Thermo-rheological structure of the northern margin of the South China Sea: Structural and geodynamic implications

Thermo-rheological structure of the northern margin of the South China Sea: Structural and geodynamic implications

Journal Pre-proof Thermo-rheological structure of the northern margin of the South China Sea: Structural and geodynamic implications Jie Hu, Yuntao T...

11MB Sizes 0 Downloads 2 Views

Journal Pre-proof Thermo-rheological structure of the northern margin of the South China Sea: Structural and geodynamic implications

Jie Hu, Yuntao Tian, Zulie Long, Di Hu, Yuping Huang, Yibo Wang, Shengbiao Hu PII:

S0040-1951(20)30021-4

DOI:

https://doi.org/10.1016/j.tecto.2020.228338

Reference:

TECTO 228338

To appear in:

Tectonophysics

Received date:

7 June 2019

Revised date:

17 January 2020

Accepted date:

19 January 2020

Please cite this article as: J. Hu, Y. Tian, Z. Long, et al., Thermo-rheological structure of the northern margin of the South China Sea: Structural and geodynamic implications, Tectonophysics(2020), https://doi.org/10.1016/j.tecto.2020.228338

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2020 Published by Elsevier.

Journal Pre-proof

Thermo-rheological structure of the northern margin of the South China Sea: structural and geodynamic implications Jie Hu1,2,3, Yuntao Tian4,5*, Zulie Long6, Di Hu1,2,3, Yuping Huang6, Yibo Wang1,2,3, Shengbiao Hu1,2,3

State Key Laboratory of Lithospheric Evolution, Institute of Geology and Geophysics,

of

1

ro

Chinese Academy of Sciences, Beijing 100029, China

University of Chinese Academy of Sciences, Beijing 100049, China

3

Innovation Academy for Earth Science, Chinese Academy of Sciences, Beijing

re

-p

2

4

lP

100029, China

Guangdong Provincial Key Laboratory of Geodynamics and Geohazards, School of

Southern Marine Science and Engineering Guangdong Laboratory, Zhuhai 519082,

China 6

Jo ur

5

na

Earth Sciences and Engineering, Sun Yat-sen University, Guangzhou 510275, China

Shenzhen Branch of China National Offshore Oil Corporation, Shenzhen 518000,

China

1

Journal Pre-proof

Abstract Rheological properties of continental lithosphere are key controls on the behavior of continental deformation. Using thermal structure, constrained by surface heat flow data and measured thermal properties of rocks, the present study calculates different thermo-rheological structure scenarios for the ocean–continent transition (OCT) at the

of

northern margin of the South China Sea, using two different models: a conventional

ro

model, taking into account frictional sliding and power-law creep, and a model that

-p

additionally includes a high-pressure brittle-fracture mechanism. Two compositions of

re

the lower part of the lithosphere are considered: a soft case with felsic granulite lower

lP

crust and wet peridotite lithospheric mantle, and a hard case with mafic granulite lower crust and dry peridotite lithospheric mantle. The former scenario shows a major

na

rheological change from a “jelly sandwich” to a “Christmas tree” type of rheology from

Jo ur

north to south along the margin. This complex rheological structure explains lateral changes in earthquake distribution and geometries of extensional faults of the OCT at the northern margin of the South China Sea. Further, our analyses indicate that the initial lithospheric rheology profile probably has only one ductile layer in the lower part of upper crust. Such an initial lithospheric rheology model predicts focused extension to form asymmetric margins, which is the case for the SCS. Keywords: Ocean-continent transition; Crustal strength; Thermo-rheology; South China Sea; Pearl River Mouth basin

2

Journal Pre-proof

1. Introduction Lithosphere rheology plays a key role in controlling the behavior of continental deformation (Ranalli and Adams, 2013). Owing to rheological variations, oceanic and continental lithosphere are characterized by different types of deformation. Rheological properties of continental lithosphere are known to vary, with two end-member

of

rheological models having been proposed, known as the “jelly sandwich” and “crème

ro

brulée” models (Burov and Watts, 2006). The former model proposes that two hard,

-p

possibly brittle, load-bearing layers (upper crust and uppermost mantle) are separated

re

by a soft ductile layer (lower crust) (Chen, 2017; Ranalli, 1995; Ranalli and Murphy,

lP

1987). The latter considers the existence of only one hard load-bearing layer, comprising the upper crust (or the whole crust in some cases), overlying a ductile

na

mantle (Chen, 2017; Jackson et al., 2008; Maggi et al., 2000). Oceanic lithosphere is

Jo ur

rigid, and its rheology model is a simple and uniform “Christmas tree” style on account of its thinness (5–7 km) and homogeneous mafic composition (Brace and Kohlstedt, 1980; Duretz and Gerya, 2013; Popov and Sobolev, 2008; Wang et al., 2000). At the ocean–continent transition (OCT), marked lateral variations in crustal geometry and composition have been reported (Allen and Allen, 2013; Chen et al., 2008; Pauselli et al., 2010; Pauselli and Ranalli, 2017;Shi et al., 2000). For example, studies in the northern margin of the South China Sea (SCS) have indicated that the middle–lower crust gradually decrease in thickness from continent to ocean (Shi et al., 2000). Such transitions might result in a significant lateral change in rheological structure at the 3

Journal Pre-proof

OCT. The objective of the present study was to develop a rheological model that explains the earthquake distribution and the geometries of extensional faults at the northern margin of the SCS. We present new thermal conductivity (72 core samples from 11 wells) and heat production measurements (29 samples from 13 wells) from the

of

Pearl River Mouth basin (PRMB) in the SCS. Combining previous heat flow

ro

measurements and ocean bottom seismometer (OBS1993) seismic profiles, we

-p

calculate thermal and rheological structures. Our rheology calculations consider

re

various rheological models and compositional models for the lower crust and

lP

lithosphere mantle. Results indicate a major rheological change from a “jelly sandwich” to a “Christmas tree” type of rheological structure at the transition from the continental

na

margin to the oceanic crust. This complex rheological structure explains lateral changes

Jo ur

in earthquake distribution, which cluster at depths less than 15 km, and the listric geometries of extensional faults, which merge into a detachment at a depth of ~10–15 km in the continental crust.

2. Geological and geophysical background The SCS is the largest marginal sea in the western Pacific region and is surrounded by the western Pacific Plate to the east, the Eurasian Plate to the north, and the Indian– Australian Plate to the west and south (Fig.1a) (Guo et al., 2016). The northern continental margin of the SCS is a rifted continental margin with an OCT extending for ~400 km (Xia et al., 1994; Xia et al., 2018; Zhou et al., 1995). The northern margin of 4

Journal Pre-proof

the SCS shows three sets of faults, trending NE, ENE, and NW, of which the ENE-trending faults control regional subsidence and the basin trend (Shi et al., 2000; Ye et al., 2018b; Zhou et al., 2018). The SCS developed as a result of Cenozoic rifting and seafloor spreading at the southern margin of the South China block (Fan et al., 2017; Taylor and Hayes, 1983;

of

Xia et al., 2018; Ye et al., 2018a). Formation of the SCS can be divided into three major

ro

stages: 1) breakup of the SCS at 34 Ma; 2) mid-ocean ridge jumping at 23.8 Ma; and 3)

-p

spreading cessation at 15.5 Ma (Ye et al., 2018a; Zhao et al., 2018). The pre-Cenozoic

re

metamorphic igneous rock basement is covered by a >11-km-thick sequence of

lP

Paleogene syn-rift lacustrine sediments and Neogene post-rift neritic–abyssal-facies sediments (Huang, 2005; Huang et al., 2018). In addition, Plio-Pleistocene volcanoes in

na

upper-crust fracture zones in the northern OCT suggest that the margin has been the site

Jo ur

of significant tectonic activity (Lüdmann and Wong, 1999). Thirty years of geothermal investigations have provided more than 450 heat flow data points at the northern margin of the SCS (Shi et al., 2017). The distribution of data is non-uniform, with fewer observations in the continental slope and basin area We compiled 177 heat flow data points for the northern margin of the SCS and southern China (17°–24°N, 110°–120°E) and used kriging interpolation to obtain a heat flow contour map (Fig. 1b). From this map, a heat flow profile was constructed along profile OBS1993 for inverting thermal structures (Fig. 2). Previous seismic studies have imaged the complex crustal structures at the 5

Journal Pre-proof

northern margin of the SCS (Pin et al., 2001; Wang and Li, 2009; Xiao et al., 2018; Zhao et al., 2010). The NNW–SSE-striking seismic profile OBS1993 (Fig.1a) released by Pin et al (2001) and reinterpreted by Fan et al (2019) shows marked differences in crustal and lithospheric structures and properties between the northern and southern parts of the OCT (Fan et al., 2019; Pin et al., 2001).

of

Along profile OBS1993, the crust thins gradually from 22 to 8 km with a drastic

ro

change in the lower part of the continental slope from the continental shelf to the

-p

oceanic basin (Fig. 2) (Pin et al., 2001). The average heat flow values increase from ~70

re

mW/m2 in the continent to 100 mW/m2 in the basin (Fig.2) (Jiang et al., 2016; Tang et

lP

al., 2014a; Tang et al., 2014b). The crustal structure along profile OBS1993 is divided into four layers (Fig. 2). The uppermost layer comprises sediments, with the thickest

na

being deposited in grabens in the continental shelf. The upper and lower crust beneath

Jo ur

the sediments are significantly thinned toward the basin interior. A high-velocity layer (HVL) (Fig. 2) has been identified at the base of the crust, as shown by both expanded seismic profile data and OBS data (Wu et al., 2011; Yan et al., 2006). The origin of the HVL is unclear. Wan et al. (2017) suggested that the HVL underlying the OCT represented extension-related decompression melting during the Cenozoic (Wan et al., 2017). In petrology study, the Vp/Vs ratio is particularly sensitive to quartz content. Vp/Vs ratios for different crystalline basement compositions vary from 1.71 in granite (felsic) and 1.78 in granodiorite to 1.84 in gabbro (mafic), but the Vp/Vs of quartz can be as low as 1.48(Christensen and Mooney, 1995; Domenico, 1984; Holbrook et al., 6

