Contents lists available at ScienceDirect

Journal of Asian Earth Sciences

journal homepage: www.elsevier.com/locate/jseaes

Full length article

Curie point depths in Northeast China and their geothermal implications forthe Songliao Basin

Jian Wanga, Chun-Feng Lib,c,⁎

a Key Laboratory of Crustal Dynamics, Institute of Crustal Dynamics, China Earthquake Administration, Beijing 100085, ChinabDepartment of Marine Sciences, Zhejiang University, Zhoushan 316021, Chinac Laboratory of Marine Mineral Resources, Qingdao National Laboratory for Maine Science and Technology, Qingdao 266237, China

A R T I C L E I N F O

Keywords:NGDC-720 magnetic anomalyCurie-point depthThermal lithospheric thicknessHeat flowNortheast ChinaSongliao Basin

A B S T R A C T

This work integrates processing and interpretation of the combined geomagnetic anomaly model of NGDC-720-V3.1 to better constrain thermal structures of the Northeast China lithosphere, which shows intensive extensionand active volcanism during the Mesozoic and Cenozoic. Numerical tests demonstrate that magnetic anomaliescalculated at 20 km altitude can be used for accurate Curie depth estimation. Averaged Curie depths using a 3Dfractal exponent of 3.0 vary from ∼ ±13 2.0 to ±33 4.1 km with a mean of ∼ ±22 2.4 km in Northeast China.With temperature-dependent thermal conductivity models constrained by incorporated surface heat flow andCurie depths in different geological blocks, the thermal lithospheric thicknesses are estimated mostly between 45and 90 km based on 1D steady-state thermal conduction equation. Beneath the Songliao Basin, the Moho tem-peratures are abnormally high, ranging from ca. 700 to 1000 °C, and the mantle contributes about 65% of thetotal surface heat flow. The shallow Curie points and thermal lithosphere-asthenosphere boundary, together withhigh Moho temperatures and mantle heat flow, support abnormally high geotherms in the Songliao Basin,originated from hot upper mantle upwelling triggered likely by lithospheric delamination.

1. Introduction

Heat derived from the Earth’s interior governs magmatism, me-chanical properties, and thickness of the lithosphere (e.g., Fowler,2005). It is widely accepted that the eastern China block (Fig. 1). ex-perienced intensive rifting and voluminous basaltic magmatism fromthe Mesozoic to the Cenozoic, has a thinned lithosphere (Deng et al.,2004; Li, 2010; Wan et al., 2016; Zhou, 2006). Geothermal modelingbased on S-wave tomography and surface heat flow revealed that thelithospheric thickness of eastern China is as thin as ∼80–120 km (Anand Shi, 2006; Deng and Tesauro, 2016; Wang and Cheng, 2012),suggesting high geothermal anomalies beneath eastern China. Situatedbetween the Siberian and North China cratons, Northeast China (NEChina) has undergone multiple subductions of the Paleo-Asian, Mongol-Okhotsk and Pacific Oceans since the Paleozoic, and is composed ofseveral micro-continental terranes/blocks (Fig. 1). Since the Late Me-sozoic, granitoids emplacement, extensive volcanism and lithosphericthinning have occurred in NE China, causing the Songliao Basin to havethe highest heat flow and geothermal gradient (∼30–64 °C/km) amongthe sedimentary basins in China (Chen et al., 2007; Hou et al., 2009;Liu, 1982; Liu et al., 2017; Meng, 2003; Ren et al., 2002; Tian et al.,

1992; Wang et al., 1999; Wu et al., 2002, 2011; Xu et al., 2013). Eventhough the major lithospheric structures of NE China have been wellresolved with seismic tomography since the deployment of the NE-CESSArray portable stations (e.g., Guo et al., 2016), lithosphericthermal structures of NE China and the origin of the abnormal geo-therms in the Songliao Basin are poorly understood, due to the diffi-culties in isolating the temperature from composition effects on seismicvelocity (e.g., Dalton et al., 2017).

Surface heat flow measurements are usually used for investigatingthe lithospheric thermal structure. However, heat flow cannot constrainwell the entire lithospheric thermal structure due to their shallowmeasurements, uneven distributions, and other factors such as geo-thermal circulation and intraplate magmatism (Siler and Kennedy,2016). To partially circumvent this problem, the depths to the bottomof magnetic layer, also known as Curie-point depths, estimated fromspectral analyses of magnetic anomalies, can provide complementarygeothermal information of the lithosphere (Bouligand et al., 2009; Li,2011; Li et al., 2012, 2013; Okubo et al., 1985; Ross et al., 2006; Tanakaet al., 1999; Wang and Li, 2015; Wang et al., 2016a). Hu et al. (2006)reported Curie depths of ∼20–40 km in NE China from aeromagneticanomalies using a random magnetization model. Sun et al. (2012) and

https://doi.org/10.1016/j.jseaes.2018.05.026Received 25 January 2018; Received in revised form 4 May 2018; Accepted 26 May 2018

⁎ Corresponding author at: Department of Marine Sciences, Zhejiang University, Zhoushan 316021, China.E-mail address: cfli@zju.edu.cn (C.-F. Li).

Journal of Asian Earth Sciences 163 (2018) 177–193

Available online 26 May 20181367-9120/ © 2018 Elsevier Ltd. All rights reserved.

T

Li et al. (2015) estimated Curie depths (∼15–34 km) in the Hailar andSongliao basins using the Parker method (Parker, 1973) and 4th orderwavelet approximate, respectively. At regional and global scales, Curiedepths have been evaluated based on the Earth’s Magnetic AnomalyGrid in 2-minute resolution (EMAG2, Maus et al., 2009) with a 3Dfractal magnetization correction (Li et al., 2017; Li and Wang, 2016,2018).

Over the past 40 years, numerous satellite magnetic anomaly mapswere produced from POGO, Magsat and CHAMP data (Langel andHinze, 1998; Maus et al., 2008). Excellent spatial and temporal datacoverage has been achieved thanks to the launch of CHAMP, leading toa highly accurate global lithospheric magnetic model (MF6) with aresolution of 333 km wavelength (Maus et al., 2008). Curie depths wereestimated from satellite models MF3 in Antarctica (Maule et al., 2005),MF5 in India (Rajaram et al., 2009) and MF6 in Tibet (Hemant andMitchell, 2009). Recently, combined models from satellite, marine andaeromagnetic data (e.g., NGDC-720, EMM2010 and WDMAM2.0) wereapplied to estimated Curie depths (Arnaiz-Rodríguez and Orihuela,2013; Gao et al., 2013, 2015c, 2017; Salazar et al., 2017). Model NGDC-720 version 3.1 (http://www.geomag.org/models/ngdc720.html) wasreleased by National Geophysical Data Center, and was constructedfrom the magnetic grids of EMAG2 (http://www.geomag.org/models/emag2.html) and MF6 (http://geomag.org/models/MF6.html). It re-solves the lithospheric magnetic field to a spherical harmonic degree720, corresponding to a spatial wavelength of 56 km. Since the longwavelength component is more reliably determined by satellite, thewavelengths larger than 330 km (∼degree 120) are replaced with aCHAMP satellite geomagnetic model MF6 (Maus, 2010).

Here we explore magnetic vector and gradient tensor componentsderived from the NGDC-720-V3.1 model at varying observation alti-tudes. These components, especially the vertical vectors and gradienttensors, can relocate and enhance boundaries of magnetic sources andtectonic sutures in NE China. Magnetic anomalies calculated at variousaltitudes help discern deep sources and delineate regional tectonicunits. Then we estimate the Curie depths in NE China using spectralanalyses with a fractal magnetization correction based on the NGDC-

720-V3.1 model. By integrating surface heat flow and estimated Curiedepths, we construct lithospheric temperature-dependent thermalconductivity models for different geological blocks. Based on the 1Dsteady-state heat conduction equation, we assess the thermal litho-spheric thickness, Moho temperature and mantle heat flow, and finallycharacterize the lithospheric geothermal state of NE China and in-vestigate the origin of the abnormal geotherms in the Songliao Basin.

2. Regional geological framework

Situated in the eastern part of the Central Asian Orogen Belt, NEChina has undergone subductions of the Paleo-Asian, Mongol-Okhotskand Pacific Oceans since the Paleozoic (Fig. 1) (e.g., Liu et al., 2017; Wuet al., 2011; Xu et al., 2013). There are two regional collisional suturezones, the Mongol-Okhotsk Suture Zone (MOSZ) to the northwest andthe Solonker-Xar Moron-Changchun Suture Zone (SXCS, S3 in Fig. 1) tothe south, and three local sutures within NE China. The final closure ofthe Paleo-Asian Ocean along the SXCS took place from the Late Per-mian-Early Triassic in the west to the Late Permian-Middle Triassic inthe east (Liu et al., 2017; Zeng et al., 2012). The succeeding closure ofthe Mongol-Okhotsk Ocean occurred between the Late Jurassic and thebeginning of the Early Cretaceous (Kravchinsky et al., 2002; Metelkinet al., 2010; Van der Voo et al., 2015), leading to the final amalga-mation of the Siberian and North China cratons.

The NE China Block consists of four sub-blocks, Erguna, Xing’an,Songliao and Jiamusi, from northwest to southeast/east (Fig. 1). TheErguna Block (EB) is between the MOSZ to the northwest and Xing’anBlock (XB) to the southeast. The EB as the eastern extension of theCentral Mongolian Massif accreted to the Siberia Craton in the EarlyPaleozoic (Wu et al., 2011). The XB is mainly composed of GreatXing’an Range, which is characterized by large volumes of Late Meso-zoic volcanic rocks and granitoids (Fig. 1) (Wu et al., 2002, 2011; Xuet al., 2013; Zeng et al., 2012). The EB and XB were welded togetherfrom oceanic island accretions by westward subduction along theXinlin-Xiguitu Suture (XXS) in ca. 500Ma (Liu et al., 2017). The Son-gliao Block (SB) is located in the central part of NE China, and mainly

Fig. 1. Tectonic map of Northeast China (Chen et al.,2007; Liu et al., 2017; Sun et al., 2015; Wilde, 2015;Zeng et al., 2012). The dashed purple and red linesare the sutures and major faults, respectively. Theblue line outlines the Songliao Basin. The Paleogeneto Holocene volcanoes are shown as squares, pointsand triangles in various colors. The unfilled trianglesrepresent undated volcanoes. The light and dark grayareas delineate the Cretaceous granitoids and Cen-ozoic basalts (Wang et al., 2015; Wu et al., 2011;Zeng et al., 2012), respectively. The inset shows thelocation of the study area. F1: Derbugan Fault (DF);F2: Nenjiang Fault (NF); F3: Songliao Basin CentralFault (SBCF); F4: Jiamusi-Yilan Fault (JYF); F5:Dunhua-Mishan Fault (DMF); F6: Yujinshan Fault(YF); F7: Chifeng-Kaiyuan Fault (CKF); S1: Xinlin-Xiguitu Suture (XXS), sutured around 500Ma; S2:Heihe-Hegenshan Suture (HHS), sutured between thelate Early Carboniferous and early Late Carboni-ferous; S3: Solonker-Xar Moron-Changchun Suture(SXCS), sutured between the Late Permian and EarlyTriassic in the west and between the Late Permianand Middle Triassic in the east; S4: Mudanjian-YilanSuture (MYS); JB: Jiamusi Block; ABV: Abaga Vol-canic Field; CBV: Changbai Volcanic Field; HRV:Halaha River Volcanic Field; JPV: Jingpo VolcanicField; LGV: Longgang Volcanic Field; NRV: NuominRiver Volcanic Field; WDLCV: Wudalianchi VolcanicField.

