Mtg., 19: 587.
Berry, W.B.N., Wilde, P. and Quinby-Hunt, M.S., 1989. Paleozoic (Cambrian through Devonian) anoxitropic biotopes. Palaeogeogr., Palaeoclimatol., Palaececol., 74: 3-13.
Brass, G.W., Saltzman, E., Sloan, J.L., Southam, J.R., Hay, W.W., Holzer, W.T. and Peterson, W.H., 1982. Ocean circulation, plate tectonics, and climate. Climate in Earth History, National Academy Press, Washington, D.C., pp. 83-89.
Brenchley, P.J., 1984. Late Ordovician extinctions and their relationship to the Gondwana glaciation. In: P.J. Brenchley (Editor), Fossils and Climate. Wiley, New York, N.Y., pp. 291-315.
Brenchley, P.J. and Newell, G., 1984. Late Ordovician environmental changes and their effect on faunas. In: D.L. Bruton (Editor), Aspects of the Ordovician System. Palaeontol. Contrib. Univ. Oslo, 295: 65-79.
Budyko, M.L., Ronov, A.B. and Yanshin, A.L., 1987. History of the Earth's Atmosphere. Translated from Russian by S.F. Lemeshko and V.G. Yanuta. Springer-Verlag, Berlin, 139 pp.
Davis, J.C. and Sampson, R.J., 1973. Statistics and Data Analysis in Geology. Wiley, New York, N.Y., 550 pp.
DeBaar, H.J.W., 1991. On cerium anomalies in the Sargasso Sea. Geochim. Cosmochim. Acta, 55: 2981-2983.
DeBaar, H.J.W., Bacon, M.P., Brewer, P.G. and Bruland, K.W., 1985. Rare earth elements in the Pacific and Atlantic Oceans. Geochim. Cosmochim. Acta, 49: 1943-1959.
DeBaar, H.J.W., German, C.R., Elderfield, H. and van Gaans, P., 1988. Rare earth element distributions in anoxic waters of the Cariaco trench. Geochim. Cosmochim. Acta, 52: 1203-1219.
Elderfield, H. and Greaves, M.J., 1982. The rare earth elements in seawater. Nature, 296: 214-219.
Elderfield, H. and Pagett, R., 1986. REE in icthyoliths: variations with redox conditions and depositional environment. Sci. Total Environ., 49: 175-197.
Elderfield, H. and Sholkovitz, E.R., 1987. Rare earth elements in the pore waters of reducing near-shore sediments. Earth Planet. Sci. Lett., 82: 280-288.
Elderfield, H., Hawkesworth, C.J., Greaves, M.J. and Calvert, S.E., 1981a. Rare earth element geochemistry of oceanic ferromanganese nodules and associated sediments. Geochim. Cosmochim. Acta, 45: 513-528.
Elderfield, H., Hawkesworth, C.J., Greaves, M.J. and Calvert, S.E., 1981b. Rare earth element zonation in Pacific ferromanganese nodules. Geochim. Cosmochim. Acta, 45: 1231-1234.
Fortey, R.A., 1984. Global earlier Ordovician transgressions and regressions and their biological implications. In: D.L. Bruton (Editor), Aspects of the Ordovician System. Palaeontol. Contrib. Univ. Oslo, 295: 37-50.
Frakes, L.A., 1979. Climates Throughout Geologic Time. Elsevier, Amsterdam, 310 pp.
Frakes, L.A., Francis, J.E. and Syktus, J.I., 1992. Climate Modes of the Phanerozoic: the History of the Earth's Climate over the past 600 Million Years. Cambridge University Press, Cambridge, 274 pp.
German, C.R. and Elderfield, H., 1989. Rare earth elements in Saanich Inlet, British Columbia, a seasonally anoxic basin. Geochim. Cosmochim. Acta, 53: 2561-2571.
German, C.R. and Elderfield, H., 1990. Rare earth elements in the NW Indian Ocean. Geochim. Cosmochim. Acta, 54: 1929-1940.
German, C.R., Holliday, B.P. and Elderfield, H., 1991. Redox cycling of rare earth elements in the suboxic zone of the Black Sea. Geochim. Cosmochim. Acta, 55: 3553-3558.
Goldberg, E., 1961. Chemistry in the oceans. In: M. Sears (Editor), Oceanography. American Association of Advanced Science, Washington, D.C., pp. 583-597.
Grandjean-Lecuyer, P., Feist, R. and Albarede, F., 1993. Rare earth elements in old biogenic apatite. Geochim. Cosmochim. Acta, 57: 2507-2514.
Grasshoff, K., 1975. The hydrochemistry of landlocked basins and fjords. In: J.P. Riley and G. Skirrow (Editors), Chemical Oceanography, Vol. 2, 2nd ed. Academic Press, London, pp. 455-597.
Gromet, L.P., Dymek, R.F., Haskin, L.A. and Korotev, R.L., 1984. The 'North American shale composite': its compilation, major and trace element characteristics. Geochim. Cosmochim Acta, 48: 2469-2482.
