RUSSIAN JOURNAL OF EARTH SCIENCES VOL. 7, ES3001, doi:10.2205/2005ES000179, 2005
[52] Data on shallow-water deposits and many other features indicate that variations in the sea level have existed throughout the geologic history of the Earth. A wide spectrum of sea level variation cycles of various periods has been established [Turcotte and Schubert, 1982; Veil et al., 1977]. Global sea level variations can be due to climatic processes, in particular, formation and thawing of glaciers, when water is first gathered on continents and then comes back into oceans. The ocean water volume can also vary if the amounts of water transported by magma from the mantle and escaping into the mantle in subduction zones are different. These processes change the sea level synchronously for all continents by up to a few hundred meters. Moreover, regional uplifts and subsidences of land due to tectonic processes are possible on each continent and their amplitudes also reach hundreds of meters. These changes can evidently be asynchronous in both continents and their regions.
![]() |
Figure 6 |
[54] We compare the origin of global sea level variations with theoretical inferences of the theory of floating continents, in particular, with the above results derived from a numerical evolution model involving two continents. The green-colored curves in the upper parts of Figure 2 and Figure 4 depict the model topography of the mantle surface with floating continents, (i.e. variations in the seafloor topography h(x) at various time moments).
[55] The sea level coincides with the geoid surface. The topographic inequalities of the geoid are on the order of a few tens of meters, whereas the amplitude of variations in the seafloor topography reaches a few kilometers. Therefore, given a constant ocean water volume, the averaged seafloor surface displaced by an average ocean depth can be taken, in a first approximation, as a sea level.
[56] In the absence of water, the mantle surface topography hm can be found from the relation connecting the surface height with the vertical stress hm=(p + Szz)/(rm g), where rm is the mantle density. Here, the pressure p and the stress Szz are calculated in the process of the solution of the convection equations (see Appendix). The topography h of the seafloor covered by water is calculated with regard for the water buoyancy force: h = (p+Szz)/(rm - rw)g, and its amplitude increases by about one-third. The floating continents rise or subside following the variations in the topography of the mantle surface relative to its average level. The continents jut out above the ocean surface. Variations in their position relative to the average ocean level are controlled by the mantle topography in the absence of water, rather than by the seafloor topography, because a rise in a continent decreases its buoyancy thereby compensating the difference between the topographies of the seafloor and the mantle surface.
[57] The black segments in Figure 2 and Figure 4 show schematically the position of continents relative to the average ocean level (the thickness of the continents is neglected here). Calculations of mantle convection still involve a number of uncertainties, and the goal of this study was to gain qualitative constraints rather than to attain numerical coincidence with results of observations.
[58] In our model, even at the moment when continents are placed onto the mantle, the surface of the latter (and, accordingly, the seafloor) is irregular. The seafloor is uplifted above mantle upwellings and subsides above downwellings. Model recalculations with parameters of the real mantle yield a variations in the amplitude of the seafloor topography of about 2-5 km. As is known, the ocean depth averages about 4 km (about 2 km in ridge zones and about 6 km in subduction zones). The model results reveal an interesting, previously unknown feature: since the continent sizes are comparable with the wavelength of the convecting mantle topography, a solid floating continent does not drift strictly horizontally but experiences variations in its tilts (as a ship at sea). Thus, if Eurasia were in the place of the Pacific plate, the difference between heights of its edges would amount to a few kilometers. Smoothing the topography of the mantle surface by floating continents decreases, to an extent, their tilts.
[59] As seen from Figure 2, the level of the newly formed supercontinent averages zero because the mantle downwelling attenuated and a new hot flow has not arisen as yet. However, before its breakup, the supercontinent as a whole is uplifted by about 0.5 km, being subjected to a bending force that appears to facilitate its subsequent breakup.
[60] As is evident from Figure 4, which shows the situation arising immediately after the breakup of the supercontinent, the mantle flow still uplifts the diverging continents. Figuratively speaking, the continents seem to go down from a hill. Finally, after a large ocean forms between the diverging continents, they subside to a zeroth level (Figure 4, t = 340 Myr).
[61] Thus, our model shows that ocean level fluctuations can be due to not only global variations in the water amount and other possible factors but also due to uplifts and subsidences of individual continents and should be regarded as relative, rather than absolute, variations. In a first approximation, various factors responsible for sea level fluctuations can be estimated additively. Note that the position of continents is measured from the mean level of the mantle surface calculated at a given time moment. This allows one to automatically take into account the fact that, when continents rise onto ridges, the displaced water moves toward deep sea trenches, and vice versa, when continents occupy the places of deep sea trenches, water is displaced toward mid-ocean ridges.