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

178

consists of Great Xing’an Range in the southwest, the Lesser Xing’anRange in the northeast, the Zhangguangcai Range in the east and theSongliao Basin in the central area. Geochronological data indicate thatthe Mesozoic granitoids in the southwest section of the block (GreatXing’an Range) are mostly of Late Jurassic and Early Cretaceous (Wuet al., 2011). The final amalgamation of the XB and SB along the Heihe-Hegenshan Suture (HHS) occurred in the late Early Carboniferous-earlyLate Carboniferous (Liu et al., 2017). During the Late Mesozoic, thesubduction of the Pacific Plate resulted in the accretion of the JiamusiBlock and Nadanhada Terrane on the Eastern Asian margin, formingtwo lithospheric faults, the Yilan-Yitong and Dunhua-Mishan Faults(Wu et al., 2011; Xu et al., 2017).

The mega-rifting of the Songliao Basin, the main tectonic unitwithin the Songliao Block (SB), occurred from 150 to 105Ma, followedby anomalous subsidence between 110 and 80Ma, and then regionaluplift and basin inversion lasted until 64Ma (Li and Liu, 2015; Wanget al., 2016b). The basin is the only one that contains two Cretaceoussedimentary layers of ∼6 km thick in NE Asia. The lower layer wasdeposited during the Early Cretaceous (syn-rift stage), and the upperlayer was filled between the Early and Late Cretaceous (post-rift stage).After the mid-Danian (ca. 64Ma), the basin has neither undergoneprominent deposition nor deformation (Huang et al., 2011; Wang et al.,2016b). A large set of seismic reflection profiles reveals that the totalsedimentary thickness of the basin is up to∼8 km (Li and Liu, 2015). InNE China, the Songliao Basin lithosphere experienced the greatest ex-tension and thinning (e.g., Meng, 2003; Ren et al., 2002), to be as thinas ∼60–100 km (Wang et al., 2016b), resulting in a high geothermalgradient with a maximum of 62 °C/km (Tian et al., 1992). The Mohodepths range mostly between 29 and 34 km (Wang et al., 2016b) withtwo uplift regions in the northern part of the basin. The origin andmechanism of lithospheric thinning and basin subsidence are still underdebate, and can be attributed to (1) back-arc extension (Wang et al.,2002; Yan et al., 2002), (2) pull-apart (Ren et al., 1998; Wang et al.,1998), (3) basin and range system (Wang et al., 2007), (4) mantleplume (He and Santosh, 2016; Xiao et al., 2004), (5) lithospheric de-lamination (Wu et al., 2005), (6) decratonization (Yang et al., 2010),and (7) subduction-related asthenospheric upwelling (Wang et al.,2016b).

During the Late Mesozoic, intrusive rocks were widespread in NEChina with Jurassic granitoids mainly developed in the east andCretaceous granitoids in the west, manifesting a westward youngingtrend (Wu et al., 2011). The identified A-type granites were emplacedduring the Permian, Late Triassic to Early Jurassic, and Early Cretac-eous, and were closely related to post-collisional slab break-off, litho-spheric delamination and extension (Wu et al., 2002). Cenozoic in-traplate volcanic rocks occurred around the Songliao Basin are mainlycomposed of tholeiitic and alkali basalts with ages varying from ca.80Ma to 300 years ago (Chen et al., 2007). Most volcanoes were activeduring the last 20Ma, and some are still active with rather young agesless than thousands of years (e.g., Wudalianchi volcanic field) (Fig. 1).

3. Magnetic anomalies and gradient tensors

Lithospheric magnetization is the vectorial addition of induced andremanent components in the crust and uppermost mantle. Inducedmagnetization controlled by the magnitude and direction of the am-bient magnetic field is dominant within the continental crust (Counilet al., 1991; Hemant and Maus, 2005), whereas remanent magnetiza-tion dominates in the oceanic crust (Cohen and Achache, 1994; Dymentand Arkani-Hamed, 1998), and is difficult to estimate due to the com-plicated geologic, tectonic and thermal evolution. However, inducedmagnetization is generally considered to dominate over remanentcomponent, especially in plutonic and sedimentary rocks (e.g., Hinzeet al., 2012). Therefore, remanent magnetization is usually ignored inmodeling and inversion of magnetic data (Hemant and Mitchell, 2009;Kamm et al., 2015). Furthermore, Curie depth estimation does not

concern whether the magnetization is induced or remanent.We calculate the vector and gradient tensor components using a

harmonic geomagnetic model. These components represent directionalfilters and thus emphasize structures in particular orientations.Especially, the vertical vector and gradient tensor components corre-spond to anomalies reduced to the pole and help delineate steepboundaries (Munschy and Fleury, 2011; Schmidt and Clark, 2006).

The Earth’s internal geomagnetic field potential function V r θ λ( , , )is generally written in a spherical harmonic series (e.g., Blakely, 1996;Langel and Hinze, 1998)

∑ ∑= ⎛⎝

⎞⎠

+= =

+V r θ λ a a

rg mλ h mλ P θ( , , ) ( cos sin ) (cos )

n

N

m

n n

nm

nm

nm

1 0

1

(1)

where a is the Earth’s radius (6371.2 km), r is the geocentric co-ordinates, θ and λ are the colatitude and longitude, respectively. gn

m andhn

m are called the Gauss coefficients, and P θ(cos )nm is the Schmidt quasi-

normalized form of associated Legendre functions of degree n and orderm. Considering the vertical component Fz with Z downward along thelocal vertical, and the two horizontal components Fx and Fy with Xpositive toward the north and Y positive toward the east, the vectorcomponents of the geomagnetic field can be derived from

= = ∑ ∑ +

= − = ∑ ∑ −

= = − + ∑ ∑ +

∂∂

= =

+ ∂∂

∂∂

= =

+

∂∂

= =

+

( )( )( )

Fx g mλ h mλ

Fy g mλ h mλ

Fz n g mλ h mλ P θ

( cos sin )

( sin cos )

( 1) ( cos sin ) (cos )

.

rVθ

n

N

m

nar

nnm

nm P θ

θ

r θVλ

n

N

m

nar

nnm

nm mP θ

θ

Vr

n

N

m

nar

nnm

nm

nm

1

1 0

2 (cos )

1sin

1 0

2 (cos )sin

1 0

2

nm

nm

(2)

The total geomagnetic field can be calculated from= + +F Fx Fy Fz2 2 2 , and accordingly the lithospheric magnetic

anomalies can be obtained by subtracting the core field (to degree 15)(Maus et al., 2008) from the total magnetic field (degree of 720).

Then, the magnetic gradient tensor FGT components can be calcu-lated from the spatial rate of change of the three vector componentsFx Fy Fz( , , ) (Nelson, 1988)

=

⎡

⎣

⎢⎢⎢⎢

⎤

⎦

⎥⎥⎥⎥

=⎡

⎣

⎢⎢

⎤

⎦

⎥⎥

∂∂

∂∂

∂∂

∂∂

∂∂

∂∂

∂∂

∂∂

∂∂

FFxx Fyx FzxFxy Fyy FzyFxz Fyz Fzz

.GT

Fxx

Fyx

Fzx

Fxy

Fyy

Fzy

Fxz

Fyz

Fzz (3)

We compare the EMAG2 model with magnetic anomalies and gra-dient tensors calculated from the NGDC-720 model using the Eqs. (2)and (3). The EMAG2 model shows that the Mongol-Okhotsk SutureZone (MOSZ) is dominated by long wavelengths and large negativeamplitudes, which are similar to those found in the Himalaya area (Qiuet al., 2017). These strong negative anomalies may be originated fromaccretionary wedges, alteration and/or destruction of magnetic rocksdue to collision between different terranes. Regions to the northwest ofthe MOSZ are characterized by small wavelengths and medium-highpositive amplitudes induced by the post-collisional magmatism (Fig. 2).The Sikhote-Alin Accretionary Complex (SAAC) in the east of the Mu-danjiang-Yilan Suture shows a wide magnetic quiet zone. Some youngvolcanoes (< 3Ma) show weak or small negative magnetization due tothermal demagnetization (Bouligand et al., 2014; Wang and Li, 2015),for instance, the Wudalianchi and Abaga volcanic fields. The SongliaoBasin exhibits a background of weak magnetic magnetization due tothick sediments, and has some small areas with high anomalies(> 100 nT) along and around the Nenjiang (F2), Songliao Basin Central(F3) and Jianusi-Yilan (F4) faults, reflecting Cretaceous intrusion ofgranitoids and volcanic rocks (Fig. 1). Surrounding the basin, threepatches of high positive magnetic anomalies with amplitudes up to∼250 nT are observed (M1, M2 and M3 on Fig. 2). corresponding to theCretaceous granitoids (Fig. 1). On the total-field magnetic anomalies

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

179

calculated from the NGDC-720 model at the same datum of 4 km as theEMAG2 grid (Fig. 3). NGDC-720 exhibits less high-frequency compo-nents than EMAG2 (Fig. 2).

To suppress local signals from shallow sources and enhance regionaland/or deep seated magnetic features, we perform upward continuationof EMAG2 by 20 km after the reduction to the pole (Fig. 4a). and alsocalculate the total-field magnetic anomalies of the NGDC-720 model at20, 200 and 400 km altitudes (Fig. 4b–d). The upward continuationenhances the three prominent positive magnetic anomalies (M1, M2and M3). They start to join together when observed at 200 km altitude(Fig. 4c). At 400 km altitude, they show as a dome-like magnetic feature(Fig. 4d). indicating a possible common deep-seated source. Gradienttensors of the NGDC-720 model at the altitudes of 200 and 400 kmenhance edges of magnetic sources (Fig. 4e–f), and again show a unifiedmagnetic anomaly up to ∼500–600 km in radius observed at 400 kmaltitude (Fig. 4f).

