An integrated kinematic and geochemical model to determine lithospheric extension and mantle temperature from syn-rift volcanic compositions
Introduction
Melting of the mantle occurs at rifted continental margins in response to decompression as mantle rocks are brought toward the surface by lithospheric extension. This melting is controlled by several parameters including: the initial mantle temperature and source composition; the rate of continental extension; the final degree of lithospheric stretching (βmax); and the initial lithospheric thickness (Bown and White, 1995). Several studies have attempted to estimate some of these parameters from estimates of rift-related magmatic volume (e.g., (Whitmarsh et al., 2001, Hopper et al., 2003, Barton and White, 1997, Kelemen and Holbrook, 1995, Niu and Batiza, 1991)). This volume is usually estimated from wide-angle seismic data. Igneous products, added to the lower crust (i.e., underplated or intruded complexes) are identified as regions of high seismic velocity (White and McKenzie, 1989) relative to normal continental crust (e.g., (Christensen and Mooney, 1995)). Extrusive volcanics (i.e., flood basalts) are identified as high amplitude seaward dipping reflector sequences on the continent-ocean transition. Examples of both intrusive and extrusive magmatic rocks are found on “volcanic” margins such as the East Greenland margin (Korenaga et al., 2000); (Holbrook et al., 2001) the Faeroes margin and Hatton Bank (White et al., 2008). In contrast, such features are absent at “non-volcanic” rifted margins, and the presence at some of these margins of exhumed serpentinized mantle, well-documented by both geophysical studies and by direct sampling in the southern Iberia Abyssal Plain (e.g., (Dean et al., 1999); (Whitmarsh et al., 2001)), shows that melting can be suppressed by different syn-rift conditions.
In both types of rifted continental margin, there are difficulties in measuring accurately the volume of magmatic products using seismic reflection techniques alone. This is due to the high impedance contrast between magmatic intrusions and the surrounding lower continental crustal material, as well as limited spatial resolution, which makes it difficult to image small and discontinuous bodies of intruded material. However, it is possible to put constraints on the total intrusive volume from the average lower-crustal velocity if the end member velocities of the pre-existing crust and the intrusive rocks are known (White et al., 2008). More importantly, the volume of magmatic products is not constrained uniquely by the tectonic process alone; rather it is also subject to variations in mantle temperature, composition and strain rate.
An alternative approach to determining the rifting conditions (encompassing tectonic geometry, strain rate, mantle temperature and composition) is to use the composition of the volcanic products and, where available, upper mantle residues. (Williamson et al., 1995) showed, through modelling, how the rare earth element compositions of syn-rift volcanics on the Labrador margin of eastern Canada and the North Sea Rift are sensitive to mantle temperature, the degree of stretching and the duration of rifting. Using a similar principle, we have developed an integrated kinematic and geochemical model to determine the geochemical compositions of syn-rift melts and their residual mantle source rocks under a variety of rifting conditions. From this model, we demonstrate that the major, minor and rare-earth element geochemistry of the syn-rift melts and their mantle residues are sensitive independently to the extension history and mantle temperature. By inverting the approach, we compare the compositions of syn-rift volcanic products from around the North Atlantic margin with those predicted by our model and hence to infer the most probable rifting conditions prevalent during their genesis.
Section snippets
Rifted margin melting model
Various tectonic models for lithospheric stretching at continental margins have been proposed including pure shear models (McKenzie, 1978), simple shear models (Wernicke, 1985), and models involving depth-dependent stretching in different forms (e.g., (Davis and Kusznir, 2004)). The simplest models to account for the main features of most rifted margins assume pure shear at a uniform and finite rate. We employ one of these variants, the one-dimensional lithospheric stretching model of (Bown and
Major element composition
Three different major element parameterisations have been implemented (Watson and McKenzie, 1991); (Niu and Batiza, 1991); (Niu, 1997). The first two use similar methods to determine melt composition from experimentally derived partition coefficients. In contrast, (Watson and McKenzie, 1991) use empirical fits to data, from laboratory melting experiments, to derive functions, thus describing effective bulk partition coefficients of each element and the source mantle mineralogy as melting
Rare earth element composition
The REE composition is determined using the parameterisation of (McKenzie and O'Nions, 1991). (McKenzie and O'Nions, 1991) depleted earth source composition is given in Table 2. The composition of the melt and residue is determined using the same equations as for the major elements in (Niu and Batiza, 1991) and (Niu, 1997). The partition coefficients are dependent on the plagioclase, spinel and garnet stability fields. (McKenzie and O'Nions, 1991) provide the proportion of each mineral present
Model results
We have run our model for a range of mantle potential temperatures (1200–1500 °C) to a stretching factor βmax = 50, which is found to simulate infinite stretching and hence steady-state melt production at a mid-ocean ridge. To explore the effect of strain rate on melt volume and composition, we have, for each temperature step, run the model for a range of rift durations of 1–50 m.y. Results for long rift durations describe behaviour at low strain rates, while results for short durations describe
Comparison with actual rocks
In the following section, we compare the results of our model against actual volcanic products recovered from rifted margins. Like most mantle melting models, our model predicts the composition of primary melts, generated under a variety of conditions. These melts undergo variable degrees of fractional crystallisation before being erupted (Kinzler and Grove, 1992); (Stolper, 1980), the effects of which are to increase the concentrations of incompatible elements and reduce the concentrations of
Summary and conclusions
We have developed a numerical model that integrates kinematic and geochemical parameters during continental stretching and rifting to allow us to predict the composition of magmatic products generated under a range of continental rifting conditions. Using a modular scheme, we have integrated the lithospheric stretching model of (Bown and White, 1995), with three different major element parameterisations: (Watson and McKenzie, 1991); (Niu and Batiza, 1991) and (Niu, 1997). We also incorporate a
Acknowledgements
This project was funded by the Natural Environment Research Council's Ocean Margins (LINK) programme, grant reference number NER/T/S/2000/00650.