Journal Pre-proof

1992). Hence, the Vp/Vs ratios provides some information about lithology. In the study area, Zhao et al. (2010) concluded that measured Vp/Vs ratios of 1.74–1.76 for the upper crust likely reflect felsic (granitic) rock, Vp/Vs ratios of 1.81–1.87 in the lower crust suggest a mafic composition, whereas Vp/Vs ratios of HVL ranges from 1.76 to 1.94 with an average value of 1.85, indicating a mafic composition (gabbro or

of

mafic gneisses) (Zhao et al., 2010). Further, the heat flow trend is independent of the

ro

thickness of the HVL (Fig. 2), indicating that this layer generates only small amounts of

thermo-rheological structure.

na

3.1. Thermal conductivity

lP

3. Methods and measurements

re

-p

heat. Therefore, in the present study the HVL is assumed to be gabbro when calculating

We collected 72 core samples from 11 wells in the PRMB for thermal conductivity

Jo ur

measurements. The optical scanning method suggested by the International Society for Rock Mechanics (ISRM) was used (He et al., 2008; Popov et al., 2016). A TCS (Thermal Conductivity Scanner, Germany) was used to measure variations in thermal conductivity among inhomogeneous samples of variable size (1–70 cm) and shape. Within a conductivity range of 0.1–70 W·m−1·K−1, this equipment has high precision (1.5%) and accuracy (1.5%, 0.95 significance level) (Popov et al., 2012). The factor G = standard deviation/mean and InhomoFactor = (maximum-minimum)/mean are used to describe the uncertainty. The thermal conductivity of rocks can be influences by other factors such as pressure, temperature, porosity, and water saturation (Pribnow et 7

Journal Pre-proof

al., 1996). In this study, we focused on the temperature effect, which was the main factor of thermal conductivity in high temperature basin, and ignored the potential water saturation correction and pressure effect (because of lacking rock porosity and pressure data). The temperature was tested by MDT (Modular Formation Dynamics Tester Tool) from Shenzhen Branch of China National Offshore Oil Corporation. We

of

applied the empirical correction suggested by Sekiguchi et al (1984). The corrected

ro

thermal conductivity, 𝐾(𝑊 ∙ 𝑚−1 ∙ 𝐾 −1 ), at absolute temperature, T(K), is given by T 𝑇

1

1

𝑚 (K 0 − 𝐾𝑚 ) ( − ) + 𝐾𝑚 K = 𝑇 0−𝑇 𝑇 𝑇 0

𝑚

(1)

-p

𝑚

re

Where K m = 1.05 𝑊 ∙ 𝑚−1 ∙ 𝐾 −1 , a calibration coefficient

lP

𝐾0 = thermal conductivity at laboratory temperature, 𝑇0 𝑇0 = temperature (K) at which 𝐾0 was measured (293K in this study);

na

𝑇𝑚 = 1473𝐾, a calibration coefficient

Jo ur

The collected core samples cover a wide range of rock types (mudstone, siltstone, sandstone, gritstone, conglomerate, diorite, and amphibolite). The corrected conductivity values of sedimentary rocks range from 1.1 to 2.8W·m−1·K−1 with a mean of 2.1 ± 0.4W·m−1·K−1 (Table 1) and show a normal distribution (Fig. 3). 3.2. Radiogenic heat production To investigate heat production, we measured 29 samples from 13 wells in the PRMB covering the majority of the strata (Table 2). The contents of the radioactive heat-generating elements of the rocks were tested at the China Nuclear Industry Geological Analysis and Testing Center, Beijing, China. U and Th contents were 8

Journal Pre-proof

measured using a solution inductively coupled plasma–mass spectrometry (ICP–MS) method, and K content analyses were determined using atomic absorption spectroscopy (Wang et al., 2019). The density was tested by BSD-TD1 automatic density analyzer from Beishide Instrument (Beijing), which uses inert gas (He) replacement method. Heat production was calculated as follows:

of

A = 0.1325𝜌(0.718𝑐𝑈 + 0.193𝑐𝑇ℎ + 0.262𝑐𝐾 ) (2)

ro

where A is radioactive heat generation (in μW/m3 ); cU and cTh are the U and Th

re

g/cm3 (Rybach and Buntebarth, 1982).

-p

contents (in ppm), respectively; ck is the K content (in wt.%); and ρ is density in

Calculated heat production values range from 0.61 to 3.56 μW·m−3, with a mean

lP

value of 1.65 ± 0.91 (standard deviation) μW·m−3.

na

4. Modeling methods and results

Jo ur

4.1. Temperature structure 4.1.1. Thermal parameters

Four types of data were used to calculate the temperature profile: (1) surface heat flow, (2) seabed temperature, (3) heat production, and (4) thermal conductivity. Surface heat flow was obtained from the heat flow map (Fig. 1b). Seabed temperature is negatively related to seawater depth. At seawater depths of >2.8 km, the seabed temperature is 2.2 °C, at shallower depths, the temperature is linearly related to the logarithm of depth (Feng et al., 1996; Wang et al., 2005). The depth– temperature model is 9

Journal Pre-proof

{

𝑙𝑛𝑍 = −1.3361𝑙𝑛𝑇0 + 2.0339 (for 𝑍 < 2.8) , 𝑇0 = 2.2 (for 𝑍 ≥ 2.8)

(3)

where Z is seawater depth (in km) and T0 is seabed temperature in °C. Thermal conductivity and heat production of the sedimentary layer used the average values of measurements, presented above. For deeper layers, values obtained from previous studies were used (Table 3), and the upper crust, lower crust, HVL and

of

mantle are treated as granite, granulite, gabbro and peridotite, respectively.

ro

4.1.2. Method

-p

As there is too much uncertainty to calculate a reliable non-steady-state

re

temperature distribution for the northern margin of the SCS, we computed a

lP

steady-state thermal structure for seismic profile OBS1993, with an assumption that heat transfer by convection and advection is negligible. Under the assumption of

Jo ur

equation is

na

two-dimensional (2-D) purely conductive steady-state heat transfer, the governing



𝜕𝑥

𝜕𝑇



𝜕𝑇

(K 𝜕𝑥 ) + 𝜕𝑧 (K 𝜕𝑧 ) + 𝐴 = 0 , (4)

where T is temperature, K is thermal conductivity, A is heat production, and x and z are horizontal and vertical coordinates, respectively. In the calculation of Equation 4, the domain of our model is 0 ≤ x ≤ 400 km and 0 ≤ z ≤ 50 km. The top boundary condition is an irregular boundary (seabed) set as the seabed temperature (Equation 5a), thermal insulating conditions are set along the sides of model (Equation 5b), and a constant heat flow is set as the bottom boundary (Equation 5c), as follows: 10

Journal Pre-proof 𝑇0 = 𝑇(𝑠𝑒𝑎𝑏𝑒𝑑) (5𝑎) 𝜕𝑇 𝜕𝑇 𝑞𝑥 = −𝐾 = = 0, when x = 0 and x = 400 km (5𝑏) 𝜕𝑥 𝜕𝑥 𝜕𝑇 −𝐾 = 𝑄𝑚 (5𝑐) { 𝜕𝑥 where T0 is the temperature at the seabed in the calculation grid, T(seabed) is the seabed temperature calculated from Equation 3, and qx and Qm are the lateral and bottom heat flow, respectively.

of

The solution of temperature field requires an inverse problem to be solved first,

ro

(i.e., to find the bottom heat flow Qm), which is solved by optimizing Qm to fit the

-p

surface heat flow observations (Fig. 2). The misfit function is the sum of the L1 norm

re

of computed (Qcomputed) and observed (Qobserved) seabed heat flow values:

lP

Misfit = ∑ |𝑄𝑜𝑏𝑠𝑒𝑟𝑣𝑒𝑑 − 𝑄𝑐𝑜𝑚𝑝𝑢𝑡𝑒𝑑 | (6)

na

2-D calculation of temperature distribution was performed using a MATLAB finite-difference method following the approach of Gerya (2009), with a calculation

Jo ur

grid of 201 × 101 in size (Gerya, 2009). The optimization toolbox in MATLAB was used, along with the constrained nonlinear minimization solver (fmincon) function and sequential quadratic programming (SQP) algorithm. In the optimization, the upper and lower bounds are the observed surface heat flow and 1 mW/m2, respectively. The calculation grid of 201 × 101 means that the bottom boundary has 201 values. We sampled 21 nodes equidistantly and established bottom heat flow (BHF) as the optimization variable, with the bottom boundary being obtained by spline interpolation. BHF0 is the initial value and can be any value within the bounds. In the present study, the median of the bounds was used. The workflow is shown in Fig. 4, 11

Journal Pre-proof

with the dashed boxes indicating improvements made to the original workflow after initial trials. After running the optimization, two problems were identified:1) huge variations in BHF over a spatial scale of several tens of kilometers, and 2) variations in computed surface heat flow over a spatial scale of a few kilometers (Fig. 5a). For the first problem,

of

the misfit was 421.5, and the mean error at each point was only 2.1 . Although this

ro

might represent a good mathematical solution, it is not a reasonable geological solution

-p

because of the large variation over a fairly short distance. To address this problem, we

re

added a function to discard BHF outliers and to smooth BHF (Fig. 4 workflow, upper

lP

dashed box), which made variations in BHF less extreme. For the second problem, the variations were considered to be caused by the internal boundary and lateral conduction.

na

For example, the top and right side of a particular sediment node was seawater whose

Jo ur

temperature was fixed or showed a small variation. The lateral conduction of heat caused the heat flow of the top of the node to be low and the heat flow of the right of the node to be larger, thereby increasing the misfit. A denser grid can decrease the amplitude but raise the number of variations and significantly increase the duration of the calculation up to several weeks for an optimization process. Generally, heat flow is calculated as the product of geothermal gradient and thermal conductivity, such as in Equation 5b (1D steady-state thermal conduction equation). However, heat flow by definition is terrestrial heat flow density, which measures the flow of heat from the hot interior of Earth to its cooler exterior per unit area and per unit time (Beardsmore, 2001; 12

Journal Pre-proof

Haenel et al., 2012). For this purpose, we introduced a moving average of heat flow with a window of five nodes (values) (Fig. 4 workflow, lower dashed box), which not only makes the calculation more realistic (an area-based rather than a point-based definition of heat flow) but also reduces the size of the variations. In addition, in the later stages of iteration, while the algorithm was computing the minimum value of

of

misfit, the misfit reduction was very small and the BHF became increasingly extreme

ro

and decreasingly realistic, so we loosened the stopping criteria. After several trials,