4. Curie-point depths as thermal proxies

Crustal magnetism depends strongly on temperature distribution.Above the Curie temperature (∼550 °C but slightly changes with tita-nium content and pressure), minerals lose their ferromagnetism and areconsidered non-magnetic. Thus, Curie depths correspond to the basaldepths of magnetic layer and can be inverted from magnetic anomalies.

Curie temperature is usually taken as 580 °C for magnetite, which isconsidered to be the principal magnetic minerals in the continentalcrust (Frost and Shive, 1986). Curie temperature decreases as the tita-nium content increases in titanomagnetite, another magnetic mineralcontributing long wavelength magnetic anomalies, and can be as low as∼520 °C in the deep crust and uppermost mantle (Shuey et al., 1977).On the other hand, Curie temperature increases as pressure/depth in-creases, however, this effect on magnetite and titanomagnetite is small(Friedman et al., 2014). Consequently, Curie temperature is taken as550 °C in this study.

4.1. Methodology

Spectral peak method (Shuey et al., 1977; Spector and Grant, 1970)and the centroid method (Okubo et al., 1985; Tanaka et al., 1999) areamong the commonly used techniques for estimating Curie-pointdepths from magnetic data. Conventionally, both methods assumerandom magnetization (e.g., Blakely, 1996), although magnetization isnot completely uncorrelated but rather has a fractal distribution(Pilkington and Todoeschuck, 1993). In this study, we employ thecentroid approach with a fractal magnetization model (Li et al., 2013)to estimate the Curie depths in NE China. Assuming a 2D horizontalfractal magnetization but a constant vertical magnetization, the 1Dradial amplitude spectrum of magnetic anomalies can be written as(Blakely, 1996; Li et al., 2013)

= − −−

+ − − −

A k Z Z β C πk Zβ

πk

e

ln[ ( , , , )] |2 |1

2ln |2 |

ln[1 ],

T t b DP

tDP

πk Z Z

Δ 3 13

|2 | ( )b t (4)

where A TΔ is the radially averaged amplitude spectrum of the total-fieldmagnetic anomalies, = +k k kx y

2 2 is the wave number with kx andbeing the wave numbers in x and y directions, respectively. Zt and Zbare the top and bottom depths to the magnetic source, respectively. β D

P3

is the fractal exponent defined for power spectrum of 3D magnetizationand C1 is a constant. The Curie depths are estimated by the followingtwo steps in different wave number bands. The first step is the esti-mation of the top depth at middle- to high-wavenumber band using

× ≈ −−πk A k Z β C πk Zln[|2 | ( , , )] |2 | ,βT t D

Pt

( 1) 2Δ 3 2D

P3 (5)

and the second step is to estimate the centroid depth Z( )0 at low-wavenumber band using

× ≈ −−πk A k Z β πk C πk Zln[|2 | ( , , ) |2 |] |2 |βT D

P( 1) 2Δ 0 3 3 0D

P3 (6)

where C2 and C3 are constants. The Curie depths Z( )b are then obtainedby

= −Z Z Z2b t0 (7)

Fig. 2. Total-field magnetic anomalies of EMAG2 at an altitude of 4 km (Maus et al., 2009). M1, M2 and M3 represent three deep magnetic sources with high positiveamplitudes. For other notations refer to Fig. 1.

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

180

4.2. Numerical simulation test

Bouligand et al. (2009) and Li et al. (2013) tested numerically themethodology on synthetic magnetic anomalies simulated from differentmagnetization thicknesses with a 3D fractal exponent of 3.0 andwindow size of ×200 200 km2. To illustrate the effects of window sizesand observation elevations, we further test on synthetic magneticanomaly models calculated at various altitudes from a 3D fractalmagnetization layer of 30 km thick using different window sizes. Herewe follow the simulation procedure proposed by Li et al. (2013) andconstruct a fractal magnetization grid of × ×500 500 30 km3. The meanand standard deviation of the magnetization is set to be ±0 0.3 A/m.

For simplification, we calculate the magnetic anomalies observed atthe North Pole. Fig. 5a shows the simulated 3D fractal magnetizationmodel with depth to the top =Z 2tm km, depth to the bottom

=Z 32bm km and =β 3.0DP

3 . Magnetic anomalies are calculated at thealtitudes of 0, 4, 10, 20 and 30 km, respectively (Fig. 5b–f). Long wa-velength components increase as the observation altitudes increase.

We next calculate the wavenumber-scaled radially averaged spectraof the synthetic magnetic anomalies observed at various elevations andusing different window sizes (Fig. 6). We find that the simulated Curiedepths Zbs are very close to, but slightly smaller than, the Zbm of theinput model. This is caused by the finite size of the synthetic model weused here. The theoretical models are infinite in horizontal extensions(Blakely, 1996; Li et al., 2013). It can be seen that the spectra calculated

using larger windows ( ×200 200 km2 and ×300 300 km2) (Fig. 6b andc) tend to be more stable than using a small window ( ×100 100 km2)(Fig. 6a), and therefore can reduce the statistical errors of linear fitting.The peaks in Fig. 6a are induced by using a small window size( ×100 100 km2) that causes long-wavelength deficiencies. The esti-mated Curie depths increase only slightly as the window size increases,but the spatial resolution of the estimated Curie depths decrease ob-viously. This can be overcome by taking the average of Curie depthsestimated from the three window sizes (Table 1).

In our synthetic modeling, the wavenumber-scaled radially aver-aged spectra calculated at 20 km altitude are relatively stable with allthe three windows (Fig. 6). and the Curie depth estimated at this alti-tude is close to the model depth (Table 1). As described above, calcu-lated magnetic anomalies at various altitudes change less rapidly withincreasing observation elevation (Fig. 5). and above about 20 km alti-tude shallow sources and small wavelength magnetic signals are sup-pressed effectively. Hence, we can estimate Curie depths in NE Chinabased on the magnetic anomalies calculated from the NGDC-720 modelat the 20 km altitude (Fig. 4b).

4.3. Curie-point depths and uncertainties

Curie depth estimation depends strongly upon the wavenumberband selected for the straight-line fitting. The smallest wave numbers ofthe spectra contain the primarily information of the deepest sources and

Fig. 3. (a) Total-field magnetic anomalies of NGDC-720-V3.1 (Maus, 2010); (b)-(d) are the XΔ , YΔ and ZΔ components of the vector magnetic anomalies. Thealtitude of the magnetic anomalies for (a)-(d) is 4 km. For other notations refer to Figs. 1 and 2.

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

181

their bottoms, and most researches took ∼0.003–0.008 km−1 and∼0.03–0.05 km−1 as the lower and upper wave number limits for es-timating Z0, respectively (Table A1 in Appendix A). In practical im-plementations, however, the first few points may be smaller than thefollowing ones in spectral amplitudes, and if so, our algorithm canabnegate automatically the first few points (Fig. 6a) (Li et al., 2013,

2017). We test the fitting of Z0 from the NGDC-720 model observed atthe 20 km altitude, starting from the second points only (wave numberof ∼0.01–0.04 km−1) for all the moving window. The whole Zb valuesare too small and inconsistent with true geology (Fig. 7a), caused by theselection from the second points that rules out the information of thedeepest sources and results in large relative errors of Zb. In this study,

Fig. 4. (a) Upward continued anomalies of EMAG2 by 20 km after the reduction to the pole. Total-field magnetic anomalies of NGDC-720-V3.1 are calculated at thealtitudes of (b) 20 km; (c) 200 km and (d) 400 km. Magnetic gradient tensor Fzz components are calculated at the altitudes of (e) 200 km and (f) 400 km, respectively.For other notations refer to Figs. 1 and 2.

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

182

we calculate Zt and Z0 using the wave number bands of ca. 0.03 ∼0.08and ca. 0.003 ∼0.03 km−1, respectively.

Numerically, we estimate that the statistic errors of the averagedCurie depths range from ca. 1.3 to 4.1 km (Fig. 7b) (Appendix B), andmost of the relative errors is less than∼14% (Fig. 7c). The errors in thisstudy based on the NGDC-720 model are compatible with those

calculated from the EMM2010 (Arnaiz-Rodríguez and Orihuela, 2013)and EMAG2 (Speranza et al., 2016). Recently, Li et al. (2017) presenteda global model of Curie depths (GCDM) based on EMAG2, and we findthat 90% of the differences between the Curie depths in this study andthe GCDM model are less than about ± 6 km (Fig. 7d).

Another uncertainty in Curie depth estimation is the selection of the

Fig. 5. Synthetic modeling of 3D fractal magnetization and their magnetic anomalies calculated at various observation altitudes. (a) The magnetization has aGaussian distribution with a mean value of 0 A/m and a standard deviation of 0.3 A/m. The fractal exponent β( )D

P3 , the top Z( )tm and bottom Z( )bm depths to the

magnetic sources of the input model are taken as 3.0, 2 km and 32 km, respectively. All the magnetic anomalies are calculated at the North Pole with varying altitudesof (b) 0 km, (c) 4 km, (d) 10 km, (e) 20 km, and (f) 30 km. The black squares depict the three windows of ×100 100, ×200 200 and ×300 300 km2 used for Curie depthestimations.

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

183

fractal exponent. Assuming a random 2D magnetization model( =β 1.0D

P3 ) will result in a deeper bound on Curie depth estimation (Li

et al., 2012), and we find that with =β 1.0DP

3 the estimated Curie depthwill vary from ∼35 to 50 km in NE China. These values indicate thatalmost the uppermost 10 km of mantle in NE China is magnetized,which is rather unrealistic. Recently, Witter et al. (2018) verified fromthermal modeling that the Curie depths calculated by Li et al. (2017)with a 3D fractal exponent of 3.0 are more accurate for Yukon, Canada.Therefore, to achieve accurate and stable solution of Curie depths, wefix the fractal exponent β D

P3 at 3.0. Same as those used by Li et al.

