Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 375 (2094)

Global Water Cycle and the Coevolution of the Earth's Interior and Surface Environment

Affiliations

Global Water Cycle and the Coevolution of the Earth's Interior and Surface Environment

Jun Korenaga et al. Philos Trans A Math Phys Eng Sci.

Abstract

The bulk Earth composition contains probably less than 0.3% of water, but this trace amount of water can affect the long-term evolution of the Earth in a number of different ways. The foremost issue is the occurrence of plate tectonics, which governs almost all aspects of the Earth system, and the presence of water could either promote or hinder the operation of plate tectonics, depending on where water resides. The global water cycle, which circulates surface water into the deep mantle and back to the surface again, could thus have played a critical role in the Earth's history. In this contribution, we first review the present-day water cycle and discuss its uncertainty as well as its secular variation. If the continental freeboard has been roughly constant since the Early Proterozoic, model results suggest long-term net water influx from the surface to the mantle, which is estimated to be 3-4.5×1014 g yr-1 on the billion years time scale. We survey geological and geochemical observations relevant to the emergence of continents above the sea level as well as the nature of Precambrian plate tectonics. The global water cycle is suggested to have been dominated by regassing, and its implications for geochemical cycles and atmospheric evolution are also discussed.This article is part of the themed issue 'The origin, history and role of water in the evolution of the inner Solar System'.

Keywords: continental freeboard; mantle convection; oceans.

Conflict of interest statement

We declare we have no competing interests.

Figures

Figure 1.
Figure 1.
Schematic for the set-up of freeboard modelling. (a) The top part of the model. The zero-height continental section corresponds to where the present-day sea-level crosses continental topography. A positive h reduces the area of exposed landmasses Aex. (b) The entire model view. Note that in this paper we show the hypsometry of ocean basins in a reverse order (i.e. decreasing topography with increasing cumulative surface area) to indicate a likely passive margin structure.
Figure 2.
Figure 2.
(a) Two contrasting thermal histories (blue for slower plate tectonics (PT) and green for faster plate tectonics in the past, both taken from [2]) compared with the variation of mantle potential temperatures estimated for the last 3.5 Gyr (dots) [52]. These estimates are based on the composition of primary melts for Precambrian non-arc basalts, and their overall trend converges smoothly to the potential temperature of present-day MORB source (approx. 1350°C [53]), supporting the interpretation that the non-arc basalts formed by the melting of ambient mantle. (b) Corresponding surface heat flux (solid) and internal heat production (dashed). (c) Corresponding plate velocity. (d) Continental growth models: [54] (yellow), [55] (black) and [56] (magenta).
Figure 3.
Figure 3.
(a) Thicknesses and (b) densities of oceanic crust (cyan) and depleted oceanic lithospheric mantle (blue) as a function of mantle potential temperature. After [60].
Figure 4.
Figure 4.
Representative results of freeboard modelling with slower plate tectonics in the past with the continental growth models of (a,d,g) Armstrong [54], (b,e,h) McLennan & Taylor [56] and (c,f,i) Campbell [55]. (ac) Sea level, (df) the areal fraction (with respect to the Earth’s surface area) of exposed continents above sea level, and (gi) the zero-age depth of seafloor. In all panels, thick black curves denote a reference case, which assumes the net water influx of 3×1014 g yr−1, half-space cooling for seafloor subsidence, 5 km thicker continental crustal at 2.5 Ga, time-varying continental crust density, ΔρAcl of −5 kg m−3, and reduced continental topography in the past. Other curves denote sensitivity tests in which only one of those assumptions is modified: constant volume of water in the oceans (dashed cyan), the net water influx of 4.5×1014 g yr−1 (solid cyan), the plate model for subsidence (dashed tan), constant continent crustal thickness (solid orange), constant continental crustal density (dashed orange), ΔρAcl of 0 kg m−3 (solid green) and constant continental topography (dashed orange). In (df), thick dotted curves denote the total areal fraction of continental crust for the reference case. Even when continental mass does not change, this fraction can vary if continental thickness changes. The ridge depth shown in (gi) is d0h, i.e. the depth with respect to the sea level. The ridge shallowing for the case of constant ocean volume is caused by the lowering of the sea level. CL, continental lithosphere.
Figure 5.
Figure 5.
Same as figure 4 but with faster plate tectonics in the past.
Figure 6.
Figure 6.
Hypsometry snapshots from the reference cases shown in figures 4 and 5 at present (solid), 1 Ga (green), 2 Ga (cyan) and 3 Ga (magenta). Corresponding sea levels are shown in dashed lines. The hypsometry is measured with respect to the (present day) zero-height continental section, the relative location of which in the continental area does not change with time (figure 1a). The absolute height of the reference point, with respect to the centre of the Earth, varies to satisfy the conservation of mass in the solid Earth.
Figure 7.
Figure 7.
Compilation of epicontinental sedimentary records for exposed areas larger than 30 000 km2. The data are available from the authors upon request and will be described in detail in a forthcoming publication. Only broad trends are discussed herein.
Figure 8.
Figure 8.
Schematic of the links between continental emergence and major biogeochemical cycles. The light grey area represents modern continental crust and fluxes and the dark grey represents the Archean continental crust and fluxes. Limited continental exposure might have lead to high weathering intensities in a CO2 rich atmosphere and decreased phosphorus fluxes to the oceans. Lower phosphorous fluxes to the oceans would have limited primary productivity and oxygen release to the atmosphere. Traditionally, it was assumed there was greatly increased hydrothermal activity in the Archean, which would have reduced dissolved Mg levels, but there is no empirical support for this ‘fast tectonic’ model.

Similar articles

See all similar articles

Cited by 6 articles

See all "Cited by" articles

LinkOut - more resources

Feedback