-p

we increased the lower bound to 30 mW/m2, which was more suitable given the

re

geological context.

lP

After the above trials of and improvements to the optimization process, the misfit yielded was 428.8 and average error was 2.13, which were only a little bit larger than

na

in the original optimization and acceptable because the error was less than that

Jo ur

associated with observed heat flow values (~5%). Although the error between computed and observed heat flow might be caused by thermal fluids or volcanism (thermal advection), the computed heat flows are in agreement with the large-scale geology of the studied region and are therefore sufficiently realistic. 4.1.3. Temperature distribution Surface heat flow can be divided into two components: crustal heat flow, which is due mainly to radioactive heat production; and mantle heat flow, which is due mainly to cooling of the mantle and core. For the studied region, the mantle heat flow increases from ~60% of the total heat flow in the north to ~90% in the south (Fig. 6a), 13

Journal Pre-proof

reflecting that the northern margin of the SCS is under a tensile stress field (crustal thinning), with the majority of heat flow deriving from the mantle. We then used the computed BHF to solve Equation 4, for which the resulting temperature profile along OBS1993 is shown in Fig. 7. Along OBS1993, continental crust is hotter than oceanic crust, and the temperature of the Moho shows marked lateral

of

variation along the profile from oceanic crust to continent crust, being higher (~800 K)

ro

in the north (continental) than in the south (oceanic) (~480 K) because of thick crust in

-p

the north. Conversely, the geothermal gradient increases from continental to oceanic

re

crust, which is consistent with oil company well-logging data (Tang et al., 2014b). The

the OCT.

na

4.2. Rheological structure

lP

marked spatial variation in temperature is a key control on the rheological structure of

Jo ur

Rheology describes the response of materials to an imposed stress system (Pauselli et al., 2010; Ranalli, 1995; Watts and Burov, 2003). The strength of the lithosphere is controlled by the temperature, crustal composition, and crustal thickness (Kusznir and Park, 1987; Panza and Raykova, 2010). For any depth and temperature range, the mechanism of deformation that requires the smallest critical stress predominates (Gong et al., 2018; Ranalli, 1995). The rheological profile does not focus on the absolute value of but the relative strength and deformation mechanism of the lithospheric strength. 4.2.1. Method 14

Journal Pre-proof

As well as the constrained temperature distribution, density was also required to ascertain the rheological structure of the lithosphere. We measured the density of 29 samples (Table 2) and used the mean of these data as the sediment density. Hao et al. (2008) used Hearn’s method to construct a density model of the crust for the northeastern SCS. Those authors generated average densities for the upper crust, lower

of

crust, and HVL of 2.40–2.65, 2.75–2.85, and 2.92 g/cm3, respectively (Table 3). In the

ro

present study, we used the medians of those value ranges, and the adopted lithospheric

-p

mantle value of 3.32 g/cm3 following Clark and Ringwood (1964).

re

In the widely used conventional model (CM), the rheological structure of the

lP

lithosphere is determined by comparing the critical stress 𝜎𝐹 required for brittle frictional sliding (Coulomb’s/Byerlee’s law) with the critical stress 𝜎𝐷 required for

na

ductile creep (power-law creep, sometimes including Dorn creep at high stresses) at a

Jo ur

constant steady-state strain rate (Byerlee, 1978, Ranalli, 1995). For brittle frictional sliding, we have σ1 − 𝜎3 ≡ 𝜎𝐹 = 𝑓𝜌𝑔𝑧(1 − 𝜆) . (7a)

Considering the inhomogeneity of lithospheric density, Equation 7a can be rewritten as 𝑧

σ1 − 𝜎3 ≡ 𝜎𝐹 = 𝑓𝑔(1 − 𝜆) ∫ 𝜌(𝑧)𝑑𝑧 , (7b) 0

where σ1 and 𝜎3 are the maximum and minimum principal stresses, respectively; σF is the critical stress difference in the frictional case; f is a parameter related to the standard (Anderson) stress state (f = 3.0 and 0.75 for compressional and extensional 15

Journal Pre-proof

regimes, respectively); ρ is the material density; g is acceleration due to gravity; z is depth; and λ is the pore fluid factor. For dislocation creep, we have 1

σ1 − 𝜎3 ≡ 𝜎𝐷 =

𝜖 𝑛 𝐸 (𝐴̇ ) exp (𝑛𝑅𝑇)

, (8)

where σF is the critical stress difference in the dislocation creep case; 𝜖̇ is strain rate;

of

A, n, and E are creep parameters; T is absolute temperature (the pressure dependence is

ro

negligible at lithospheric depths and is therefore disregarded); and R is the gas constant.

-p

With increasing pressure, the critical stress difference in the brittle field becomes

re

nonlinearly dependent on pressure, temperature, and strain rate (Zang et al., 2007).

lP

Therefore, Pauselli et al. (2010) suggested a parallel set of models (ZMs) that take into account high-pressure brittle failure:

𝑃 𝑚

𝑇 𝛽

𝜖̇

na

σ1 − 𝜎3 ≡ 𝜎𝑃 = 𝐵0 [1 + 𝐾 (𝐵 ) ][1 + 𝛼 (𝑙𝑔 𝑇 ) ][1 + 𝛾𝑙𝑔 𝜖 ] (9) 0

0

𝑜

Jo ur

where B0 , K, m, α, β, γ are empirical parameters; P, T and 𝜖̇ are confining pressure (MPa), temperature (K) and strain rate ( 𝑠 −1 ), respectively; T0 = 298 K (room temperature); and 𝜖0 = 10−5 𝑠 −1 is a normalizing parameter. However, this equation is experimentally validated only for P ≤ 0.8 GPa and T ≤ 900 and 1100 °C, respectively, for crustal and upper mantle rocks. Thus, for the CM the predominant mechanism (and the relevant critical stress) is given by 𝜎𝑠 = 𝑀𝑖𝑛[𝜎𝐹 , 𝜎𝐷 ] (10a) For the high-pressure stress model (ZM; Pauselli et al., 2010), the value is given 16

Journal Pre-proof

by 𝜎𝑠 = 𝑀𝑖𝑛[𝜎𝐹 , 𝜎𝐷 , 𝜎𝑃 ] (10b) As the study area is in an extensional state, f takes a value of 0.75. It is assumed that the pore fluid pressure is hydrostatic and is taken as a typical value of λ = 0.37 following the usage by Wang et al (2011) (Wang et al., 2001).

of

Strain rate is also needed to obtain a rheological profile. Michela et al (2001) used a

ro

GPS network to study crustal motion and block behavior in SE Asia and chose baseline

-p

variations (<4 mm/yr) and small interstation strains (<6.34 × 10−16 s−1) to define a “rigid”

re

Sundaland platelet (Michel et al., 2001). We consider that the present study area and the

lP

Sundaland platelet are subjected to the same stress field, and the distance between them is small, so their strain rates might be similar (within the same magnitude). Further, a

na

strain rate of one order greater of less does not result in significant influence in strength

Jo ur

distribution. In this work, we used a strain rate of 10−16 s−1. To better investigate thermal state and rheological structure at the northern margin of the SCS, the variable composition of the lower part of the lithosphere was taken into account as follows: (1) a soft case with a felsic granulite lower crust and wet peridotite lithospheric mantle; and (2) a hard case with a mafic granulite lower crust and dry peridotite lithospheric mantle (Chen et al., 2014; Shi et al., 2003). Under a power-law regime (Equation 8), the upper crust (including sediments) and HVL are assumed to be controlled by wet quartz and gabbro whose creep parameters follow those of plagioclase An75, respectively (Gerya, 2009; Ranalli, 1995). The lower crust and the 17

Journal Pre-proof

lithospheric mantle are taken to have the rheologies of felsic granulite and wet peridotite, respectively, in the soft case, and of mafic granulite and dry peridotite in the hard case (Wang et al., 2012; Wilks et al.,1990). Under high-pressure failure (Equation 9), quantitative models are available for four rock types (granite, gabbro, peridotite and basalt). For these reasons , both lower crust and HVL are regarded as gabbro, whereas

of

the upper crust (including sediments) and lithosphere are assigned the parameters for

ro

granite and peridotite, respectively (Pauselli et al., 2010). All parameter values used in

-p

the calculations are given in Table 4.

re

4.2.2. Rheological profile

lP

We constructed four rheological models considering two compositional combinations and two deformation models along profile OBS1993 based on seismicity,

na

heat flow, and measured thermal properties of rocks. Taking the petrology study into

Jo ur

account (Section 4.2.1), the study area is more likely to be the hard case (Table 4), which is presented below. The soft cases are described in the supplementary material. All these rheological models are compared with earthquake distribution and fault geometry in the following sections (Fig. 8). In the hard CM, there is an obvious “jelly sandwich” model in the northern part of the study area and a “Christmas tree” model in the southern part. However, the “jelly sandwich” is caused mainly by the significant compositional difference between the HVL and continental lithosphere. The brittle field has a maximum strength σs of ~500 MPa at a depth of 39 km. There are also two ductile layers (the bottom part 18

Journal Pre-proof

of the upper crust and the HVL), and the entire lower crust is brittle. In the hard ZM, the maximum rheological strength of the lithospheric mantle is clearly reduced relative to the CM. High-pressure brittle failure occurs only in the lithospheric mantle, and the brittle regime is slightly thicker and deeper in the hard ZM compared with the hard CM.

of

There are significant lateral differences in rheological structure in both hard

ro

models for a given thermal state. The different types of deformation change the

-p

thickness of the brittle layer and its strength but have little influence on the general

re

rheological structure. The rheological profile changes from “jelly sandwich” type to

lP

“Christmas tree” type without a clear boundary, and the ductile layer in the upper crust and HVL successively disappear from north to south in the OCT. When high-pressure

na

stress is taken into account, the brittle regime in the rheological structure can be divided