(2017), we calculate Zb with three window sizes, i.e., ×98.8 98.8,×195 195, and ×296.4 296.4 km2. By averaging the results of the three

windows, we reduce the error and improve the resolution of Zb.Averaged Curie depths Z( )b of NE China range from ∼ ±13 2.0 to

±33 4.1 km, with a mean of ∼ ±22 2.4 km (Fig. 8). Large Zb are found inthe MOSZ and SAAC, which show long wavelengths of negative or weakmagnetic anomalies, originated from subduction-accretion and collisionbetween different blocks. Remarkably small Zb are found within theSongliao Basin, which is magnetically rather weak and quiet (Figs. 2–4).These patches of small Zb are not homogeneous everywhere within the

basin and exhibit a north-south elongation, implying a high averagecrustal thermal gradient in the basin. Zhao et al. (2015) calculated thethermal gradients at 3 km depth in the basin based on borehole ob-servations and reported a high thermal gradient of ca. 50 °C/km. Ye(2017) estimated the temperature at 10 km depth in the basin and alsorevealed a high thermal anomaly. These results are all consistent to oursmall Curie depths. The small Zb in the northern part of the basinroughly match the locations of Moho upwelling (Fig. 8). indicatingdynamic effects from the upper mantle. In the southeastern portion ofthe basin developed three Paleogene volcanoes exhibiting large Zb andweak magnetization. The three magnetic anomalies M1, M2 and M3around the basin show distinctly large Zb, corresponding to extensiveintrusions of Cretaceous granitoids. Some small areas with shallow Zb

along the DF (F1) and XXS (S1) correlate with the Pleistocene basaltsand volcanisms (Figs. 1 and 8). Other regions with Cenozoic basalts tothe east of the DMF (F5) and MYS (S4) also show small Zb. It should bepoint out that features of small Zb with horizontal scales less than∼50 km, like the N-S elongated features to the northeast of M1, may beartefact, since our smallest window size used in this work is∼ ×100 100 km2. Numerous Quaternary volcanoes, for instance, theregionally distributed active Wudalianchi volcanic field, show rela-tively small Zb (Fig. 8). However, some other isolated Quaternary vol-canoes, including the Longgang and Changbai volcanoes (Fig. 1) are notwell captured due to the low resolution of the data calculated at 20 kmaltitude.

5. Lithospheric thermal structures in NE China

5.1. Correlations between Curie depths and heat flow

We correlate our Curie depths with the sparsely and unevenly dis-tributed heat flow measurements. We compile recently published

Fig. 6. Curie depths estimated from the wavenumber-scaled radially averaged amplitude spectra of the synthetic magnetic anomalies according to Fig. 5. The inputCurie depth model is =Z 32bm km, and Zbs are the simulated Curie depths from the synthetic magnetic anomalies observed at the elevations of 0, 4, 10, 20 and 30 km,respectively. The window sizes used for Curie depth estimations are (a) ×100 100 km2, (b) ×200 200 km2, and (c) ×300 300 km2. The shadows show the wavenumberbands for linear fitting of Z0. Note that the first points of the spectra in (a) calculated from magnetic anomalies observed at altitudes of 0 and 4 km are abnegatedautomatically, and the linear fitting starts from the second points.

Table 1Curie depths estimation from the synthetic magnetic data (bottomdepth= 32 km) calculated at varying altitudes shown in Fig. 5.

Window size (km2) Altitude (km)

0 4 10 20 30

×100 100 30.79 28.78 27.61 32.29 31.31×200 200 30.74 30.85 32.23 28.56 20.68×300 300 31.68 32.41 35.14 34.79 29.79

Mean 31.07 30.68 31.66 31.88 27.26

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

184

surface heat flow in continental China (Hu et al., 2001; Jiang et al.,2016; Wang and Huang, 1990) and from the Global Heat Flow Database(http://www.heatflow.und.edu/data.html) for Russia and the JapanSea (Fig. 9a). We choose high quality heat flow data larger than20mW/m2 in order to remove local effects of hydrothermal circulationand erosion. We interpolate the original heat flow data by a 0.5°gridding interval using an algorithm of minimum curvature with atension factor of 0.5 (Smith and Wessel, 1990).

Despite data gaps, there are many similarities between Curie depths(Fig. 8). and heat flow (Fig. 9a). The highest surface heat flow regionsare found in the Japan Sea and the Songliao Basin, both of which showsmall Zb. The Neogene and Pleistocene ABV displays medium heat flowand small to medium Curie depths. Cenozoic basalts along the Dunhua-Mishan fault (F5) show high heat flow and small Zb, whereas the SAACexhibits low heat flow and large Zb. There are some poor correlations inthe JPV and CBV volcanic fields due partly to our limited resolution ofthe magnetic data.

Curie point is considered as an inherent physical property of mag-netic minerals. As descried afore, there is in general an inverse re-lationship between heat flow and Curie depths. Cenozoic volcanoesdeveloped in the study area are volumetrically small compared to large

igneous provinces (Fig. 1) (Chen et al., 2007). We can employ the 1Dsteady-state heat conduction equation (Li, 2011) to provide a first-orderconstraint on regional lithospheric geothermal field

= −−

+ −−

+− −

−q k T TZ Z

h H e eZ Z

h H esc s

b sr s

Z h Z h

b sr s

Z h2 b r s rs r

(8)

Here k is the bulk average vertical thermal conductivity for themagnetic layer, hr is the characteristic drop-off of heat production, Hs isthe heat production rate at the surface, and Ts and Tc are the tempera-tures at the surface elevation Zs and Curie depth, respectively.

A logical way of estimating the geothermal parameters is to performa nonlinear regression in the least squares sense on the data points ofheat flow vs. Curie depths. This method has been employed in the Sulu-Dabie orogen (Li et al., 2012) and the Antarctica (Martos et al., 2017).Assuming =T 5s °C and =T 550c °C, the optimal values in the wholestudy area for k, hr and Hs are 2.51W/(m °C), 14.84 km and 1.91 μW/m3, respectively. Given that the NE China block consists of several sub-blocks that welded together during the Phanerozoic, it is necessary tohandle the optimizations for different geologic units to produce morerealistic geothermal parameters (Fig. 9b–e and Table 2). In the

Fig. 7. (a) Averaged Curie depths estimated from NGDC-720-V3.1 observed at the 20 km altitude with a wave number band of ca. 0.01–0.04 km−1 for Z0 estimation.The window sizes and fractal exponent are applied in concert with those in mapping Fig. 8 (b) and (c) are the statistic errors and relative errors of the Curie depthsaccording to Fig. 8, respectively. (d) Differences between Curie depths in this study and the global Curie depths model GCDM (Li et al., 2017). The cold colors ofpositive values indicate that our Curie depths are larger than the GCDM model. For other notations refer to Figs. 1 and 2.

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

185

continental domain, the bulk average thermal conductivities for themagnetic layer vary from ∼2.1 to 2.4W/(m °C), in concert with thoseof granite and basalt (e.g., Turcotte and Schubert, 2002). Surface heatproductions range between ∼2.0 and 2.2 μW/m3, similarly to those ofthe global granitic rocks in all ages (Artemieva et al., 2017). Accordingto the exponentially decreasing model of the heat production (e.g.,Turcotte and Schubert, 2002), and the surface heat productions and thescale depths listed in Table 2, the heat productions in the lower crustare ∼0.22–0.34 μW/m3, in line with estimations by Hacker et al.(2015). The Japan Sea is floored with extended continental crust andoceanic crust, where heat conduction is more dominant over radio-active heat production. Our best optimal values in the Japan Sea(Table 2) give rise to the radioactive heat flow less than 5mW/m2.

5.2. Lithospheric thickness and Moho temperature

With the best fitting k for the magnetic layer, we can extrapolate atemperature-dependent thermal conductivity model for the lithosphere

= ⎧⎨⎩

+ ⩽ ⩽− >

k T za T z T z

b T z T z[ ( )]

[1 0.001 ( )] 0 ( ) 550[1 0.00025 ( )] ( ) 550 (9)

where a and b are constant coefficients for the magnetic crust and non-magnetic lithosphere, respectively. For example, =a 2.6 and =b 1.4 forthe Erguna Block (Table 2), and the conductivity decreases from ca. 2.6to 1.7W/(m °C) for the magnetic layer, and then increases to ∼2.1W/(m °C) at 1300 °C. The bulk average thermal conductivity estimatedfrom the least square fitting is consistent with this temperature-de-pendent thermal conductivity model. Then the 1D stable thermal

conduction equation can be rewritten as (Turcotte and Schubert, 2002)

⎛⎝

⎞⎠

= − −ddz

k T z dT zdz

H e[ ( )] ( )s

z hr(10)

in which z is the depth, Hs and hr are listed in Table 2. We then cal-culate the depths to the base of thermal lithosphere using the mantlesolidus (Lachenbruch, 1978; Pollack and Chapman, 1977)

≈ +T z1050 3LAB (11)

whereTLAB represents the temperature at the lithosphere-asthenosphereboundary (LAB).

Based on surface heat flow and Curie depths, the calculated thermallithospheric thicknesses in NE China range mostly between ca. 45 and90 km (Fig. 10). The depths to the asthenosphere estimated from themantle-derived Pliocene and Quaternary basaltic rocks in NE China aremainly approximately 50–85 km (Deng et al., 2004). Recently, Daltonet al. (2017) also reported a thin lithosphere (∼60–80 km) in NE Chinabased on Rayleigh wave attenuation, which is less sensitive to litho-spheric compositions.

We further build an integrated geothermal profile along theManzhouli-Suifenhe Global Geosciences Transect (GGT) (e.g., Wanget al., 2016b) (Figs. 10 and 11). The thermal lithospheric thicknessesvary from ∼45 to 90 km (Fig. 11d) consistent to those calculated byGornov et al. (2009). The thinnest lithosphere is immediately beneaththe central Songliao Basin with a thickness of ∼50 km, close to theresult revealed by the seismological survey along the GGT (Wang et al.,2016b). Using the above temperature-dependent thermal conductivitymodel, we can calculate the Moho temperature (Fig. 11d) at the Mohodepth (Fig. 8). The abnormally high Moho temperatures vary mainly

Fig. 8. Averaged Curie depths estimated from NGDC-720-V3.1 observed at the 20 km altitude using three window sizes of ×98.8 98.8, ×195 195and ×296.4 296.4 km2

with a constant 3D fractal exponent of 3.0. The black dashed lines are the Moho contours according to Wang et al. (2016b). The inset is the histogram of averagedCurie depths. For other notations refer to Figs. 1 and 2.

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

186

(caption on next page)

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

187

from ca. 700 to 1000 °C, with the highest values around 1000 °C be-neath the Songliao Basin, in accordance with geothermal modeling(Gornov et al., 2009) and pyroxene geotherms (Deng and Zhao, 1991).We also calculate the mantle heat flow across the Moho, ranging be-tween ∼40 and 50mW/m2 (Fig. 11c). These values imply that themantle beneath the Songliao Basin contributes about 65% of the totalsurface heat flow.

