Craton is an ancient and stable continental block on the earth. The lithospheric mantle of the craton is a treasure house of diamonds, which was formed in Archean 2.5 billion years ago and Proterozoic 1.6-25 billion years ago (Peslier et al., 201; Li et al., 20 1 1). Craton-type lithospheric mantle has obvious characteristics: huge thickness (up to 300km), low density, loss of basaltic composition, high fire resistance, low temperature gradient (Lee et al., 20 1 1), and lithospheric mantle.
The lithospheric mantle of the ancient craton was formed after the crust-mantle differentiation event in the early stage of earth evolution. They are the residual refractory components (mainly refractory gabbro) after the original mantle (mainly composed of meltable lherzolite) is partially melted and a large amount of basaltic melt is extracted. Therefore, the lithospheric mantle of the ancient craton has obvious characteristics of rich magnesium, poor iron and high fire resistance. After basaltic melt is extracted in large quantities, the density of lithospheric mantle in ancient craton is less than that of the underlying asthenosphere, so it can "float" on the asthenosphere for a long time like a ship. The lithospheric mantle of craton has mechanical resistance (Arndt et al., 2009), which can prevent the destruction caused by mantle convection, so it can exist for a long time.
How was the lithospheric mantle of the ancient craton formed? Although scientists have done a lot of research on this issue, there are still obvious differences in understanding the genesis of lithospheric mantle in craton. The focus of disagreement is mainly two different academic viewpoints: melting mode and accumulation mode.
The first viewpoint (melting model) holds that the lithospheric mantle of craton is formed by the rising of mantle material diapir, mantle inversion or mantle plume, and the material and high melting degree of mantle transition zone (Boyd,1989; Stein and Hofmann, 1994), so they represent highly depleted basaltic melts, relatively dry and low-density molten residues. This model is easily accepted by everyone, because it can reasonably reveal many observed phenomena, such as the low density (the result of high melting) and buoyancy (Boyd,1989; Stein and Hofmann, 1994), the thickness of lithosphere is positively correlated with age, and the discovery of peridotite xenoliths with anhydrous mantle shows that the lithosphere mantle has enough viscosity and is not affected by the underlying asthenosphere (Peslier et al., 20 10). According to this model, it can be predicted that the density at the top of the mantle melting column should be relatively low and the density at the bottom of the lithosphere mantle should be relatively high because of the high melting degree of the upper mantle materials during decompression melting. However, the chemical stratification of lithosphere mantle predicted by melting model is quite different from the observation data of other scholars and the results predicted by working model (Lee et al., 20 1 1).
The second academic view (accumulation model) on the genesis of the lithospheric mantle of the craton holds that the lithospheric mantle of the ancient craton was originally formed by the superposition of subducted oceanic crust and its depleted mantle originally formed on the mid-ocean ridge (Helmstedt and Schulze,1989; Bill et al., 20 18). The model also successfully explains some geological phenomena, such as the huge thickness of the lithospheric mantle, the loss of basaltic melt in the lithospheric mantle relative to the marine lithosphere, and the geochemical characteristics of eclogite xenoliths in kimberlites in craton. However, there are some problems in this accumulation model, for example, it is difficult to explain the inconsistency between a relatively large amount of ancient oceanic crust materials in the marine lithosphere and relatively rare eclogite xenoliths in kimberlite mantle xenoliths (Arndt et al., 2009; Li et al., 20 1 1). In addition, because the ancient subduction plate contains a considerable thickness of oceanic crust, it may be affected by plate fracture and/or plate retraction, so that multiple accumulations cannot occur (Perchuk et al., 20 19).
In order to solve the above controversial scientific problems, a new alternative model-thermal-mechanical model (Sizova et al., 2010; Perchuk et al., 20 19), has been further developed recently. Recently, Perchuk et al. (2020) published a paper in Nature and established a two-dimensional high-resolution magma-thermal-mechanical model. Through the simulation study, they found that after the Archean plate structure started, the hot mantle layer with positive buoyancy and toughness and loss basaltic melt under the subduction ocean plate could not subduct synchronously with the ocean plate during the subduction process (Figure 1), and with the decrease of geothermal gradient, the high viscosity loss mantle layer invading the mainland also cooled down and merged with the mainland to form the lithospheric mantle of the ancient craton. According to the thickness estimation of continental lithosphere mantle, it shows that the dynamic mechanism demonstrated by this model is effective in the main formation period of ancient craton lithosphere. Therefore, the subduction of the oceanic plate with high depletion of the Precambrian mantle layer is a prerequisite for the formation of the extra-thick lithospheric mantle under the land. Due to the existence of ultra-thick lithospheric mantle, the ancient craton was preserved for a long time in the subsequent plate tectonic process. This study provides new evidence for the second view: the lithospheric mantle of ancient craton was formed by plate accumulation caused by early oceanic plate subduction.
Main references
The origin of Archean continental lithospheric mantle: some petrological constraints [J].Lithos, 2009,109 (1-2): 61-71.
Zhang Xuejun, Zhang Xuejun. Formation of Craton Lithosphere during the Tectonic Evolution of Chinese mainland Plate [J]. Chinese mainland Science Press, 2002. Geology, 2018,46 (6): 487-490.
Journal of China Geo University, 2003, 10( 1). Earth and Planetary Science Express,1989,96 (1-2):15-26.
Kimberlite and its mantle samples from South Africa: the significance of Archean structure and lithosphere evolution [J]. Kimberlite and related rocks, 1989, 1: 358-368.
Li Zhentai, Lu Feiping, Chen Yingjie. Formation and destruction of continental mantle [J]. China Geological Publishing House, 2002. Annual Journal of Earth and Planetary Science, 201,39: 59-90.
Zhang Zhiyong, et al. Craton keel in Precambrian plate tectonics [J]. Nature, 2020,586 (7829): 395-401.
Precambrian hotter mantle but colder subduction: what does it mean? [J]。 Precambrian research, 20 19, 330: 20-34.
Et al. Water content of olivine in Chinese mainland lithosphere and life span of craton [J]. Nature, 20 10/0,467 (731): 78-81.
[J].Lithos, 20 10,16 (3-4): 209-229.
Mantle plume and episodic crustal growth [J]. Beijing: Science Press, 200 1. Nature,1994,372 (6501): 63-68.
Proofreading: Zhang Tengfei