Jo ur

into two regions: one dominated by Byerlee’s law (in crust), and the other by high-pressure brittle failure (in mantle), which greatly reduces rheological strength and deepens the brittle deformation region. 5. Discussion 5.1. Comparison with earthquake distribution Rheological properties of the crust can be assessed by considering the seismogenic layer, where earthquakes occur, as revealed by seismicity (Panza and Raykova, 2010; Wu and Zhang, 2012). The maximum earthquake depth might not be a direct proxy for the depth of the brittle–ductile transition (BDT), but is expected to be related to 19

Journal Pre-proof

depth-dependent brittle strength (Burov, 2011). No obvious relationship between the seismogenic layer and the long-term strength of the lithosphere has been found (Chen et al., 2012; Handy and Brun, 2004; Watts and Burov, 2003; Watts et al., 2013), but seismicity might reflect the short-term rheology of the lithosphere, which can provide insights into the rheological behavior of the lithosphere at different time scales (Chen et

of

al., 2014). Accordingly, we combined seismic events from China Earthquake

ro

Datacenter (CED, data since 2008 onwards) and the United States Geological Survey

-p

(USGS, data before 2008) and analyzed the depth distribution of earthquakes. The

re

earthquake data points are plotted in Figure 9a.

lP

To compare the crustal rheology, calculated above, and the earthquake distribution in the study area, we plotted the earthquake foci within the PRMB (defined as an area of

na

3 degrees to the west and 2.5 degrees to the east, that is the area within the orange

Jo ur

rectangle in Fig.9a) onto the OBS1993 rheology line. Earthquakes outside of PRMB may have different origins, for examples, the cluster of events to the east (at a longitude of ~120oE), are related to the subduction beneath the Luzon arc (Kao et al., 1998; Galgana et al., 2007). Further, earthquakes in the north margin of the PRMB may be controlled by a littoral fault zone (Cao et al., 2018). Comparison between the earthquakes (42 events) within the PRMB and the hard-case rheology profiles (CM and ZM, shown in Fig. 8), it is found that most events fall into brittle layers in the upper and middle crust at depth <15 km and between 15-25 km (Fig. 8a,b); whereas at depths 10-15 km and 25-30 km, less earthquakes have 20

Journal Pre-proof

occurred (Fig. 8a,b). Therefore, the earthquake data confirm these modeling results, showing two ductile layers in upper crust and HVL respectively, and three brittle layers in the upper crust, lower crust and mantle, respectively (Fig. 8). Further, the soft-case rheology models indicate a thicker weaker layer in the lower crustal at depth between 18-25 km in the northern part of the OBS1993 profile (Fig. S1).

of

The lower crustal brittle region is as thin as ~2 km, which seems inconsistent with the

ro

cluster of the earthquakes in lower crust (15-20 km) (Fig.S1). We therefore suggest

-p

this rheology model is not suitable for the study area.

re

It should be noted that the earthquake depths do not strictly correspond to the

lP

locations of the brittle layers in Fig. 8. This is probably because of lateral differences in crustal structure and rheology between the earthquake locations and the studied profile

na

OBS1993. Further, these real-time earthquake depth reports from CED and USGS may

Jo ur

not be accurate and needs to be relocated (Fang et al., 2013; Huang et al., 2008; Zhu et al., 2005). But, the big-picture of these earthquakes are consistent with our rheology calculations, as discussed above. However, the earthquake data doesn’t help distinguish the ZM and CM models. Both predict two ductile layers at similar depths. 5.2. Comparison with extensional fault geometry Rocks at Earth’s shallow subsurface generally exhibit brittle behavior, but with increasing temperature and pressure with depth, brittle behavior transitions to ductile (Panza and Raykova, 2010; Park, 2013; Zang et al., 2007). Fault geometry is also found dependent on crustal rheology. In the study area, along the seismic profile DSRP2002, 21

Journal Pre-proof

the interpreted faults (solid red lines) end at 10–14 km depth (Fig. 10) in the northern part of the profile. In the southern part, the interpreted faults also end in the upper crust (Huang, 2005). At a nearby site, Zhou et al. (2018) found a continent-ward-dipping, low-angle fault plane reflection at depth by 3D seismic data. It can be traced to ~7.5s (TWT) or ~16 km (calculated by time-depth conversion equation therein), which is

of

consistent with a deep-seated, undulating reflector band and may represent ductile

ro

shear fabrics in the lower crust (Zhou, 2018). The depth (~10-16 km) where fault merge

-p

coincides with the rheologically weak layers in the lower part of the upper crust, as

re

shown by our preferred rheological model (hard cases). Deformation in this layer

lP

should be ductile in nature.

Zuber et al. (1986) and Clift et al. (2002) predicted a ductile layer in the lowermost

na

crust in the study area based on their rheological investigations. According to our

Jo ur

rheological study, this ductile layer in the lowermost crust is likely to be the HVL, mapped by OBS studies.

The presence of two ductile layers in the study area may prevent the fault from cutting through the Moho in areas with limited extension, such as the study area of this work. Earlier studies suggested deep-seated faults in the study area. For example, Huang et al (2005; 2018) predicted that Main Baiyun Sag (MBS) is limited in north and south by two sets of deep-seated boundary faults, which is likely to have penetrated Moho, and that the southern boundary of the South Baiyun Sag (SBYS) is defined as a set of deep-seated faults, which were speculated as a lithosphere fault. In our 22

Journal Pre-proof

rheological study, the faults in MBYS are unlikely to penetrate Moho because of the presence of two ductile layers in upper crust and HVL (Huang, 2005; Huang et al., 2018). As the entire crust is brittle in southern part, the faults in the southern margin of the SBYS may cross through the Moho. Further, in seismic reflection profiles of the OCT in the northeastern SCS, no detachment faults cutting through the entire crust

of

were observed (Clift and Lin, 2001; Zuber et al., 1986).

ro

5.3 Implication for the rifting of northern margin of SCS

-p

As reviewed in Peron-Pinvidic and Manatschal (2019), lithospheric structure is

re

one of the most important factors controlling the rheological properties and evolution

lP

of magma-poor rifted continental margins. For continental margins with a weak lower

na

crust, crustal rifting lead to the decoupling of upper and lower lithosphere over a wide area (Huismans and Beaumont, 2011). Such a lower crust may flow laterally, which

Jo ur

can explain the subsidence in the OCT (Clift et al., 2002). If the weak zone is located in the middle crust, upper crustal extension would be compensated by mid-crustal shear zone (Franke et al., 2014; Li et al., 2019). Our results suggest that the lower crustal rheology properties of the northern SCS is controlled by its composition (Figs. 8 and S1 in the online supplementary material). As shown by previous seismic studies, the lower crust and the underlying HVL are characterized by relatively high Vp/Vs ratios (1.76 to 1.94), indicating a mafic composition (Zhao et al., 2010). Such a model suggests a high rheological strength in the lower crust and a weak zone in the lower part of the upper crust (Fig. 8). Our 23

Journal Pre-proof

rheological model has some important implications for the formational dynamics of the marginal sea, as discussed below. Before rifting, the study area was likely characterized by a thicker crust than the present-day. Using the less-rifted northern part of the OBS1993 profile as an analog, the initial lithospheric rheology structure can be estimated as the following two

of

scenarios with or without the HVL. If the HVL was formed before the SCS rifting, the

ro

upper crust consists of a brittle layer in the upper part and a ductile layer in the lower

-p

part, whereas the lower crust is mostly brittle underlain by a second ductile layer in

re

the HVL (Fig. 8b). Lateral ductile flow of HVL may mechanically decouple the lower

lP

crust and accommodate the extension deformation during rifting (Li et al., 2019). Further, during lithospheric stretching and rifting, deformation of the ductile HVL

na

should be more uniform than brittle layers, as shown by numerical models (Brune et

Jo ur

al., 2017; Li et al., 2019; Tetreault and Buiter, 2018). However, this speculation is inconsistent with the current lithospheric geometry of the study area, which includes a pinch-out HVL and a relatively homogeneous lower crust (Fig. 2a). Therefore, we favor a second scenario that the HVL was formed during or after the rifting. If so, using the northern part of the profile as an analog, the initial lithospheric rheology profile has only one ductile layer in the lower part of upper crust, without the one that is currently located within the HVL. Based on the modeling of Tetreault and Buiter (2018), at a high extension rate (such as ~5.0 cm/a for the SCS, Hayes and Nissen, 2005) and a low ratio of ductile thickness in the crust, deformation at layers is likely 24

Journal Pre-proof

to be more coupled, leading to focused strain to form narrow initial rifts and asymmetric margins (Tetreault and Buiter, 2018). Such a prediction is consistent with the crustal architecture in the SW part of the SCS, with an asymmetric OCT (~60 km and no more than 30 km wide on the northern and southern sides of the basin) (Pichot et al., 2014).

of

Conclusions

ro

On the basis of surface heat flow, 72 new measured thermal conductivity and 29

-p

heat production measurements, the temperature distribution and rheological structure

re

of the northern part of the SCS have been studied.

lP

1) Using a new work flow for constraining the bottom heat flow, we derived a 2D

na

steady-state temperature distribution along the 400-km-long OBS1993 profile. Based on the temperature profile, four rheological models considering two

Jo ur

compositional combinations and two deformation models are acquired. 2) The model with mafic granulite lower crust and dry peridotite lithospheric mantle predicts two ductile layers at depths of ~10-15 km and 20-25 km, where lower-seismicity occurs and faults merge. Based on the rheological study, we infer that the faults in the northern SCS are unlikely to penetrate through the Moho. 3) Byerlee’s law, which does not consider temperature, would result in unrealistically high critical stresses in the lithospheric mantle. The ZM models, used in this work, significantly reduce the strength and deepen the brittle region. However, there is no strong evidence from earthquake distribution and faults geometry to distinguish 25

Journal Pre-proof

these rheology models. 4) The HVL is likely to form during or after the rifting. If so, the initial lithospheric rheology profile probably has only one ductile layer in the lower part of upper crust, without the one that is currently located within the HVL. Such an initial lithospheric rheology model would result in an extensional basin with asymmetric

lP

Acknowledgments

re

-p

ro

of

margins, which is the case for the SCS.

na

This work was financially supported by the National Major Science and Technology

Jo ur

Projects program in the 13th Five-Year Plan under the title “The Evolution of Paleotemperature and Paleopressure and the Process of Generation, Ejection, and Accumulation of Petroleum” (2016ZX05026-003-006), National Natural Science Foundation of China (NSFC) grants (U1701641) and the Guangdong Province Introduced Innovative R&D Team (2016ZT06N331). We are grateful to Prof. Lijuan He, Prof. Xiaobin Shi, and Assoc. Prof. Shaowen Liu for their professional and patient guidance. Figs. 1 and 9 were generated with the Cartopy package developed by the Met Office.