The regions with uplifted Curie depths in the Songliao Baisn roughlycoincide with low and negative magnetic amplitudes of NGDC-720

observed at 20 km (Fig. 11b). This Zb-magnetic anomaly pattern is alsoobserved in the Venezuela Basin (Arnaiz-Rodríguez and Orihuela,2013). Deep thermal perturbation shallows the Curie depth andtherefore weakens the rock magnetization. Based on the small Curiedepth, a very thin thermal lithosphere and high Moho temperatures andmantle heat flow are obtained beneath the Songliao Basin. Ambientnoise tomography revealed low Rayleigh phase and group velocityanomalies at depths of ∼20 km (Kim et al., 2016). Guo et al. (2018)also found low velocities in the central-north Songliao Basin at depth of25 km from the joint inversion of surface and body waves. These lowvelocities may imply partial melting in the lower crust, consistent withour estimated small Curie depths of ∼15–17 km and the high Mohotemperatures of ∼700–1000 °C in the basin. Guo et al. (2014) found athin lithosphere of ∼70–90 km in NE China from S-wave receiverfunctions. These values are slightly larger than our thermal modelingresults, but reveal a similar variation pattern. The thickness discrepancyis expected, because the thermal LAB constraints the top, whereas theseismic LAB constraints the bottom, of the transitional layer in themantle (Artemieva, 2011). Kang et al. (2016) found a vertically arrayed‘fast-slow-fast’ S-wave anomaly pattern between 40 and 140 km be-neath the Songliao Basin. The lower fast velocity layers extended from∼80 to 140 km in all three vertical transects across the basin may re-flect downwelling by lithospheric delamination, whereas the low ve-locity layers distributed between ∼60 and 80 km correspond to the

Fig. 9. (a) Surface heat flow map of Northeast China with a 0.5° gridding interval. Only heat flow measurements larger than 20 mW/m2 are considered as primarilyfrom thermal conduction. The red points are the total number of 352 original heat flow measurements complied from continental China (Hu et al., 2001; Jiang et al.,2016; Wang and Huang, 1990) and from Global Heat Flow Database (http://www.heatflow.und.edu/data.html). Correlations between surface heat flow and Curiedepths with reference to the 1D steady state heat conduction model for different geological units: (b) Erguna Block, (c) Xing’an and Songliao Block, (d) Songliao Basinand (e) Japan Sea. The green curves are the best least-square fitting models. The dashed blue curves are the theoretical conductivities ranging between 1 and 4W/(m °C). The red points are the original heat flow measurements vs. Curie depths. For other notations refer to Fig. 1. (For interpretation of the references to colour inthis figure legend, the reader is referred to the web version of this article.)

Table 2Least square optimization of geothermal parameters for different geologicalblocks in NE China.

Fig. 10. Estimated thermal lithospheric thickness of Northeast China based on the geothermal parameters listed in Table 2. The white line is the Manzhouli-SuifenheGlobal Geosciences Transect (GGT, Wang et al., 2016b). For other notations refer to Figs. 1 and 2.

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

188

asthenospheric upwelling triggered by lithospheric delamination. Therelict thin lithospheric lid (upper fast velocity layers) immediatelybelow the Moho with a thickness of ∼20 km (Kang et al., 2016) in theSongliao Basin, Xing’an and Erguna blocks is also in line with our re-sults (Fig. 11d). Accordingly, our Curie depth estimation and geo-thermal modeling support asthenospheric upwelling triggered by li-thospheric delamination in the Songliao Basin (Kang et al., 2016).

6. Conclusions

In this study, numerical simulation tests on the magnetic anomaliescalculated at 20 km altitude from a 3D fractal magnetization modeldemonstrate that the wavenumber-scaled radially averaged spectra arerelatively stable using three different window sizes ( ×100 100,

×200 200 and ×300 300 km2, respectively), and lead to an invertedCurie depth close to the model depth. Therefore, we estimate the NEChina Curie depths from the NGDC-720 model at the 20 km altitude.Small Zb of ca. 15–17 km are found in the Songliao Basin, especially inits northern part.

By integrating Curie depths and surface heat flow, we estimate the

thermal lithospheric thickness to be ∼45–90 km in NE China based on1D steady-state thermal conduction equation. Constrained by Mohodepths, the high Moho temperatures vary from ca. 700 to 1000 °C andthe mantle contributes about 65% of the total surface heat flow in theSongliao Basin. The small Curie and thermal LAB depths and high Mohotemperature and mantle heat flow together support hot upper mantleupwelling triggered by lithospheric delamination beneath the SongliaoBasin, resulting in abnormal high heat flow observed at the surface.

Acknowledgements

We would like to thank the Editor Dapeng Zhao, and FabioSperanza, Mariano S. Arnaiz-Rodríguez and an anonymous reviewer fortheir insightful reviews and constructive comments. Data processingand mapping are supported by the USGS potential field software(Cordell et al., 1993; Phillips, 1997), and by the Generic Mapping Tools(GMT) software (Wessel et al., 2013). This research is funded by theNational Science Foundation of China (Grant Nos. 41704086 and41776057), and by the Institute of Crustal Dynamics, China EarthquakeAdministration (Grant No. ZDJ2016-01).

Fig. 11. Geothermal profiles along the Manzhouli-Suifenhe Global Geosciences Transect (Wang et al., 2016b). (a) Topography and tectonic units. (b) Magneticanomalies of EMAG2 at the 4 km altitude (in red) and NGDC-720 at the 20 km altitude (in blue). (c) Measured surface heat flow (blue points) within a 0.5° intervalalong the transect, and calculated mantle heat flow (red squares). (d) Thick solid red and dashed blue curves show the Curie and thermal LAB depths, and coloredcurve shows the Moho temperature. Thin solid red and dashed blue curves are the error bounds of the Curie and thermal LAB depths, respectively. (For interpretationof the references to colour in this figure legend, the reader is referred to the web version of this article.)

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

189

Appendix

See Table A1.

Appendix B

Errors in Curie depth estimation

The statistical error is calculated for each moving window from the standard deviation σ( ) between the spectrum and the linear fit at the wavenumber bands k(Δ ) of 0.005–0.03 km−1 and 0.03–0.7 km−1 for Z0 and Zt, respectively. We use the error ε( ) definition suggested by Okubo andMatsunaga (1994)

=ε σπ k2 Δ (B1)

Numerically, we can calculate the errors of the averaged Curie depths estimated from three different windows in this study by the followingexpression

∑= +=N

ε εE 1 [(2 ) ( ) ]i

Ni

ti

10

2 2

(B2)

where =N 3, and ε0 and εt are the errors of Z0 and Zt, respectively.

Table A1Summary of previous Curie depth estimations using the centroid method.

Region Data Height (km) Zt band (km−1) Z0 band (km−1) (km) Reference

East and Southeast Asia CCOPa 0 ∼0.05–0.1 ∼0.005–0.03 9–46 Tanaka et al. (1999)Western Argentina Ground data 0 ∼0.02–0.06 ∼0.005–0.02 20–40 Ruiz and Introcaso (2004)Central-Southern Europe Aeromagnetic data 3 0.05–0.1 0.005–0.05 17–33 Chiozzi et al. (2005)Eastern China CCOP 0 ∼0.05–0.1 ∼0.005–0.05 ∼20–49 Li et al. (2009)Bulgaria Ground data 3 0.05–0.1 0.005–0.045 14–35 Trifonova et al. (2009)South China Sea CCOP 0 ∼0.05–0.1 ∼0.005–0.05 ∼14–48 Li et al. (2010)Central Turkey Aeromagnetic data 0.6 ∼0.05–0.8 ∼0.05–0.25 13.7 Maden (2010)Sinai Peninsula Ground data 0 ∼0.064–0.167* ∼0.008–0.056* 15–25 Aboud et al. (2011)Germany Aeromagnetic data 1 ∼0.008–0.08* ∼0.002–0.035* 13–45 Bansal et al. (2011)Nankai subduction zone CCOP 0 ∼0.05–0.1 ∼0.005–0.05 ∼17–51 Li (2011)Mexican subduction zone NAMAGb 0.305 ∼0.095–0.14* ∼0.008–0.032* 9–37 Manea and Manea (2011)Central India Aeromagnetic data 1.5 ∼0.002–0.1* ∼0.002–0.02* 22–43 Bansal et al. (2013)North Atlantic EMAG3c & EMAG2d 5 & 4 ∼0.03–0.08 ∼0.005–0.03 ∼5–45 Li et al. (2013)Taiwan Aeromagnetic & marine data Unknown ∼0.04–0.07 ∼0.005–0.03 6–17 Hsieh et al. (2014)Southeast Tibet POMME-6.2e 15 ∼0.04–0.08* ∼0.006–0.07* 20–34 Gao et al. (2015a)Tarim Basin NGDC-720-V3f Unknown 0.05–0.1 0.005–0.05 28–48 Gao et al. (2015b)Western North America NAMAG 0.305 ∼0.03–0.08 ∼0.005–0.03 5.4–42 Wang and Li (2015)Arabian shield Aeromagnetic data 0.8 ∼0.003–0.036 ∼0.003–0.027 ∼17–57 Aboud et al. (2016)Western Turkey Aeromagnetic data 0.6 ∼0.01–0.025* ∼0.003–0.01* ∼6–12 Bilim et al. (2016)Eastern Brazil Aeromagnetic data 1 ∼0.006–0.01*(?) ∼0.0005–0.006*(?) 18–59 Correa et al. (2016)SW Indian Ocean EMAG2 4 ∼0.098–0.286 ∼0.008–0.092 10–30 Gailler et al. (2016)E & SE Asia EMAG2 & CCOP 4 ∼0.03–0.08 ∼0.005–0.03 ∼5–45 Li and Wang (2016)South California Aeromagnetic data 0.305 ∼0.01–0.02* ∼0.003–0.01* 5–23 Mickus and Hussein (2016)Western Desert Aeromagnetic data 7 ∼0.008–0.02*(?) ∼0.0008–0.007*(?) ∼25–33 Saada (2016)Alps & Po Plain ENIg & EMAG2 2.5 & 4 ∼0.04–0.16 ∼0.01–0.07 ∼6–40 Speranza et al. (2016)Southern Turkey MTAh 0.6 ∼0.011–0.025*(?) ∼0.003–0.008*(?) ∼7–14 Bilim et al. (2017)Lesser Antilles Arc Aeromagnetics & EAMG2 4 ∼0.096–0.178 ∼0.012–0.088 18–32 Gailler et al. (2017)West Himalayan syntaxis NGDC-720-V3 20 ∼0.04–0.1* ∼0.016–0.04* 20–44 Gao et al. (2017)Global EMAG2 4 ∼0.03–0.08 ∼0.005–0.03 ∼5–45 Li et al. (2017)SW Caribbean WDMAM V2.0i 5 ∼0.07–0.18* ∼0.003–0.04 ∼6–56 Salazar et al. (2017)Pacific EMAG2 4 ∼0.03–0.08 ∼0.003–0.03 ∼5–40 Li and Wang (2018)

a Geological Survey of Japan and Coordinating Committee for Coastal and Offshore Geoscience Programs in East and Southeast Asia (1996).b Magnetic Anomaly Map of North America (http://mrdata.usgs.gov/magnetic/).c World Digital Magnetic Anomaly Map of 3-arc-minute resolution (http://geomag.org/) (Maus et al., 2007).d World Digital Magnetic Anomaly Map of 2-arc-minute resolution (http://geomag.org/) (Maus et al., 2009).e Field model from CHAMP with MF6 and NGDC-720 (http://geomag.org/).f Field model combined from MF6 and EMAG2 (http://geomag.org/) (Maus, 2010).g Italian aeromagnetic map by the national oil company AGIP (now Eni).h General Directorate of Mineral Research and Exploration.i The second version of the World Digital Magnetic Anomaly Map (http://wdmam.org/) (Lesur et al., 2016).* Converted from rad/km.