[62] This model also shows that variations in the sea level occur synchronously only during the lifetime of a supercontinent. In other epochs, different continents diverge at different velocities and overlie mantle surface depressions at different times.
[63] The model of mantle convection with floating continents presented in this paper is fairly general and actually independent of initial conditions. Given a different initial configuration, the continents also become, with time, close to a descending mantle flow, as was accepted in our calculations to save the computation time.
![]() |
Figure 7 |
[65] A giant plume that should have formed beneath Pangea at about 220-250 Ma uplifted the supercontinent and broke it up. For this reason, the sea level relative to Pangea dropped in the Mesozoic. The continents moved to their places near subduction zones at about 100 Ma and the relative sea level rose.
[66] Continents were also uplifted and the relative sea level was low in the Precambrian, during the breakup of Rodinia. Later, in the Paleozoic, when the continents were already located above mantle downwellings and therefore occupied a lower position, the sea level was high. A schematic curve of the sea level variation obtained from our modeling data correlated with time according to [Scotese et al., 1988] is shown in the upper panel of Figure 6.
[67] Overall, the calculated results qualitatively agree with observations, notwithstanding the simplicity of the model. However, a certain shift in phase is observed. According to the Veil curve, the sea level dropped in the Paleozoic and rose in the Mesozoic about 50 Myr earlier. The stratigraphic data used in [Veil et al., 1977] mainly characterize North America. However, the results of the continental drift reconstruction indicate that North America broke off from Pangea at about 175 Ma, whereas South America formed as late as 125 Ma. Consequently, the drift of North America in the Phanerozoic was ahead of other continents. As regards the present time, the present Eurasia is only approaching a subduction zone, the edge of South America overrode a subduction zone, and North America has already moved over a former mid-ocean ridge and this can lead to an uplift of this continent. Therefore, the sea level variations relative to North America (particularly, its western margin) can be some 50 Myr ahead of sea level variations relative to other continents.
[68] It is still unclear whether the sea level drop over the last 50 Myr fixed by the Veil curve is related to an uplift of the western part of the North American continent. The modeling results are only preliminary. On the other hand, the data of observations are averaged over all continents, whereas the model under consideration indicates that continents can be uplifted simultaneously but subsidence periods are specific to each continent.
[69] The interpretation proposed here for the global variation in the sea level on a time scale of a few hundred million years is at variance with the generally accepted ideas. Turcotte and Schubert [1982] and Schubert et al. [2001] explain a sea level rise in terms of water displaced by the system of mid-ocean ridges. The latter arise due to enhancement of tectonic activity caused by chaotic mantle convection.
[70] The geologic time scale of the Paleozoic, Mesozoic, and Cenozoic has long been used by geologists. One might try to correlate these periods with global variations in the sea level and epochs of great extinctions and revivals of the Earth's organic world. During and after the Rodinia breakup in the Precambrian, continents were uplifted and the primitive life developed very slowly. At about 650 Ma, the ocean rose by nearly 500 m, while the present land level amounts to about 800 m. Therefore, the ocean flooded nearly half of the land [Schubert et al., 2001; Turcotte and Schubert, 1982], and water ruined land protozoans and stimulated the development of new life forms, in particular, the marine animal kingdom prevailing in the Paleozoic.
[71] The ocean receded before the breakup of Pangea, approximately 220-250 Ma. The land area became 1.5 times larger [Turcotte and Schubert, 1982]. New lowlands should have been covered by luxuriant vegetation, which stimulated the development of numerous species of giant phytophagous land animals. Finally, at 60-100 Ma (the beginning of the Cenozoic), continents started again subsiding and the sea level rose by a few hundred meters. This resulted in extinction of the land species existing at that time. It is interesting that the sea level variations affected about 50% of land, and great biomass changes were also on the order of 50%. A curve of relative amounts of plant and animal species [Khain, 1994; Sepkoski, 1987] presented in the lower panel in Figure 7 correlate in part with the sea level variation curve.
[72] As is known, a giant asteroid fell onto the Earth at 65 Ma; it produced a 180-km crater and continental Ir anomalies. This should have led to a climate change and rapid extinction of many species, particularly, the largest ones. Actually, a sharp narrow minimum observed in the curve in Figure 7 exactly coincides in time with the asteroid fall, but the amplitude of the minimum is as low as about 10%. Possibly, the extinction of Mesozoic animals was gradual due to a slow rise in the sea level, and the asteroid brought this process to its accomplishment.
Citation: 2005), Evolution of mantle plumes and uplift of continents during the Pangea breakup, Russ. J. Earth Sci., 7, ES3001, doi:10.2205/2005ES000179.
Copyright 2005 by the Russian Journal of Earth Sciences (Powered by TeXWeb (Win32, v.2.0).