References (47)
- et al.
Modelling the composition of melts formed during continental breakup of the Southeast Greenland margin
Earth Planet. Sci. Lett.
(2008) - et al.
U–238–Th–230 disequilibria and mantle melting processes: a discussion
Chem. Geol.
(2000) Experimental melting studies on a model upper mantle composition at high pressure under water-saturated and water-unsaturated conditions
Earth Planet. Sci. Lett.
(1973)- et al.
The fingerprint of seawater circulation in a 500-meter section of ocean crust gabbros
Geochim. Cosmochim. Acta
(1999) - et al.
Mantle thermal structure and active upwelling during continental breakup in the North Atlantic
Earth Planet. Sci. Lett.
(2001) - et al.
Decompression melting at rifted margins: comparison of model predictions with the distribution of igneous rocks on the eastern Canadian margin
Earth Planet. Sci. Lett.
(1994) Some remarks on the development of sedimentary basins
Earth Planet. Sci. Lett.
(1978)- et al.
On the regional variations of heat flow, geotherms, and lithospheric thickness
Tectonophysics
(1977) - et al.
Melt extraction in the Earth's mantle: constraints from U–Th–Pa–Ra studies in oceanic basalts
Earth Planet Sci. Lett.
(2006) - et al.
Crustal structure of the Edoras Bank margin and mantle thermal anomalies beneath the North Atlantic
J. Geophys. Res.
(1997)
The effect of finite extension rate on melt generation at continental rifts
J. Geophys. Res.
Data report: composition of basaltic lavas from the seaward-dipping reflector sequence recovered during deep sea drilling project leg 81 (Hatton Bank)
From continental extension to seafloor spreading: crustal structure of the Goban Spur rifted margin, southwest of the UK
Geophys. J. Int.
Seismic velocity structure and composition of the continental crust: a global view
J. Geophys. Res.
Depth-dependent lithospheric stretching at rifted continental margins
Deep structure of the ocean-continent transition in the southern Iberia Abyssal Plain: II. The IAM-9 transect at 40°20'N
J. Geophys. Res.
Volcanic rocks from the southeast Greenland margin at 63°N: composition, petrogenesis, and mantle sources
Petrogenesis of mid-ocean ridge basalts
Structure of the SE Greenland margin from seismic reflection and refraction data: implications for nascent spreading center subsidence and asymmetric crustal accretion during North Atlantic opening
J. Geophys. Res.
Anhydrous partial melting of peridotite at 0–15 kb pressure and the genesis of tholeiitic basalts
Contrib. Miner. Petrol.
Origin of thick, high-velocity igneous crust along the US East-Coast Margin
J. Geophys. Res.
Primary magmas of Mid-Ocean ridge basalts 2. Applications
J. Geophys. Res.
Global correlations of ocean ridge basalt chemistry with axial depth and crustal thickness
J. Geophys. Res.
Cited by (6)
Investigating the Goban Spur rifted continental margin, offshore Ireland, through integration of new seismic reflection and potential field data
2020, TectonophysicsCitation Excerpt :Superimposed on these plate motions, the variable widths of each of the crustal domains across the two margins may also reflect highly variable rifting rates. At the Goban Spur, lower mantle temperatures are supported by geochemical models, suggestive of relatively slower rifting than along other northern Atlantic margins (Dean et al., 2009). Meanwhile, inferred complexities in the tectonic processes along the northeastern Flemish Cap margin also make it difficult to determine the rifting rate.
Upper mantle temperature and the onset of extension and break-up in Afar, Africa
2015, Earth and Planetary Science LettersCitation Excerpt :This material, due to its buoyancy, does not participate in the melting as it remains above the solidus and is moved to the sides due to the divergent boundary condition. Partitioning of REEs between the solid mantle and partial melt is calculated from the melt depletion, temperature and pressure within the melt region assuming incremental batch melting (see Dean et al., 2008; Gibson and Geist, 2010; Armitage et al., 2011; and the Supplementary Material). Geochemical and isotopic evidence from erupted melts shows that mantle source beneath Afar is fertile compared to the depleted upper mantle source of MORBs (e.g. Schilling, 1973; Schilling et al., 1992; Barrat et al., 1998; Rooney et al., 2012a; Ferguson et al., 2013), and that this fertile source has been a long-lived feature of the mantle here (Pik et al., 1999).
The Role of Crustal Strength in Controlling Magmatism and Melt Chemistry During Rifting and Breakup
2018, Geochemistry, Geophysics, GeosystemsThin oceanic crust and flood basalts: India-Seychelles breakup
2012, Geochemistry, Geophysics, GeosystemsLithospheric controls on melt production during continental breakup at slow rates of extension: Application to the North Atlantic
2009, Geochemistry, Geophysics, Geosystems