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

190

References

Aboud, E., Alotaibi, A.M., Saud, R., 2016. Relationship between Curie isotherm surfaceand Moho discontinuity in the Arabian shield, Saudi Arabia. J. Asian Earth Sci. 128,42–53.

Aboud, E., Salem, A., Mekkawi, M., 2011. Curie depth map for Sinai Peninsula, Egyptdeduced from the analysis of magnetic data. Tectonophysics 506, 46–54.

An, M.J., Shi, Y.L., 2006. Lithospheric thickness of the Chinese continent. Phys. EarthPlanet. In. 159, 257–266.

Arnaiz-Rodríguez, M.S., Orihuela, N., 2013. Curie point depth in Venezuela and theEastern Caribbean. Tectonophysics 590, 38–51.

Artemieva, I.M., 2011. The lithosphere: An interdisciplinary approach. CambridgeUniversity Press.

Artemieva, I.M., Thybo, H., Jakobsen, K., Sorensen, N.K., Nielsen, L.S.K., 2017. Heatproduction in granitic rocks: Global analysis based on a new data compilationGRANITE2017. Earth.-Sci. Rev. 172, 1–26.

Bansal, A.R., Anand, S.P., Rajaram, M., Rao, V.K., Dimri, V.P., 2013. Depth to the bottomof magnetic sources (DBMS) from aeromagnetic data of Central India using modifiedcentroid method for fractal distribution of sources. Tectonophysics 603, 155–161.

Bansal, A.R., Gabriel, G., Dimri, V.P., Krawczyk, C.M., 2011. Estimation of depth to thebottom of magnetic sources by a modified centroid method for fractal distribution ofsources: An application to aeromagnetic data in Germany. Geophysics 76, L11–L22.

Bilim, F., Akay, T., Aydemir, A., Kosaroglu, S., 2016. Curie point depth, heat-flow andradiogenic heat production deduced from the spectral analysis of the aeromagneticdata for geothermal investigation on the Menderes Massif and the Aegean Region,western Turkey. Geothermics 60, 44–57.

Bilim, F., Aydemir, A., Ates, A., 2017. Tectonics and thermal structure in the Gulf ofIskenderun (southern Turkey) from the aeromagnetic, borehole and seismic data.Geothermics 70, 206–221.

Blakely, R.J., 1996. Potential Theory in Gravity and Magnetic Applications. CambridgeUniversity Press.

Bouligand, C., Glen, J.M.G., Blakely, R.J., 2009. Mapping Curie temperature depth in thewestern United States with a fractal model for crustal magnetization. J. Geophys.Res.-Sol Ea 114.

Bouligand, C., Glen, J.M.G., Blakely, R.J., 2014. Distribution of buried hydrothermalalteration deduced from high-resolution magnetic surveys in Yellowstone NationalPark. J. Geophys. Res.-Sol. Ea 119, 2595–2630.

Chen, Y., Zhang, Y.X., Graham, D., Su, S.G., Deng, J.F., 2007. Geochemistry of Cenozoicbasalts and mantle xenoliths in Northeast China. Lithos 96, 108–126.

Chiozzi, P., Matsushima, J., Okubo, Y., Pasquale, V., Verdoya, M., 2005. Curie-pointdepth from spectral analysis of magnetic data in central-southern Europe. Phys. EarthPlanet. In. 152, 267–276.

Cohen, Y., Achache, J., 1994. Contribution of induced and remanent magnetization tolong-wavelength oceanic magnetic anomalies. J. Geophys. Res.-Sol. Ea 99,2943–2954.

Cordell, L., Phillips, J.D., Godson, R.H., 1993. USGS potential-field geophysical softwarefor PC and compatible microcomputers. Lead. Edge 12, 290.

Correa, R.T., Vidotti, R.M., Oksum, E., 2016. Curie surface of Borborema Province, Brazil.Tectonophysics 679, 73–87.

Counil, J.L., Cohen, Y., Achache, J., 1991. The global continent-ocean magnetizationcontrast. Earth Planet. Sci. Lett. 103, 354–364.

Dalton, C.A., Bao, X.Y., Ma, Z.T., 2017. The thermal structure of cratonic lithosphere fromglobal Rayleigh wave attenuation. Earth Planet. Sci. Lett. 457, 250–262.

Deng, J.F., Mo, X.X., Zhao, H.L., Wu, Z.X., Luo, Z.H., Su, S.G., 2004. A new model for thedynamic evolution of Chinese lithosphere: 'continental roots-plume tectonics'. Earth-Sci. Rev. 65, 223–275.

Deng, J.F., Zhao, H.L., 1991. The thermal structure of the upper mantle in Eastern China -Inferred from the petrological mode. Acta Geol. Sin.-Engl. 4, 195–201.

Deng, Y.F., Tesauro, M., 2016. Lithospheric strength variations in Mainland China:Tectonic implications. Tectonics 35, 2313–2333.

Dyment, J., Arkani-Hamed, J., 1998. Contribution of lithospheric remanent magnetiza-tion to satellite magnetic anomalies over the world's oceans. J. Geophys. Res.-Sol. Ea103, 15423–15441.

Fowler, C.M.R., 2005. The Solid Earth: An Introduction to Global Geophysics, second ed.Cambridge University Press.

Friedman, S.A., Feinberg, J.M., Ferré, E.C., Demory, F., Martín-Hernández, F., Conder,J.A., Rochette, P., 2014. Craton vs. rift uppermost mantle contributions to magneticanomalies in the United States interior. Tectonophysics 624–625 15–23.

Frost, B.R., Shive, P.N., 1986. Magnetic mineralogy of the lower continental crust. J.Geophys. Res. Solid Earth 91, 6513–6521.

Gailler, L., Arcay, D., Münch, P., Martelet, G., Thinon, I., Lebrun, J.-F., 2017. Forearcstructure in the Lesser Antilles inferred from depth to the Curie temperature andthermo-mechanical simulations. Tectonophysics 706–707, 71–90.

Gailler, L.S., Lenat, J.F., Blakely, R.J., 2016. Depth to Curie temperature or bottom of themagnetic sources in the volcanic zone of la Reunion hot spot. J. Volcanol. Geoth. Res.324, 169–178.

Gao, G., Kang, G., Li, G., Bai, C., Wu, Y., 2017. An analysis of crustal magnetic anomalyand Curie surface in west Himalayan syntaxis and adjacent area. Acta Geod. Geophys.52, 407–420.

Gao, G.M., Kang, G.F., Bai, C.H., Li, G.Q., 2013. Distribution of the crustal magneticanomaly and geological structure in Xinjiang, China. J. Asian Earth Sci. 77, 12–20.

Gao, G.M., Kang, G.F., Bai, C.H., Wen, L.M., 2015a. Study on crustal magnetic anomaliesand Curie surface in Southeast Tibet. J. Asian Earth Sci. 97, 169–177.

Gao, G.M., Kang, G.F., Li, G.Q., Bai, C.H., 2015b. Crustal magnetic anomaly and Curie surfacebeneath Tarim Basin, China, and its adjacent area. Can. J. Earth Sci. 52, 357–367.

Gao, G.M., Kang, G.F., Li, G.Q., Bai, C.H., 2015c. Crustal magnetic anomaly in the Ordosregion and its tectonic implications. J. Asian Earth Sci. 109, 63–73.

Gornov, P.Y., Goroshko, M.V., Malyshev, Y.F., Podgornyi, V.Y., 2009. Thermal structureof lithosphere in Central Asian and Pacific belts and their adjacent cratons, from dataof geoscience transects. Russ Geol Geophys+ 50, 485–499.

Guo, Z., Cao, Y., Wang, X., John Chen, Y., Ning, J., He, W., Tang, Y., Feng, Y., 2014. Crustand upper mantle structures beneath Northeast China from receiver function studies.Earthquake Sci. 27, 265–275.

Guo, Z., Chen, Y.J., Ning, J.Y., Yang, Y.J., Afonso, J.C., Tang, Y.C., 2016. Seismic evi-dence of on-going sublithosphere upper mantle convection for intra-plate volcanismin Northeast China. Earth Planet. Sci. Lett. 433, 31–43.

Guo, Z., Wang, K., Yang, Y., Tang, Y., Chen, Y.J., Hung, S.H., 2018. The origin and mantledynamics of quaternary intraplate volcanism in Northeast China from joint inversionof surface wave and body wave. J. Geophys. Res. Solid Earth 123, 2410–2425.

Hacker, B.R., Kelemen, P.B., Behn, M.D., 2015. Continental lower crust. In: Jeanloz, R.,Freeman, K.H. (Eds.), Annual Review of Earth and Planetary Sciences, Vol. 43.Annual Reviews, Palo Alto, pp. 167-+.

He, C.S., Santosh, M., 2016. Seismic tomographic evidence for upwelling mantle plume inNE China. Phys. Earth Planet. In. 254, 37–45.

Hemant, K., Maus, S., 2005. Why no anomaly is visible over most of the continent-oceanboundary in the global crustal magnetic field. Phys. Earth Planet. In. 149, 321–333.

Hemant, K., Mitchell, A., 2009. Magnetic field modelling and interpretation of theHimalayan-Tibetan Plateau and adjoining north Indian Plains. Tectonophysics 478,87–99.