Hallam, A., 1992. Phanerozoic Sea-Level Changes. Columbia University Press, 266 pp.
Haskin, L.A., Frey, R.A., Schmitt, F.A. and Smith, R.H., 1966. Meteoric, solar and terrestrial rare-earth distributions. In: L.H. Ahrens, F. Press, S.K. Runcorn and H.C. Urey (Editors), Physics and Chemistry of the Earth. Pergamon, Oxford, pp. 167-321.
Jacobs, L., Emerson, S. and Skei, J., 1985. Partitioning and transport of metals across the O2 /H2S interface in a permanently anoxic basin: Framvaren Fjord, Norway. Geochim. Cosmochim. Acta, 49: 1433-1444.
Jacobs, L., Emerson, S. and Huested, S.S., 1987. Trace metal geochemistry in the Cariaco Trench. Deep-Sea Res., 34: 965-981.
Leggett, J.K., 1985. Deep-sea pelagic sediments and palaeooceanography: a review of recent progress. In: P.J. Brenchley and B.P.J. Williams (Editors), Sedimentology, Recent Developments and Applied Aspects> Geol. Soc. London, Spec. Publ., 18: 95-118.
Leggett, J.K., McKerrow, W.S., Cocks, L.R.M. and Rickards, R.B., 1981. Periodicity in the early Paleozoic marine realm. J. Geol. Soc. London, 138: 167-176.
Liu, Y.-G., Miah, M.R.U. and Schmitt, R.A., 1988. Cerium: a chemical tracer for paleo-oceanic redox conditions. Geochim. Cosmochim. Acta, 52: 1361-1371.
Liu, Y.-G., Miah, M.R.U. and Schmitt, R.A., 1989. Author's Reply. Reliability of the reported stability constant for CePO , as related to Ce redox formulations in seawater. Geochim. Cosmochim. Acta, 53: 1477-1479.
MacKenzie, F.T. and Pigott, J.D. , 1 982. Phanerozoic ocean atmosphere-sediment history; secular variations and tectonic control. Int. Congr. Sedimentol., 11: 114.
Murray, R.W., Buchholz ten Brink, M.R., Jones, D.L., Gerlach, D.C. and Russ, G.P., 111, 1990. Rare earth elements as indicators of different marine depositional environments in chert and shale. Geology, 18: 268-271.
Murray, R.W., Buchholz ten Brink, M.R., Gerlach, D.C., Russ, G.P., 111 and Jones, D.L., 1991a. Rare earth, major, and trace elements in chert from the Franciscan complex and Monterey Group: assessing REE sources to finegrained marine sediments. Geochim. Cosmochim. Acta, 55: 1875-1895.
Murray, R.W., Buchholz ten Brink, M.R., Brumsack, H.J., Gerlach, D.C. and Russ, G.P., 111, l991b. Rare earth elements in Japan Sea sediments and diagenetic behavior of Ce/Ce*: results from ODP leg 127. Geochim. Cosmochim. Acta, 55: 2453-2466.
Murray, R.W., Buchholz ten Brink, M.R., Gerlach, D.C., Russ, G.P., 111 and Jones, D.L., 1992a. Rare earth major, and trace element composition of Monterey and DSDP chert and associated host sediment: assessing the influence of chemical fractionation during diagenesis. Geochim. Cosmochim. Acta, 56: 2657-2671.
Murray, R.W., Buchholz ten Brink, M.R., Gerlach, D.C., Russ, G.P., 111 and Jones, D.L., 1992b. Interoceanic variation in the rare earth, major, and trace element depositional chemistry of chert: perspectives gained ffom the DSDP and ODP record. Geochim. Cosmochim. Acta, 56: 1897-1913.
Piper, D.Z., 1974. Rare earths in the sedimentary cycle: a summary. Chem. Geol., 14: 285-304.
Quinby-Hunt, M.S. and Wilde, P., 1994. Thermodynamic zonation in the black shale facies based on ironmanganese-vanadium content. Chem. Geol.., 113: 297-317.
Quinby-Hunt, M.S., Wilde, P. and Berry, W.B.N., 1989. Elemental geochemistry of low calcic black shales-statistical comparison with other shales. U.S. Geol. Surv. Circ., 1037: 8-15.
Quinby-Hunt, M.S., Wilde, P. and Berry, W.B.N., 1991. The provenance of low-calcic black shales. Mineral. Deposita, 26: 113-121.
Ruddiman, W.F., Shackleton, N.J. and Mclntyre, A., 1986. North Atlantic sea-surface temperatures for the last 1.1 million years. In: C.P. Summerhayes and N.J. Shackleton (Editors), North Atlantic Palaeoceanography. Geol. Soc. London, Spec. Publ., 21: 155-173.
Scotese, C.R., 1986. Phanerozoic reconstructions: a new look at the assembly of Asia. Univ. Texas Inst. Geophys. Tech. Rep. 66, 54 pp.