26

Journal Pre-proof

References Allen, P.A., Allen, J.R., 2013. Basin Analysis: Principles and Application to Petroleum Play Assessment, 3rd Edition. Brace, W.F., Kohlstedt, D.L., 1980. Limits on lithospheric stress imposed

of

by laboratory experiments. Journal of Geophysical Research 85, 6248-6252.

ro

Burov, E., Watts, A., 2006. The long-term strength of continental

-p

lithosphere:" jelly sandwich" or" crème brûlée"? GSA today 16, 4.

re

Burov, E.B., 2011. Rheology and strength of the lithosphere. Marine and Petroleum Geology 28, 1402-1443.

lP

Brune, S., Heine, C., Clift, P.D., Pérez-Gussinyé, M., 2017. Rifted margin

na

architecture and crustal rheology: Reviewing Iberia-Newfoundland, Central

257-281.

Jo ur

South Atlantic, and South China Sea. Marine and Petroleum Geology 79,

Byerlee, J., 1978. Friction of rocks, Rock friction and earthquake prediction. Springer, pp. 615-626. Cao, J., Xia, S., Sun, J., Fang, Z., Wan, K., Xu, H., 2018. Offshore Fault Geometrics in the Pearl River Estuary, Southeastern China: Evidence from Seismic Reflection Data. Journal of Ocean University of China 17, 799-810. Chen, L., 2017. Layering of subcontinental lithospheric mantle. Science Bulletin 62, 1030-1034. 27

Journal Pre-proof Chen, L., Berntsson, F., Zhang, Z., Wang, P., Wu, J., Xu, T., 2014. Seismically constrained thermo-rheological structure of the eastern Tibetan margin: Implication for lithospheric delamination. Tectonophysics 627, 122-134. Chen, L., Tao, W., Zhao, L., Zheng, T., 2008. Distinct lateral variation of

of

lithospheric thickness in the Northeastern North China Craton. Earth and

ro

Planetary Science Letters 267, 56-68.

-p

Chen, W.-P., Hung, S.-H., Tseng, T.-L., Brudzinski, M., Yang, Z., Nowack,

re

R.L., 2012. Rheology of the continental lithosphere: Progress and new perspectives. Gondwana Research 21, 4-18.

lP

Christensen, N.I., Mooney, W.D., 1995. Seismic velocity structure and

na

composition of the continental crust: A global view. Journal of Geophysical

Jo ur

Research 100, 9761-9788.

Clark, S.P., Ringwood, A., 1964. Density distribution and constitution of the mantle. Reviews of Geophysics 2, 35-88. Clift, P., Lin, J., 2001. Preferential mantle lithospheric extension under the South China margin. Marine and Petroleum Geology 18, 929-945. Clift, P., Lin, J., Barckhausen, U., 2002. Evidence of low flexural rigidity and low viscosity lower continental crust during continental break-up in the South China Sea. Marine and Petroleum Geology 19, 951-970. Domenico, S.N., 1984. Rock lithology and porosity determination from 28

Journal Pre-proof shear and compressional wave velocity. Geophysics 49, 1188-1195. Duretz, T., Gerya, T.V., 2013. Slab detachment during continental collision: Influence of crustal rheology and interaction with lithospheric delamination. Tectonophysics 602, 124-140. Fan, C., Xia, S., Cao, J., Zhao, F., Sun, J., Wan, K., Xu, H., 2019. Lateral

of

crustal variation and post-rift magmatism in the northeastern South China Sea

ro

determined by wide-angle seismic data. Marine Geology 410, 70-87.

-p

Fan, C., Xia, S., Zhao, F., Sun, J., Cao, J., Xu, H., Wan, K., 2017. New

re

insights into the magmatism in the northern margin of the South China Sea: Spatial features and volume of intraplate seamounts. Geochemistry,

lP

Geophysics, Geosystems 18, 2216-2239.

na

Fang, L., Wu, J., Wang, W., Lü, Z., Wang, C., Yang, T., Cai, Y., 2013.

Jo ur

Relocation of the mainshock and aftershock sequences of M S 7.0 Sichuan Lushan earthquake. Chinese Science Bulletin 58, 3451-3459 (in Chinese with English abstract).

Feng, Z., Feng, W., Xue, W., 1996. Evolution of Geological Hazards and Seafloor Engineering Geological Condition of Northern South China Sea. Hehai University Press (in Chinese). Franke, D., Savva, D., Pubellier, M., Steuer, S., Mouly, B., Auxietre, J.-L., Meresse, F., Chamot-Rooke, N., 2014. The final rifting evolution in the South China Sea. Marine and Petroleum Geology 58, 704-720. 29

Journal Pre-proof G. R. Beardsmore, J.P.C., 2001. Crustal Heat Flow: a guide to measurement and modeling. Cambridge University Press. Galgana, G., Hamburger, M., McCaffrey, R., Corpuz, E., Chen, Q., 2007. Analysis of crustal deformation in Luzon, Philippines using geodetic observations and earthquake focal mechanisms. Tectonophysics 432, 63-87.

of

Gerya, T., 2009. Introduction to numerical geodynamic modelling.

ro

Cambridge University Press.

structure,

mechanical

anisotropy

and

lithospheric

re

thermo-rheological

-p

Gong, W., Jiang, X., Zhou, H., Xing, J., Li, C., Yang, K., 2018. Varied

Sciences 163, 108-130.

lP

deformation of the southeastern Tibetan Plateau. Journal of Asian Earth

na

Guo, L.L., Li, S.Z., Zhao, S.J., Zhang, G.X., Suo, Y.H., Liu, H., Hui, G.G.,

Jo ur

Fu, S.Y., Dai, L.M., Cao, H.H., Li, X.Y., Wang, P.C., Wang, Q., 2016. Final breakup of continental block and opening of oceanic lithosphere: insights from deep crustal structure and tectonic evolution of the ocean-continent transition zone in the northern South China Sea. Geological Journal 51, 318-330. Haenel, R., Stegena, L., Rybach, L., 2012. Handbook of terrestrial heat-flow density determination: with guidelines and recommendations of the International Heat Flow Commission. Springer Science & Business Media. Handy, M.R., Brun, J.P., 2004. Seismicity, structure and strength of the continental lithosphere. Earth and Planetary Science Letters 223, 427-441. 30

Journal Pre-proof Hayes, D.E., Nissen, S.S., 2005. The South China sea margins: Implications for rifting contrasts. Earth and Planetary Science Letters 237, 601-616. He, L., Hu, S., Huang, S., Yang, W., Wang, J., Yuan, Y., Yang, S., 2008. Heat flow study at the Chinese Continental Scientific Drilling site: Borehole

of

temperature, thermal conductivity, and radiogenic heat production. Journal of

ro

Geophysical Research 113, B02404.

-p

He, L., Wang, K., Xiong, L., Wang, J., 2001. Heat flow and thermal history

re

of the South China Sea. Physics of the Earth and Planetary Interiors 126, 211-220.

lP

Holbrook, S.W., Mooney, W.D., Christensen, N.I., 1992. The seismic

Jo ur

1-43.

na

velocity structure of the deep continental crust. Continental Lower Crust 23,

Huang, C., 2005. Deep crustal structure of Baiyun Sag, northern South China Sea revealed from deep seismic reflection profile. Chinese Science Bulletin 50, 1131-1138.

Huang, K., Zhong, G., He, M., Liu, L., Wu, Z., Liu, X., 2018. Growth and linkage of a complex oblique-slip fault zone in the Pearl River Mouth Basin, northern South China Sea. Journal of Structural Geology 117, 27-43. Huang, Y., Wu, J., Zhang, T., Zhang, D., 2008. Relocation of the M8. 0 Wenchuan earthquake and its aftershock sequence. Science in China Series 31

Journal Pre-proof D: Earth Sciences 51, 1703-1711 (in Chinese with English abstract). Huismans, R., Beaumont, C., 2011. Depth-dependent extension, two-stage breakup and cratonic underplating at rifted margins. Nature 473, 74-78. Jackson, J., McKENZIE, D., Priestley, K., Emmerson, B., 2008. New

of

views on the structure and rheology of the lithosphere. Journal of the

ro

Geological Society 165, 453-465.

-p

Jiang, G.Z., Gao, P., Rao, S., Zhang, L.Y., Tang, X.Y., Huang, F., Zhao,

re

P., Pang, Z.H., Li-Juan, H.E., Sheng-Biao, H.U., 2016. Compilation of heat flow data in the continental area of China(4~(th) edition). Chinese Journal of

lP

Geophysics 59, 2892-2910 (in Chinese with English abstract).

na

Kao, H., Shen, S.s.J., Ma, K.F., 1998. Transition from oblique subduction

Jo ur

to collision: Earthquakes in the southernmost Ryukyu arc ‐Taiwan region. Journal of Geophysical Research: Solid Earth 103, 7211-7229. Kusznir, N., Park, R., 1987. The extensional strength of the continental lithosphere: its dependence on geothermal gradient, and crustal composition and thickness. Geological Society, London, Special Publications 28, 35-52. Li, F., Sun, Z., Pang, X., Liao, J., Yang, H., Xie, H., Zhuo, H., Zhao, Z., 2019. Low ‐Viscosity Crustal Layer Controls the Crustal Architecture and Thermal Distribution at Hyperextended Margins: Modeling Insight and Application to the Northern South China Sea Margin. Geochemistry, 32

Journal Pre-proof Geophysics, Geosystems, 20, 3248-3267. Lüdmann, T., Wong, H.K., 1999. Neotectonic regime on the passive continental margin of the northern South China Sea. Tectonophysics 311, 113-138. Maggi, A., Jackson, J., Mckenzie, D., Priestley, K., 2000. Earthquake focal

of

depths, effective elastic thickness, and the strength of the continental

ro

lithosphere. Geology 28, 495-498.