Hinze, W.J., Von Frese, R.R.B., Saad, A.H., 2012. Gravity and Magnetic Exploration:Principles, Practices and Applications. Cambridge University Press, UK.

Hou, Q.J., Feng, Z.Q., Feng, Z.H., 2009. Continental petroleum geology of Songliao Basin(in Chinese). Petroleum Industry Press, Beijing.

Hsieh, H.H., Chen, C.H., Lin, P.Y., Yen, H.Y., 2014. Curie point depth from spectralanalysis of magnetic data in Taiwan. J. Asian Earth Sci. 90, 26–33.

Hu, S., He, L., Wang, J., 2001. Compilation of heat flow data in the continental area ofChina (3rd edition) (in Chinese with English abstract). Chin. J. Geophys. 44,611–626.

Hu, X.-Z., Xu, M.-J., Xie, X.-A., Wang, L.-S., Zhang, Q.-L., Liu, S.-W., Xie, G.-A., Feng, C.-G., 2006. A Characteristic Analysis of Aeromagnetic Anomalies and Curie PointIsotherms in Northeast China. Chin. J. Geophys. 49, 1533–1542.

Huang, Q.H., Wu, H.C., Wan, X.Q., He, H.Y., Deng, C.L., 2011. New progress of integratedchronostratigraphy of the cretaceous in Songliao Basin (in Chinese with English ab-stract). J. Stratigraphy 35, 250–257.

Jiang, G., Gao, P., Rao, S., Zhang, L., Tang, X., Huang, F., Zhao, P., Pang, Z., He, L., Hu, S.,Wang, J., 2016. Compilation of heat flow data in the continental area of China (4thedition) (in Chinese with English abstract). Chin. J. Geophys. 59, 2892–2910.

Kamm, J., Lundin, I.A., Bastani, M., Sadeghi, M., Pedersen, L.B., 2015. Joint inversion ofgravity, magnetic, and petrophysical data - A case study from a gabbro intrusion inBoden, Sweden. Geophysics 80, B131–B152.

Kang, D., Shen, W.S., Ning, J.Y., Ritzwoller, M.H., 2016. Seismic evidence for lithosphericmodification associated with intracontinental volcanism in Northeastern China.Geophys. J. Int. 204, 215–235.

Kim, S., Tkalcic, H., Rhie, J., Chen, Y.L., 2016. Intraplate volcanism controlled by back-arc and continental structures in NE Asia inferred from transdimensional Bayesianambient noise tomography. Geophys. Res. Lett. 43, 8390–8398.

Kravchinsky, V.A., Cogne, J.P., Harbert, W.P., Kuzmin, M.I., 2002. Evolution of theMongol-Okhotsk Ocean as constrained by new palaeomagnetic data from the Mongol-Okhotsk suture zone, Siberia. Geophys. J. Int. 148, 34–57.

Lachenbruch, A., 1978. Heat flow in the Basin and Range province and thermal effects oftectonic extension. Pure Appl. Geophys. 117, 34–50.

Langel, R.A., Hinze, W.J., 1998. The Magnetic Field of the Earth's Lithosphere: TheSatellite Perspective. Cambridge University Press.

Lesur, V., Hamoudi, M., Choi, Y., Dyment, J., Thebault, E., 2016. Building the secondversion of the World Digital Magnetic Anomaly Map (WDMAM). Earth Planets Space68, 13.

Li, C., Liu, S.F., 2015. Cretaceous anomalous subsidence and its response to dynamictopography in the Songliao Basin, Northeast China. J Asian Earth Sci 109, 86–99.

Li, C.B., Wang, L.S., Sun, B., Feng, R.H., Wu, Y.J., 2015. Interpretations of gravity andmagnetic anomalies in the Songliao Basin with Wavelet Multi-scale Decomposition.Front Earth Sci.-Prc. 9, 427–436.

Li, C.F., 2011. An integrated geodynamic model of the Nankai subduction zone andneighboring regions from geophysical inversion and modeling. J. Geodyn. 51, 64–80.

Li, C.F., Chen, B., Zhou, Z.Y., 2009. Deep crustal structures of eastern China and adjacentseas revealed by magnetic data. Sci. China Ser. D 52, 984–993.

Li, C.F., Lu, Y., Wang, J., 2017. A global reference model of Curie-point depths based onEMAG2. Sci. Rep. 7, 9.

Li, C.F., Shi, X.B., Zhou, Z.Y., Li, J.B., Geng, J.H., Chen, B., 2017. Depths to the magneticlayer bottom in the South China Sea area and their tectonic implications. Geophys. J.Int. 182, 1229–1247.

Li, C.F., Wang, J., 2016. Variations in Moho and Curie depths and heat flow in Easternand Southeastern Asia. Mar. Geophys. Res. 37, 1–20.

Li, C.F., Wang, J., 2018. Thermal structures of the Pacific lithosphere from magneticanomaly inversion. Earth Planet. Phys. 2, 52–66.

Li, C.F., Wang, J., Lin, J., Wang, T.T., 2013. Thermal evolution of the North Atlanticlithosphere: New constraints from magnetic anomaly inversion with a fractal mag-netization model. Geochem. Geophys. Geosyst. 14, 5078–5105.

Li, C.F., Wang, J.L., Zhou, Z.Y., Geng, J.H., Chen, B., Yang, F.L., Wu, J.S., Yu, P., Zhang,X.B., Zhang, S.W., 2012. 3D geophysical characterization of the Sulu-Dabie orogenand its environs. Phys. Earth Planet. In. 192, 35–53.

Li, T., 2010. The principal characteristics of the lithosphere of China. Geosci. Front. 1,

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

191

45–56.Liu, Y.G., 1982. Relation between characteristic of geothermal field field and hydro-

carbon occurrence in Songliao Basin (in Chinese with English abstract). Pet. Explor.Dev. 26–31.

Liu, Y.J., Li, W.M., Feng, Z.Q., Wen, Q.B., Neubauer, F., Liang, C.Y., 2017. A review of thepaleozoic tectonics in the eastern part of central asian orogenic belt. Gondwana Res.43, 123–148.

Maden, N., 2010. Curie-point depth from spectral analysis of magnetic data in erciyesstratovolcano (central Turkey). Pure Appl. Geophys. 167, 349–358.

Manea, M., Manea, V.C., 2011. Curie point depth estimates and correlation with sub-duction in Mexico. Pure Appl. Geophys. 168, 1489–1499.

Martos, Y.M., Catalan, M., Jordan, T.A., Golynsky, A., Golynsky, D., Eagles, G., Vaughan,D.G., 2017. Heat flux distribution of Antarctica unveiled. Geophys. Res. Lett. 44,11417–11426.

Maule, C.F., Purucker, M.E., Olsen, N., Mosegaard, K., 2005. Heat flux anomalies inAntarctica revealed by satellite magnetic data. Science 309, 464–467.

Maus, S., 2010. An ellipsoidal harmonic representation of Earth's lithospheric magneticfield to degree and order 720. Geochem. Geophys. Geosyst. 11.

Maus, S., Barckhausen, U., Berkenbosch, H., Bournas, N., Brozena, J., Childers, V.,Dostaler, F., Fairhead, J.D., Finn, C., von Frese, R.R.B., Gaina, C., Golynsky, S., Kucks,R., Luhr, H., Milligan, P., Mogren, S., Müller, R.D., Olesen, O., Pilkington, M., Saltus,R., Schreckenberger, B., Thebault, E., Tontini, F.C., 2009. EMAG2: A 2-arc min re-solution Earth Magnetic Anomaly Grid compiled from satellite, airborne, and marinemagnetic measurements. Geochem. Geophys. Geosyst. 10.

Maus, S., Sazonova, T., Hemant, K., Fairhead, J.D., Ravat, D., 2007. National geophysicaldata center candidate for the world digital magnetic anomaly map. Geochem.Geophys. Geosyst. 8, 10.

Maus, S., Yin, F., Luhr, H., Manoj, C., Rother, M., Rauberg, J., Michaelis, I., Stolle, C.,Muller, R.D., 2008. Resolution of direction of oceanic magnetic lineations by thesixth-generation lithospheric magnetic field model from CHAMP satellite magneticmeasurements. Geochem. Geophys. Geosyst. 9, 10.

Meng, Q.R., 2003. What drove late Mesozoic extension of the northern China-Mongoliatract? Tectonophysics 369, 155–174.

Metelkin, D.V., Vernikovsky, V.A., Kazansky, A.Y., Wingate, M.T.D., 2010. Late Mesozoictectonics of Central Asia based on paleomagnetic evidence. Gondwana Res. 18,400–419.

Mickus, K., Hussein, M., 2016. Curie depth analysis of the salton Sea Region, SouthernCalifornia. Pure Appl. Geophys. 173, 537–554.

Munschy, M., Fleury, S., 2011. Scalar, vector, tensor magnetic anomalies: measurementor computation? Geophys. Prospect. 59, 1035–1045.

Nelson, J.B., 1988. Calculation of the magnetic gradient tensor from total field gradientmeasurements and its application to geophysical interpretation. Geophysics 53,957–966.

Okubo, Y., Graf, R., Hansen, R., Ogawa, K., Tsu, H., 1985. Curie point depths of the Islandof Kyushu and surrounding areas, Japan. Geophysics 50, 481–494.

Okubo, Y., Matsunaga, T., 1994. Curie point depth in northeast Japan and its correlationwith regional thermal structure and seismicity. J. Geophys. Res. Solid Earth 99,22363–22371.

Parker, R.L., 1973. The rapid calculation of potential anomalies. Geophys. J. R. Astron.Soc. 31, 447–455.

Phillips, J.D., 1997. Potential-field geophysical software for the PC, version 2.2. U.S.Geological Survey, Open-File, Report, 97-725.

Pilkington, M., Todoeschuck, J.P., 1993. Fractal magnetization of continental crust.Geophys. Res. Lett. 20, 627–630.

Pollack, H.N., Chapman, D.S., 1977. On the regional variation of heat flow, geotherms,and lithospheric thickness. Tectonophysics 38, 279–296.

Qiu, Y.D., Wang, Z.T., Jiang, W.P., Zhang, B.B., Li, F.P., Guo, F., 2017. Combining CHAMPand swarm satellite data to invert the lithospheric magnetic field in the TibetanPlateau. Sensors 17, 17.

Rajaram, M., Anand, S.P., Hemant, K., Purucker, M.E., 2009. Curie isotherm map ofIndian subcontinent from satellite and aeromagnetic data. Earth Planet. Sci. Lett. 281,147–158.