Shaw, H. F. and Wasserburg, G.J., 1 985 . Sm-Nd in marine carbonates and phosphates: implications for Nd isotopes in seawater and crustal ages. Geochim. Cosmochim. Acta, 49: 503-518.
Sholkovitz, E.R., 1988. Rare earth elements in the North Atlantic Ocean, Amazon Delta and East China Sea: reinterpretation of terrigenous input patterns to the oceans. Am. J. Sci., 288: 236-281.
Sholkovitz, E.R. and Elderfield, H., 1988. Cycling of dissolved rare earth elements in Chesapeake Bay. Global Biogeochem. Cycles, 2: 157-176.
Sholkovitz, E.R. and Schneider, D.L., 1991. Cerium redox cycles and rare earth elements in the Sargasso Sea. Geochim. Cosmochim. Acta, 55: 2737-2743.
Skirrow, G., 1975. The dissolved gases—carbon dioxide. In: J.P. Riley and G. Skirrow (Editors), Chemical Oceanography, Vol. 2, 2nd ed. Academic Press, London, pp. 1-192.
Summerhayes, C.P., 1986. Sea-level curves based on seismic stratigraphy: their chronostratigraphic significance. Palaeogeogr., Palaeoclimat., Palaeoecol., 57: 27-42.
Thomson, J., Carpenter, M.S.N., Colley, S., Wilson, T.R.S., Elderfield, H. and Kennedy, H., 1984. Metal accumulation rates in northwest Atlantic pelagic sediments. Geochim. Cosmochim. Acta, 48: 1935-1948.
Vail, P.R., Mitchum, R.M. and Thompson, S., 111, 1977. Seismic stratigraphy and global changes of sea-level, IV. Global cycles of relative changes of sea-level. In: C. Payton (Editor), Seismic Stratigraphy—Applications to Hydrocarbon Exploration. Am. Assoc. Pet. Geol. Mem., 26: 83-97.
Wang, Y.L., Liu, Y.-G. and Schmitt, R.A., 1986. Rare earth element geochemistry of South Atlantic deep sea sediments: Ce anomaly change at ~ 54 My. Geochimi. Cosmochim. Acta, 50: 1337-1355.
Watts, A.B., 1982. Tectonic subsidence, flexure and global changes of sea-level. Nature, 297: 469-474.
Wilde, P., 1987. Model of progressive ventilation of the Late Precambrian-Early Paleozoic Ocean. Am. J. Sci., 287: 442-459.
Wilde, P., 1991. Oceanography in the Ordovician. In: C.R. Barnes and S.H. Williams (Editors), Advances in Ordovician Geology. Geol. Surv. Can. Pap., 90-9: 283-298.
Wilde, P. and Berry, W.B.N., 1982. Progressive ventilation of the Oceans, 11. Potential for return to anoxic conditions in the post-Paleozoic. In: S.O. Schlanger and M.B. Cita (Editors), Nature and Origin of Cretaceous Carbon-Rich Facies. Academic Press, New York, N.Y., pp. 209-224.
Wilde, P. and Berry, W.B.N., 1984. Destabilization of the oceanic density structure and its significance to marine "extinction" events. Palaeogeogr., Palaeoclimatol., Palaeoecol., 48: 143-162.
Wilde, P., Orth, C.J., Berry, W.B.N., Quinby-Hunt, M.S., Quintana, L.R. and Gilmore, J.S., 1986. Iridium abundances across the Ordovician-Silurian stratotype. Science, 233: 399-341.
Wilde, P., Berry, W.B.N. and Quinby-Hunt, M.S., 1988. The use of statistical filters on chemical data in determining chronozonal boundaries. Geol. Soc. Am. Annu Mtg., 20(7): A219.
Wilde, P., Quinby-Hunt, M.S., Berry, W.B.N. and Orth, C.J., 1989. Palaeo-oceanography and big-geography in the Tremadoc (Ordovician) lapetus Ocean and the origin of the chemostratigraphy of Dictyonema flabelliforme black shales. Geol. Mag., 126: 19-27.
Wright, J., Miller, J.F. and Holser, W.T., 1986. Conodont chemostratigraphy across the Cambrian-Ordovician boundary: western USA and southeast China. In: R.L. Austin (Editor), Conodonts: Investigative Techniques and Applications. Ellis Horwood Ltd., Chichester, pp. 256-283.
Wright, J., Schrader, H. and Holser, W.T., 1987. Paleoredox variations in ancient oceans recorded by rare earth elements in fossil apatite. Geochim. Cosmochim. Acta, 51:631-644.
Yen, T.F., 1975. Chemical aspects of metals in native petroleum. In: T.F. Yen (Editor), The Role of Trace Metals in Petroleum. Ann Arbor Science Publishers, Ann Arbor, Mich., pp. 1-30.
Zhou, J.Y., McDuff, R. and Murray, J.W., 1982. The distribution of vanadium, chromium and manganese in the northwest Pacific. Eos (Trans. Am. Geophys. Union), 63:989-990 (abstract).