-p

Michel, G.W., Yu, Y.Q., Zhu, S.Y., Reigber, C., Becker, M., Reinhart, E.,

re

Simons, W., Ambrosius, B., Vigny, C., Chamot-Rooke, N., 2001. Crustal motion and block behaviour in SE-Asia from GPS measurements. Earth and

lP

Planetary Science Letters 187, 239-244.

na

Panza, G.F., Raykova, R.B., 2010. Structure and rheology of lithosphere

Jo ur

in Italy and surrounding. Terra Nova 20, 194-199. Park, R.G., 2013. Foundation of structural geology. Routledge. Pauselli, C., Ranalli, G., Federico, C., 2010. Rheology of the Northern Apennines: Lateral variations of lithospheric strength. Tectonophysics 484, 27-35. Pauselli C., Ranalli, G., 2017. Effects of lateral variations of crustal rheology on the occurence of post-orogenic normal faults : the Alto Tiberina Fault (Northern Apennines, Cental Utaly). Tectonophysics, 45-55. Peron-Pinvidic, G., Manatschal, G., 2019. Rifted Margins: State of the Art 33

Journal Pre-proof and Future Challenges. Frontiers in Earth Science 7: 218. Pichot, T., Delescluse, M., Chamot-Rooke, N., Pubellier, M., Qiu, Y., Meresse, F., Sun, G., Savva, D., Wong, K.P., Watremez, L., Auxiètre, J.L., 2014. Deep crustal structure of the conjugate margins of the SW South China Sea from wide-angle refraction seismic data. Marine and Petroleum Geology

of

58, 627-643.

ro

Pin, Y., Di, Z., Zhaoshu, L., 2001. A crustal structure profile across the

-p

northern continental margin of the South China Sea. Tectonophysics 338,

re

1-21.

Popov, A.A., Sobolev, S.V., 2008. SLIM3D: A tool for three-dimensional modeling

of

lP

thermomechanical

lithospheric

deformation

with

Jo ur

55-75.

na

elasto-visco-plastic rheology. Physics of the Earth and Planetary Interiors 171,

Popov, Y., Bayuk, I., Parshin, A., Miklashevskiy, D., Novikov, S., Chekhonin, E., 2012. New methods and instruments for determination of reservoir thermal properties, Proceedings, Thirty-Seventh Workshop on Geothermal Reservoir Engineering Stanford University, Stanford, California. Popov, Y., Beardsmore, G., Clauser, C., Roy, S., 2016. ISRM Suggested Methods for Determining Thermal Properties of Rocks from Laboratory Tests at Atmospheric Pressure. Rock Mechanics and Rock Engineering 49, 4179-4207. 34

Journal Pre-proof Pribnow, D., Williams, C.F., Sass, J.H., Keating, R., 1996. Thermal conductivity of water-saturated rocks from the KTB Pilot Hole at temperatures of 25 to 300°C. Geophysical Research Letters 23, 391-394. Ranalli, G., 1995. Rheology of the Earth. Springer Science & Business Media.

of

Ranalli, G., Adams, M., 2013. Rheological contrast at the continental

ro

Moho: Effects of composition, temperature, deformation mechanism, and

-p

tectonic regime. Tectonophysics 609, 480-490.

re

Ranalli, G., Murphy, D.C., 1987. Rheological stratification of the lithosphere. Tectonophysics 132, 281-295. L.,

Buntebarth,

G.,

lP

Rybach,

1982.

Relationships

between

the

na

petrophysical properties density, seismic velocity, heat generation, and

Jo ur

mineralogical constitution. Earth and Planetary Science Letters 57, 367-376. Shi, X., Chuanhai, Y., Mei, C., Yang, X., Zhao, J., 2017. Analyses of variation features and influential factors of heat flow in the northern margin of the South China Sea. Earth Science Frontiers 24, 56-64 (in Chinese with English abstract). Shi, X., Qiu, X., Xia, K., Zhou, D., 2003. Characteristics of surface heat flow in the South China Sea. Journal of Asian Earth Sciences 22, 265-277. Shi, X., Zhou, D., Zhang, Y., 2000. Lithospheric thermal-rheological structures of the continental margin in the northern South China Sea. Chinese 35

Journal Pre-proof Science Bulletin 45, 2107-2112 (in Chinese with English abstract). Tang, D.L.K., Seward, D., Wilson, C.J.N., Sewell, R.J., Carter, A., Paul, B.T., 2014a. Thermotectonic history of SE China since the Late Mesozoic: insights from detailed thermochronological studies of Hong Kong. Journal of the Geological Society 171, 591-604.

of

Tang, X.Y., Sheng-Biao, H.U., Zhang, G.C., Yang, S.C., Shen, H.L., Rao,

ro

S., Wei-Wei, L.I., 2014b. Characteristic of surface heat flow in the Pearl River

-p

Mouth Basin and its relationship with thermal lithosphere thickness. Chinese

re

Journal of Geophysics 57, 1857-1867 (in Chinese with English abstract). Taylor, B., Hayes, D.E., 1983. Origin and history of the South China Sea

na

Islands: Part 2 27, 23-56.

lP

basin. The Tectonic and Geologic Evolution of Southeast Asian Seas and

Jo ur

Tetreault, J.L., Buiter, S.J.H., 2018. The influence of extension rate and crustal rheology on the evolution of passive margins from rifting to break-up. Tectonophysics 746, 155-172. Wan, K., Xia, S., Cao, J., Sun, J., Xu, H., 2017. Deep seismic structure of the northeastern South China Sea: Origin of a high-velocity layer in the lower crust. Journal of Geophysical Research: Solid Earth 122, 2831-2858. Wang, H.B., Jin, L., Hua, G.Y., Yang, H.Y., Wei, L.X., Sha, Z.B., 2005. Estimation of the Heat Flow in the Northern of the South China Sea Based on the Seismic Data of Gas Hydrate. Geoscience 19, 67-73 (in Chinese with 36

Journal Pre-proof English abstract).. Wang, L., Li, C., Liu, F., Li, H., Lu, H., Liu, S., 2000. Thermal-rheological structure of the lithosphere beneath two types of basins in eastern and western China. Science in China Series D: Earth Sciences 43, 200. Wang, P., Li, Q., 2009. History of the South China Sea – A Synthesis, in:

of

Wang, P., Li, Q. (Eds.), The South China Sea: Paleoceanography and

ro

Sedimentology. Springer Netherlands, Dordrecht, pp. 487-496.

-p

Wang, Q., Zhang, P.Z., Freymueller, J.T., Bilham, R., Larson, K.M., Lai, X.,

re

You, X., Niu, Z., Wu, J., Li, Y., Liu, J., Yang, Z., Chen, Q., 2001. Present-day crustal deformation in China constrained by global positioning system

lP

measurements. Science 294, 574-577.

na

Wang, Y., Hu, S., Wang, Z., Jiang, G., Hu, D., Zhang, K., Gao, P., Hu, J.,

Jo ur

Zhang, T., 2019. Heat flow, heat production, thermal structure and its tectonic implication of the southern Tan-Lu Fault Zone, East–Central China. Geothermics 82, 254-266.

Wang, Z.-C., Zang, J.F., Green II, H.W., 2012. Mafic granulite rheology: Implications for a weak continental lower crust. Earth and Planetary Science Letters s 353-354:99-107. Watts, A.B., Burov, E.B., 2003. Lithospheric strength and its relationship to the elastic and seismogenic layer thickness. Earth and Planetary Science Letters 213, 113-131. 37

Journal Pre-proof Watts, A.B., Zhong, S.J., Hunter, J., 2013. The Behavior of the Lithosphere on Seismic to Geologic Timescales. Annual Review of Earth and Planetary Sciences 41, 443-468. Wilks, K.R., and Carter, N.L., 1990. Rheology of some continental lower crustal rocks. Tectonophysics, 57- 77.

of

Wu, J., Zhang, Z., 2012. Spatial distribution of seismic layer, crustal

ro

thickness, and Vp/Vs ratio in the Permian Emeishan Mantle Plume region.

-p

Gondwana Research 22, 127-139.

re

Wu, Z., Li, J., Ruan, A., Lou, H., Ding, W., Niu, X., Li, X., 2011. Crustal structure of the northwestern sub-basin, South China Sea: Results from a

lP

wide-angle seismic experiment. Science China Earth Sciences 55, 159-172.

na

Xia, K.-y., Huang, C.-l., Jiang, S.-r., Zhang, Y.-x., Su, D.-q., Xia, S.-g.,

Jo ur

Chen, Z.-r., 1994. Comparison of the tectonics and geophysics of the major structural belts between the northern and southern continental margins of the South China Sea. Tectonophysics 235, 99-116. Xia, S., Zhao, F., Zhao, D., Fan, C., Wu, S., Mi, L., Sun, J., Cao, J., Wan, K., 2018. Crustal plumbing system of post-rift magmatism in the northern margin of South China Sea: New insights from integrated seismology. Tectonophysics 744, 227-238. Xiao, H., Xue, M., Yang, T., Liu, C., Hua, Q., Xia, S., Huang, H., Le, B.M., Yu, Y., Huo, D., Pan, M., Li, L., Gao, J., 2018. The Characteristics of 38

Journal Pre-proof Microseisms in South China Sea: Results From a Combined Data Set of OBSs, Broadband Land Seismic Stations, and a Global Wave Height Model. Journal of Geophysical Research: Solid Earth 123, 3923-3942. Yan, P., Deng, H., Liu, H., Zhang, Z., Jiang, Y., 2006. The temporal and spatial distribution of volcanism in the South China Sea region. Journal of

of

Asian Earth Sciences 27, 647-659.

ro

Ye, Q., Mei, L., Shi, H., Camanni, G., Shu, Y., Wu, J., Yu, L., Deng, P., Li,

-p

G., 2018a. The Late Cretaceous tectonic evolution of the South China Sea

re

area: An overview, and new perspectives from 3D seismic reflection data. Earth-science reviews 187, 186-204.

lP

Ye, Q., Mei, L., Shi, H., Shu, Y., Camanni, G., Wu, J., 2018b. A low-angle

na

normal fault and basement structures within the Enping Sag, Pearl River

Jo ur

Mouth Basin: Insights into late Mesozoic to early Cenozoic tectonic evolution of the South China Sea area. Tectonophysics 731-732, 1-16. Zang, S.X., Wei, R.Q., Ning, J.Y., 2007. Effect of brittle fracture on the rheological structure of the lithosphere and its application in the Ordos. Tectonophysics 429, 267-285. Zhao, M., Qiu, X., Xia, S., Xu, H., Wang, P., Wang, T.K., Lee, C.-S., Xia, K., 2010. Seismic structure in the northeastern South China Sea: S-wave velocity and Vp/Vs ratios derived from three-component OBS data. Tectonophysics 480, 183-197. 39

Journal Pre-proof Zhao, Y., Ren, J., Pang, X., Yang, L., Zheng, J., 2018. Structural style, formation of low angle normal fault and its controls on the evolution of Baiyun Rift, northern margin of the South China Sea. Marine and Petroleum Geology 89, 687-700. Zhou, D., Ru, K., Chen, H.-z., 1995. Kinematics of Cenozoic extension on

of

the South China Sea continental margin and its implications for the tectonic

ro

evolution of the region. Tectonophysics 251, 161-177.