Ren, J.Y., Hu, X.Y., Zhang, J.X., 1998. The Late Mesozoic tectonic activation in theeastern Chinese continental and its evolution process (in Chinese with English ab-stract). Geotectonica Et Metallogenia 22, 51–60.

Ren, J.Y., Tamaki, K., Li, S.T., Junxia, Z., 2002. Late Mesozoic and Cenozoic rifting and itsdynamic setting in Eastern China and adjacent areas. Tectonophysics 344, 175–205.

Ross, H.E., Blakely, R.J., Zoback, M.D., 2006. Testing the use of aeromagnetic data for thedetermination of Curie depth in California. Geophysics 71, L51–L59.

Ruiz, F., Introcaso, A., 2004. Curie point depths beneath Precordillera cuyana and sierrasPampeanas obtained from spectral analysis of magnetic anomalies. Gondwana Res. 7,1133–1142.

Saada, S.A., 2016. Curie point depth and heat flow from spectral analysis of aeromagneticdata over the northern part of Western Desert, Egypt. J. Appl. Geophys. 134,100–111.

Salazar, J.M., Vargas, C.A., Leon, H., 2017. Curie point depth in the SW Caribbean usingthe radially averaged spectra of magnetic anomalies. Tectonophysics 694, 400–413.

Schmidt, P.W., Clark, D.A., 2006. The magnetic gradient tensor: Its properties and uses insource characterization. Lead. Edge 25, 75–78.

Shuey, R.T., Schellinger, D.K., Tripp, A.C., Al, L.B., 1977. Curie depth determination fromaeromagnetic spectra. Geophys. J. R. Astron. Soc. 50, 75–101.

Siler, D.L., Kennedy, B.M., 2016. Regional crustal-scale structures as conduits for deepgeothermal upflow. Geothermics 59, 27–37.

Smith, W.H.F., Wessel, P., 1990. Gridding with continuous curvature splines in tension.Geophysics 55, 293–305.

Spector, A., Grant, F.S., 1970. Statistical models for interpreting aeromagnetic data.

Geophysics 35, 293–302.Speranza, F., Minelli, L., Pignatelli, A., Gilardi, M., 2016. Curie temperature depths in the

Alps and the Po Plain (northern Italy): Comparison with heat flow and seismic to-mography data. J. Geodyn. 98, 19–30.

Sun, B., Wang, L.S., Dong, P., Wu, Y.J., Li, C.B., Hu, B., Wang, C., 2012. Integratedanalysis on gravity and magnetic fields of the Hailar Basin, NE China: Implications forBasement Structure and Deep Tectonics. Pure Appl. Geophys. 169, 2011–2029.

Sun, M., Chen, H., Zhang, F., Wilde, S.A., Minna, A., Lin, X., Yang, S., 2015. Cretaceousprovenance change in the Hegang Basin and its connection with the Songliao Basin,NE China: evidence for lithospheric extension driven by palaeo-Pacific roll-back.Geol. Soc., London, Special Publ. 413, 91–117.

Tanaka, A., Okubo, Y., Matsubayashi, O., 1999. Curie point depth based on spectrumanalysis of the magnetic anomaly data in East and Southeast Asia. Tectonophysics306, 461–470.

Tian, Z.Y., Han, P., Xu, K.D., 1992. The mesozoic cenozoic east China rift system.Tectonophysics 208, 341–363.

Trifonova, P., Zhelev, Z., Petrova, T., Bojadgieva, K., 2009. Curie point depths ofBulgarian territory inferred from geomagnetic observations and its correlation withregional thermal structure and seismicity. Tectonophysics 473, 362–374.

Turcotte, D.L., Schubert, G., 2002. Geodynamics. Cambridge University Press.Van der Voo, R., van Hinsbergen, D.J.J., Domeier, M., Spakman, W., Torsvik, T.H., 2015.

Latest Jurassic–earliest Cretaceous closure of the Mongol-Okhotsk Ocean: A paleo-magnetic and seismological-tomographic analysis. In: Anderson, T.H., Didenko, A.N.,Johnson, C.L., Khanchuk, A.I., MacDonald, J.J.H. (Eds.), Late Jurassic Margin ofLaurasia–A Record of Faulting Accommodating Plate Rotation. Geological Society ofAmerica.

Wan, T.F., Zhao, Q.L., Lu, H.F., Wang, Q.Q., Sun, C.B., 2016. Discussion on the speciallithosphere type in eastern China. Earth Sci. 5, 1–12.

Wang, J., Huang, S., 1990. Compilation of heat flow data for continental area of China(2nd edition) (in Chinese with English abstract). Seismol. Geol. 12 351-363+366.

Wang, J., Li, C.F., 2015. Crustal magmatism and lithospheric geothermal state of westernNorth America and their implications for a magnetic mantle. Tectonophysics 638,112–125.

Wang, J., Li, C.F., Lei, J.S., Zhang, G.W., 2016a. Relationship between seismicity andcrustal thermal structure in North China (in Chinese with English abstract). ActaSeismol. Sin. 38, 618–631.

Wang, P.J., Liu, W.Z., Wang, S.X., Song, W.H., 2002. Ar-40/Ar-39 and K/Ar dating on thevolcanic rocks in the Songliao basin, NE China: constraints on stratigraphy and basindynamics. Int. J. Earth Sci. 91, 331–340.

Wang, P.J., Mattern, F., Didenko, N.A., Zhu, D.F., Singer, B., Sun, X.M., 2016b. Tectonicsand cycle system of the Cretaceous Songliao Basin: An inverted active continentalmargin basin. Earth-Sci. Rev. 159, 82–102.

Wang, P.J., Xie, X.A., Frank, M., Ren, Y.G., Zhu, D.F., Sun, X.M., 2007. The CretaceousSongliao basin: Volcanogenic succession, sedimentary sequence and tectonic evolu-tion, NE China. Acta Geol. Sin.-Engl. 81, 1002–1011.

Wang, W.F., Jin, Q., Ma, Z.J., 1998. Meso-cenozoic evolution of the Tanlu fault andformation of sedimentary basins. Acta Geol. Sin.-Engl. 72, 350–362.

Wang, X.C., Wilde, S.A., Li, Q.L., Yang, Y.N., 2015. Continental flood basalts derived fromthe hydrous mantle transition zone. Nat. Commun. 6, 9.

Wang, X.K., Qiu, S.W., Song, C.C., Aleksey, K., Stepan, T., Evgeny, M., 1999. Cenozoicvolcanism and geothermal resources in Northeast China (in Chinese with Englishabstract). Geol. Rev. 45, 190–195.

Wang, Y., Cheng, S.H., 2012. Lithospheric thermal structure and rheology of the easternChina. J. Asian Earth Sci. 47, 51–63.

Wessel, P., Smith, W.H.F., Scharroo, R., Luis, J., Wobbe, F., 2013. Generic mapping tools:improved version released. Eos, Trans. Am. Geophys. Union 94, 409–410.

Wilde, S.A., 2015. Final amalgamation of the Central Asian Orogenic Belt in NE China:Paleo-Asian Ocean closure versus Paleo-Pacific plate subduction - A review of theevidence. Tectonophysics 662, 345–362.

Witter, J.B., Miller, C.A., Friend, M., Colpron, M., 2018. Curie point depths and heatproduction in Yukon, Canada, 43rd Workshop on Geothermal Reservoir Enginerring.Stanford University 1–11.

Wu, F.Y., Lin, J.Q., Wilde, S.A., Zhang, X.O., Yang, J.H., 2005. Nature and significance ofthe Early Cretaceous giant igneous event in eastern China. Earth Planet. Sci. Lett.233, 103–119.

Wu, F.Y., Sun, D.Y., Ge, W.C., Zhang, Y.B., Grant, M.L., Wilde, S.A., Jahn, B.M., 2011.Geochronology of the Phanerozoic granitoids in northeastern China. J. Asian EarthSci. 41, 1–30.

Wu, F.Y., Sun, D.Y., Li, H.M., Jahn, B.M., Wilde, S., 2002. A-type granites in northeasternChina: age and geochemical constraints on their petrogenesis. Chem. Geol. 187,143–173.

Xiao, L., Wang, F.Z., Wang, H., Pirajno, F., 2004. Mantle plume tectonics constraints onthe formation of Songliao and Bohaiwan Basins (in Chinese with English abstract).Earth Sci.-J. China Univ. Geosci. 29, 283–292.

Xu, M., Li, Y., Hou, H., Wang, C., Gao, R., Wang, H., Han, Z., Zhou, A., 2017. Structuralcharacteristics of the Yilan-Yitong and Dunhua-Mishan faults as northern extensionsof the Tancheng-Lujiang Fault Zone: New deep seismic reflection results.Tectonophysics 706–707, 35–45.

Xu, W.L., Pei, F.P., Wang, F., Meng, E., Ji, W.Q., Yang, D.B., Wang, W., 2013. Spatial-temporal relationships of Mesozoic volcanic rocks in NE China: Constraints on tec-tonic overprinting and transformations between multiple tectonic regimes. J. AsianEarth Sci. 74, 167–193.

Yan, Q.R., Gao, S.L., Wang, Z.Q., Hao, J., Xiao, W.J., Li, J.L., 2002. Geochronology andgeochemistry of volcanic rocks from the Songliao basin and their implications (inChinese with English abstract). Geochimica 31, 169–179.

Yang, J.H., O'Reilly, S., Walker, R.J., Griffin, W., Wu, F.Y., Zhang, M., Pearson, N., 2010.

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

192

Diachronous decratonization of the Sino-Korean craton: Geochemistry of mantlexenoliths from North Korea. Geology 38, 799–802.

Ye, L., 2017. Preliminary study on EGS geothermal resources evaluation and explorationin Songliao Basin (in Chinese with English abstract). Jilin Univ. 1–91.

Zeng, Q.D., Liu, J.M., Chu, S.X., Wang, Y.B., Sun, Y., Duan, X.X., Zhou, L.L., 2012.Mesozoic molybdenum deposits in the East Xingmeng orogenic belt, northeast China:characteristics and tectonic setting. Int. Geol. Rev. 54, 1843–1869.

Zhao, X.Y., Zeng, Z.F., Wu, Z.W., Wang, K., Li, J., Xu, T.F., 2015. Delineating the area ofHDR in Songliao basin using geophysical methods (in Chinese with English abstract).Prog. Geophys. 30, 2863–2869.

Zhou, X.H., 2006. Major transformation of subcontinental lithosphere beneath easternChina in the Cenozoic Mesozoic: review and prospect (in Chinese with English ab-stract). Earth Sci. Front. 13, 50–64.

J. Wang, C.-F. Li Journal of Asian Earth Sciences 163 (2018) 177–193

193