-p

Zhou, Z., 2018. The Cenozoic crustal thinning and development of

re

hyper-extended rift system in the northern South China Sea. PhD thesis, China University of Geosciences (in Chinese with English abstract).

lP

Zhou, Z., Mei, L., Liu, J., Zheng, J., Chen, L., Hao, S., 2018.

na

Continentward-dipping detachment fault system and asymmetric rift structure

Jo ur

of the Baiyun Sag, northern South China Sea. Tectonophysics 726, 121-136. Zhu, A.-L., Xu, X.-W., Zhou, Y.-S., Yin, J.-Y., Gan, W.-J., Chen, G.-H., 2005. Relocation of small earthquakes in western Sichuan, China and its implications for active tectonics. Chinese Journal of Geophysics 48, 629-636 (in Chinese with English abstract). Zuber, M.T., Parmentier, E.M., Fletcher, R.C., 1986. Extension of continental lithosphere - A model for two scales of basin and range deformation. Journal of Geophysical Research Solid Earth 91, 4826-4838.

40

Jo ur

na

lP

re

-p

ro

of

Journal Pre-proof

Fig. 1. (a) Map of the northern margin of the South China Sea. The red line denotes profile OBS1993. The blue line depicts the deep seismic reflection profile DSRP2002. Circle, triangle and x marker represent wells from which rock thermophysical properties were acquired during this study. The major faults compiled from Ye et al. (2018 a,b), Zhou et al. (2018) and Zhao (2018). The shaded area is the Pearl River Mouth Basin (PRMB). (b) Heat flow map of the northern margin of the SCS. The black 41

Journal Pre-proof

-p

ro

of

line denotes profile OBS1993. Red points denote the locations of heat flow data.

Jo ur

na

lP

re

Fig. 2. (a) Surface heat flow (from Fig. 1b) and (b) geological interpretation along profile OBS1993, the P wave velocity (km/s) model from airgun and explosive OBS data (Pin et al., 2001).

42

lP

re

-p

ro

of

Journal Pre-proof

Jo ur

na

Fig. 3. Thermal conductivity histograms for sediments.

43

Jo ur

na

lP

re

-p

ro

of

Journal Pre-proof

Fig. 4. Work flow of the model optimization. The dashed boxes are the improvements made to the original flow after initial trials. BHF is bottom heat flow. BHF0 and New BHF are initial bottom heat flow and new bottom heat flow generated by SQP algorithm, respectively.

44

lP

re

-p

ro

of

Journal Pre-proof

Jo ur

na

Fig. 5. (a) Optimization result and (b) misfit and step size evolution for the workflow without the steps removing heat flow outliners and moving average in Fig. 4 (those in dashed boxes).

45

lP

re

-p

ro

of

Journal Pre-proof

Jo ur

na

Fig. 6. (a) Optimization result and (b) misfit and step size evolution of misfit and step size for the improved workflow (Fig. 4, including the dashed boxes).

46

of

Journal Pre-proof

Jo ur

na

lP

re

-p

ro

Fig. 7. Temperature (K) distribution along the OBS1993 profile.

47

Jo ur

na

lP

re

-p

ro

of

Journal Pre-proof

Fig. 8. Rheological profiles for the hard case. (a) and (b) are 2-D rheological profile for the conventional model (CM) and high-pressure failure model (ZM). The white circle denotes the earthquake, the size of which mean the magnitude. Those earthquakes are from orange box in Fig.9a, the lateral distance is calculated from the projection of location of earthquakes to OBS1993 profile. Note that the stress scales (MPa) are different in the two cases. Upper crust includes sediments. Red lines mark the positions of the 1-D profiles in (c). (c) 1-D rheological profiles for selected locations including CM and ZM cases. UC, LC, HVL, and LM denote upper crust, lower crust, high-velocity layer, and lithospheric mantle, respectively. 48

Journal Pre-proof

of

Fig. 9. (a) Earthquakes in the study area. Size of the circle donates the magnitude from

ro

0.8 to 6.7; color donates the depth. Those in the orange box is plotted in Fig.8 a,b and

Jo ur

na

lP

re

-p

Fig.A1a,b. (b) Frequency distribution of earthquakes in orange box in panel (a).

Fig. 10 Structure along profile DSRP2002. Solid red lines are resolved faults, dashed red lines are inferred/possible faults (Huang, 2005) (see Fig. 9a for location). Aspect ratio is 4:1. Tg, seismic interfaces, represents sediment basement. The Moho stepped up from shelf to continental slope and abyssal plain, and from north to south.

49

Journal Pre-proof Table 1 Thermal conductivities of the boreholes in PRMB. The factor G = standard deviation/mean and InhomoFactor = (maximum-minimum)/mean Thermal conductivity

Corrected

(W/ (m·K)) well

Formation

thermal

Depth

G

(m)

(%)

Lithology

Temperature InhomoFactor

mean

min

conductivity (℃)

max

(W/ (m·K))

Fine Zhuhai

3089

2.658

2.511

2.822

2.757

0.117

108

2.22

3460

2.435

2.291

2.558

2.905

0.110

121

2.01

3636

3.063

2.858

3.339

4.488

0.157

128

2.42

3639

3.490

3.291

3.670

3.307

0.108

128

2.71

3648

2.904

2.811

2.994

1.832

0.063

128

2.31

3655

3.444

3.297

3.719

3.058

0.122

128

2.67

1.433

1.401

1.473

1.719

0.050

53

1.39

Sandstone Fine Zhuhai Sandstone

Sandstone NH1

Sandstone

Fine Zhuhai Sandstone

re

Sandstone

-p

Fine Zhuhai

ro

Fine Zhuhai

of

Fine Zhuhai

Mudstone

1394

Hanjiang

Mudstone

1395

1.256

1.209

1.293

2.329

0.067

53

1.23

Hanjiang

Mudstone

1401

1.615

1.515

1.687

3.494

0.107

53

1.55

Hanjiang

Mudstone

1402

1.302

1.263

1.367

2.59

0.080

53

1.27

Hanjiang

Mudstone

NH2

1405

1.222

1.147

1.364

6.06

0.178

53

1.20

1409

2.445

2.218

2.574

6.057

0.145

53

2.29

Jo ur

Sandstone

na

Fine Hanjiang

lP

Hanjiang

Hanjiang

Mudstone

1414

1.245

1.192

1.294

2.346

0.082

54

1.22

Zhuhai

Sandstone

1631

2.000

1.905

2.160

4.296

0.128

62

1.87

Sandstone

1637

1.931

1.871

1.985

1.491

0.059

62

1.81

1638

2.281

2.255

2.355

1.654

0.044

62

2.11

Zhuhai

Fine

Zhuhai

Sandstone Zhuhai

Gritstone

1639

1.672

1.636

1.701

1.588

0.039

62

1.59

Zhuhai

Gritstone

1649

1.781

1.561

1.981

7.533

0.236

62

1.68

Zhujiang

Conglomerate

2569

1.368

1.843

2.034

2.901

0.099

98

1.29

Zhujiang

Sandstone

2573

1.921

1.900

2.418

5.887

0.238

98

1.71

Zhujiang

Sandstone

2575

2.178

1.293

1.423

2.487

0.095

98

1.90

Zhujiang

Sandstone

2579

1.931

1.806

2.160

5.395

0.183

98

1.71

Enping

Sandstone

3974

3.599

3.438

4.003

1.162

0.157

140

2.71

Enping

Conglomerate

3977

3.638

3.387

3.860

4.191

0.130

140

2.73

Enping

Conglomerate

3980

3.686

3.405

4.104

5.023

0.190

140

2.76

Enping

Conglomerate

3983

3.123

3.058

3.183

4.134

0.040

141

2.40

Wenchang

Conglomerate

3754

2.920

2.683

3.407

5.883

0.248

137

2.28

NH3

NH6

NH7

50

Journal Pre-proof Wenchang

Conglomerate

3755

3.016

2.845

3.130

2.883

0.094

137

2.34

Wenchang

Conglomerate

3756

2.994

2.761

3.351

5.089

0.197

137

2.33

Wenchang

Conglomerate

3757

2.791

2.511

2.983

4.399

0.169

137

2.20

Wenchang

Conglomerate

3758

2.717

2.373

2.877

5.998

0.186

137

2.15

Wenchang

Mudstone

3760

2.589

2.460

2.717

2.598

0.099

137

2.06

Wenchang

Mudstone

3765

2.652

2.553

2.807

2.885

0.096

137

2.10

3766

3.186

2.958

3.502

4.012

0.171

137

2.45

Fine Wenchang

3770

2.521

2.423

2.648

3.073

0.089

137

2.02

Wenchang

Mudstone

3774

2.953

2.788

3.243

4.645

0.154

137

2.30

Zhujiang

Sandstone

2571

2.718

2.491

2.963

5.053

0.173

92

2.34

Zhujiang

Sandstone

2576

2.806

2.708

2.972

2.614

0.094

92

2.40

Zhujiang

Sandstone

2580

2.550

2.465

2.731

3.098

0.105

92

2.21

Zhujiang

Sandstone

2178

2.609

2.515

2.727

1.971

0.081

93

2.24

Zhujiang

Sandstone

2192

1.910

1.854

1.996

1.971

0.074

94

1.71

2195

3.243

2.835

3.578

4.91

0.229

94

2.73

2198

2.723

2.583

2.919

3.617

0.123

94

2.33

Zhujiang Sandstone Fine Zhujiang

Sandstone

2607

2.759

2.913

2.72

0.114

106

2.30

Zhujiang

Mudstone

2612

2.565

2.418

2.756

4.355

0.132

106

2.16

Zhujiang

Sandstone

2614

2.736

2.509

2.947

4.053

0.160

106

2.28

Zhujiang

Sandstone

2617

2.608

2.384

2.760

3.125

0.144

106

2.19

Zhujiang

Sandstone

Zhujiang

Sandstone

Zhujiang

Sandstone

lP

Zhujiang

2622

1.131

1.046

1.275

5.159

0.203

106

1.11

2625

1.999

1.775

2.143

5.523

0.184

107

1.74

2630

1.577

1.497

1.676

3.371

0.113

107

1.43

Jo ur

NH12

2.598

na

Sandstone

-p

Fine NH10

of

Mudstone

ro

NH8

Wenchang

re

Sandstone

Zhujiang

Sandstone

2636

1.813

1.694

1.910

3.093

0.119

107

1.61

Zhujiang

Sandstone

2638

1.990

1.927

2.043

1.73

0.058

107

1.74

Zhujiang

Sandstone

2642

2.604

2.291

2.810

5.516

0.200

107

2.18

Zhujiang

Sandstone

2645

2.408

2.161

2.635

5.844

0.197

107

2.04

Zhujiang

Sandstone

2648

2.270

1.984

2.588

8.457

0.266

108

1.94

2655

2.563

2.166

2.828

6.198

0.258

108

2.15

2657

2.094

2.002

2.175

2.648

0.083

108

1.81

2660

2.262

2.047

2.429

4.297

0.169

108

1.93

Fine Zhujiang Sandstone Zhujiang

Mudstone Fine

Zhujiang Sandstone

NH16

NH17

Enping

Mudstone

4627

2.075

1.891

2.230

3.361

0.163

168

1.66

Enping

Siltstone

4628

2.784

2.589

2.954

2.484

0.131

168

2.08

Enping

Siltstone

4630

3.267

2.791

3.684

4.627

0.273

169

2.36

Zhujiang

Fine Stone

2719

2.713

2.553

2.832

3.637

0.103

108

2.26

Zhujiang

Fine Stone

2720

2.406

2.240

2.549

7.319

0.129

108

2.04

Zhujiang

Sandstone

2723

2.590

2.419

2.801

2.714

0.148

108

2.17

51

Journal Pre-proof Zhujiang

Sandstone

2738

2.388

2.285

2.530

4.188

0.103

108

2.02

Zhujiang

Siltstone

2764

3.472

3.118

3.718

4.003

0.173

109

2.80

Zhujiang

Siltstone

2769

2.972

2.688

3.191

2.698

0.170

110

2.44

Zhujiang

Siltstone

2772

2.894

2.687

3.092

5.106

0.140

110

2.38

Zhujiang

Sandstone

2777

2.784

2.427

3.199

4.683

0.278

110

2.30

Zhujiang

Sandstone

2781

2.775

2.557

3.153

3.676

0.215

110

2.29

pre-Eocene

Diorite

4414

2.682

2.516

2.919

9.24

0.150

165

2.03

pre-Eocene

Amphibolite

4416

3.107

3.017

3.248

5.745

0.074

165

2.28

Jo ur

na

lP

re

-p

ro

of

NH14

52

Journal Pre-proof Table 2. Concentrations of U, Th, and K, rock density, and radiogenic heat production (A) data for the PRMB.

NH1

NH2

(m)

Density

Lithology

Th

U

K

A

(kg·m )

(ppm)

(ppm)

(%)

(μW·m-3)

-3

Zhuhai

3460

Mudstone

1.93

21.7

3.14

3.79

1.91

Zhuhai

3089

Sandstone

2.48

9.12

0.962

2.25

1.00

Zhujiang

1647

Gritstone

2.33

7.76

1.26

2.09

0.91

Hanjiang

1403

Fine Sandstone

2.49

6.19

1.44

1.32

0.85

Hanjiang

1395

Fine Sandstone

2.62

13.6

2.8

2.14

1.81

Zhujiang

2579

Sandstone

2.18

30.6

6.81

3.43

3.39

Zhujiang

2580

Sandstone

2.39

4.28

1.11

2.44

0.72

Wenchang

3774

Mudstone

2.40

28.2

6.57

2.96

3.48

Wenchang

3969

Sandstone

2.37

Wenchang

3756

Conglomerate

2.53

Wenchang

4158

Sandstone

2.51

Wenchang

4164

Sandstone

2.55

Wenchang

4171

Sandstone

Enping

4628

Fine Sandstone

Enping

4143

Gritstone

Zhujiang

2765

Gritstone

Zhujiang

2779

Zhujiang

8.55

3.11

3.57

4.02

0.751

1.97

0.62

18.5

4.86

4.7

2.76

12.5

3.27

3.61

1.93

2.49

10.9

2.13

2.78

1.44

2.35

11.7

3.54

3.72

1.80

2.32

re

5.03

1.3

1.7

0.72

2.07

17.7

2.66

2.63

1.66

Gritstone

2.02

3.17

2.74

1.09

0.77

3222

Sandstone

2.38

6.79

1.11

1.88

0.82

Zhujiang

2781

Gritstone

2.46

3.78

1.63

0.36

0.65

Zhujiang

2764

Fine Sandstone

2.40

14.9

2.44

1.12

1.57

Zhujiang

3347

Siltstone

2.39

12.3

2.13

2.13

1.41

Zhujiang

2410

Gritstone

2.39

4.77

2.25

1.57

0.93

Zhujiang

2407

Gritstone

2.30

13

6.79

2.13

2.42

Zhuhai

3690

Gritstone

2.34

6.6

5.47

1.7

1.76

Zhuhai

3688

Gritstone

2.49

6.71

1.48

1.91

0.94

Enping

3915

Mudstone

2.22

23

5.95

3.51

2.83

Enping

3905

Fine Sandstone

2.52

21

4.25

3.46

2.68

NH12

Zhujiang

2606

Mudstone

2.31

17.3

3.56

2.85

2.04

NH11

Wenchang

3982

Conglomerate

2.55

3.6

0.933

1.69

0.61

NH15

NH17

NH13 NH5 NH4

-p

NH16

lP

NH9

ro

22.8

Jo ur

NH7

Depth

na

NH8

Formation

of

Well

53

Journal Pre-proof Table 3. Values of parameters used in temperature distribution calculations (compiled from Afonso and Ranalli, 2004;Haenel et al., 2012; Zhang and Jia‐ Biao, 2011;Gemmer et al., 2002; Côté and Konrad, 2005 and Gerya, 2009). Thermal

Density,

conductivity, −1

(W·m ·k ) Sediments

References

production, 𝜌 (g ·

K −1

Heat A

cm )

(mW·m−2)

2.36

1.65

This article

−3

2.1 2.56

2.53

2.0

Liu et al. (2019), Beaumont et al. (2004)

Lower crust

2.6

2.80

0.4

Liu et al. (2019), He et al. (2003),Gerya, (2009)

HVL

2.2

2.92

0.31

Côté et al. (2005), Haenel et al. (2012)

Lithospheric

3.3

3.32

0.003

of

Upper crust

Afonso et al. (2004), Zhang et al.(2011)

-p

ro

mantle

Table 4. Rheological parameters: A (MPa-ns-1), n, E (kJ·mol-1) power-law creep parameters compiled from

re

Gerya, 2009; Pauselli et al., 2010; Ranalli, 1995 and Ranalli and Adams, 2013; B0 (MPa), K, m, α, β, γ high-pressure failure parameters (after Zang et al., 2007 and references therein).

Upper crust (including

Rheology

mantle

E

B0

K

m

𝛂

𝛃

γ

2.3

154

n.a

n.a

n.a

n.a

n.a

n.a

ZM

3.2 × 10−4

2.3

154

34.1

4.57

0.52

−1.128

1.732

0.035

CM-soft

8.0 × 10−3

3.1

243

n.a

n.a

n.a

n.a

n.a

n.a

CM-hard

1.4 × 104

4.2

445

n.a

n.a

n.a

n.a

n.a

n.a

ZM-soft

8.0 × 10−3

3.1

243

36.1

3.18

0.55

−2.536

2.340

0.035

ZM-hard

104

4.2

445

36.1

3.18

0.55

−2.536

2.340

0.035

3.3 ×

10−4

3.2

238

n.a

n.a

n.a

n.a

n.a

n.a

3.3 ×

10−4

3.2

238

36.1

3.18

0.55

−2.536

2.340

0.035

2.0 ×

103

4.0

471

n.a

n.a

n.a

n.a

n.a

n.a

2.5 ×

104

3.5

532

n.a

n.a

n.a

n.a

n.a

n.a

ZM-soft

2.0 ×

103

4.0

471

28.3

3.35

0.68

−1.875

1.310

0.035

ZM-hard

2.5 × 104

3.5

532

28.3

3.35

0.68

−1.875

1.310

0.035

Jo ur ZM

CM-soft Lithospheric

n

3.2 × 10−4

CM

HVL

High pressure failure parameters

CM

sediment)

Lower crust

A

na

Layer

lP

Power law creep parameters

CM-hard

1.4 ×

n.a = not applicable.

54

Journal Pre-proof

Author statement Jie Hu: Methodology, Software, Data Curation, Formal analysis, Writing-Original Draft; Yuntao Tian: Conceptualization, Methodology, Writing-Review & Editing; Zulie Long: Resources, Investigation; Di Hu: Visualization; Formal analysis; Yuping Huang: Resources, Supervision; Yibo Wang: Visualization, Validation; Shengbiao Hu: Project administration, Funding acquisition;

of

Declaration of interests

-p

ro

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

re

☐The authors declare the following financial interests/personal relationships which

Jo ur

na

lP

may be considered as potential competing interests:

Highlights 1) We present new thermal conductivity (72) and heat production (29) measurements. 2) The rheology model with mafic granulite lower crust and dry peridotite lithospheric mantle fits geological and geophysical observations. 3) Our results suggest a relatively strong lithosphere before extension, explaining the asymmetric geometry of the SCS.

55

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10