A Broadband Seismic Network in the Middle–Lower Yangtze Metallogenic Belt, Chinaby Xinfu Li, Longbin Ouyang, Hongyi Li, Yingjie Yang, Dan Zheng,Qingtian Lü, Ming Zhou, Jing Tan, Sanjian Sun, and Guibin Zhang

INTRODUCTION

The middle–lower Yangtze Metallogenic belt (MLYMB) issurrounded by the North China Block in the northwest, theDabie Orogen in the west, and the Yangtze Block in the south(Fig. 1a). The MLYMB experienced tectonic adjustment fromsouth–north compression in the late Jurassic period to exten-sion in the Cretaceous period (Ren et al., 1997), as indicated bythe extensive Cretaceous calc-alkaline to alkaline volcanism,granitoid magmatism, and synchronously rift volcanic basins.Major geologic features observed today, including the exten-sional tectonic settings in eastern China and the MLYMB, werebasically formed in this transitional period.

Seven large ore-concentration districts with more than 200kinds of polymetallic (Cu, Au, Mo, and so on) deposits are clus-tered in the narrow mineralization zone of the MLYMB. It is stillunclear how such rich mineral resources are concentrated insuch a narrow zone and what deep geodynamic processes andmagmatic activities are responsible for forming the massive met-allogenic belt. Many different models have been proposed, suchas lower-crust melting (Wang et al., 2001; Zhang et al., 2001,2002; Shi et al., 2013), thickening and delamination of lowercontinental crust (Wang et al., 2004; Hou et al., 2007), andsubduction of the paleo-Pacific plate (Ling et al., 2009). A com-prehensive seismological investigation can provide useful con-straints on the structure and dynamics of the crust andupper mantle in this region and can help to evaluate candidatemodels for the MLYMB formation.

Although there are many geophysical surveys in this region(e.g., Chang et al., 1991; Lü et al., 2003, 2004, 2010, 2013;Shi et al., 2013), these surveys mostly focused on imaging shal-low depths, and only a few geophysical investigations for deep-seated structures have been carried out (e.g., Shi et al., 2012,2013; Jiang et al., 2013). In 2011, the SinoProbe program con-ducted a 450 km long active source seismic experiment alongthe profile from Lixing to Yixing (Fig. 2), and a linear broad-band seismic array along the same profile were also deployedfrom November 2009 to August 2011 by the Chinese Acad-emy of Geological Sciences in the middle–lower Yangtze Riverto study the Moho discontinuity (Shi et al., 2013). These stud-ies focused on imaging structures along a linear seismic profile,unable to provide a 3D perspective on regional tectonics.

Because of low seismic activity in the middle–lower Yang-tze region and sparse seismic station coverage in the past, only afew surface-wave tomographic studies have been carried out inthis area (Song et al., 1993; Xu et al., 2000; Huang et al., 2003,2009). Song et al. (1993) and Xu et al. (2000) obtained 3Dshear-wave velocity structures in the crust and upper mantlebeneath east continental China, and their lateral resolutionsare about 4° × 4° and 5° × 5°, respectively. Huang et al. (2003)conducted Rayleigh-wave tomography of China and adjacentregions, and the resolution is about 4° × 4° for short periodsand 6° × 6° for long periods. Recently, several researchers con-ducted ambient noise tomography in the Dabie Orogen (Luoet al., 2012) and South China (Zhou et al., 2012). Their lateralresolutions at short-to-intermediate periods are about0:3° × 0:3°, and 0:5° × 0:5°, respectively.

To explore the crustal and upper-mantle structure of theMLYMB and better understand the genesis of metallic deposits,ChinaUniversity of Geosciences (Beijing) (CUGB) deployed atemporary broadband seismic network from May 2012. In thispaper, we first describe the middle–lower Yangtze broadbandseismic Network (MLYN) (Fig. 2) and then present some dataexamples and preliminary results.

STATION DISTRIBUTION

Our seismic network extends and complements previous seis-mic experiments in this region (e.g., Lü et al., 2004; Liu et al.,2010; Shi et al., 2013; Xu et al., 2014). The first phase ofMLYN consists of 20 stations operating from 20 May 2012to May 2014. All the 20 stations were deployed in the south-west part of the MLYMB with an average station spacing ofabout 50 km (Fig. 2). Our field deployment setup generallyfollows that for the USArray deployment (http://www.usarray.org/public/about/how#3; last accessed February2015; Fig. 1b). Each station includes a three-component Gür-alp CMG-3ESPC broadband sensor with a response from 60 sto 50 Hz and a RefTek 24-bit 130-1 digitizer. The field sta-tions are powered by solar panels. The second phase of MLYNbegan in June 2014 and is expected to finish in December2015. The 20 seismometers in the first phase were redeployedto the northeast part of the MLYMB along with 15 additionalCMG-3ESPCD seismometers, with the total number of sta-tions reaching 35. As shown in Figure 2, the distribution of

doi: 10.1785/0220140216 Seismological Research Letters Volume 86, Number 3 May/June 2015 941

the provincial stations (black triangles) in the northeastern partof the study area is relatively dense with a station spacing about60–70 km. However, in the rest area the average station spac-ing is about 150–200 km. Especially in the central and south-west part of the MLYMB, the station coverage is sparse and onlya few stations are available. Our deployment added many tem-

porary stations in the central and southwest part of theMLYMB, and the average station spacing can reach ∼50 km inthe entire region. With such station spacing, we expect to ob-tain a reasonably high-resolution crustal and upper-mantlevelocity structure in the MLYMB.

EXAMPLE DATA AND NOISE LEVEL

Figure 3a,b shows waveform examples (instrument-correcteddisplacement) from a teleseismic event and a local event re-corded by station AQ10 in the first phase. The parametersof these two events are shown in Table 1. We also computepower spectrum density levels of background noise in seismo-grams following Bendat and Piersol (1971) and Otnes andEnochson (1972) and compare them with the new low noiselevel models and the new high noise level models (NHNM)(Peterson, 1993). As shown in Figure 4a, the background noiselevels at the station AQ04 is low, especially at periods longer

115°

115°

117°

117°

119°

119°

121°

121°

29° 29°

31° 31°

33° 33°

114°

114°(a)

(b)

116°

116°

118°

118°

120°

120°

122°

122°

28° 28°

30° 30°

32° 32°

34° 34°

North China Block

Dabie Orogen

Yangtze Block

Nanjing

Tongling

Jiujiang

Hefei

Hangzhou

XGF

NZ

NW

TL

LZ

GC

AQ

JR

EDN

MLYMBTL

F

YCF

▴ Figure 1. (a) Schematic map of the tectonic background in themiddle–lower Yangtze Metallogenic belt (MLYMB) (study region).The blue patches are the known metallogenic regions. NZ, Ningz-hen; NW, Ningwu; LZ, Luzong; TL, Tongling; AQ, Anqing; GC,Guichi; JR, Jiurui; EDN, Edongnan; XGF, Xiangfan-Guangji fault;YCF, Yangxin-Changzhou fault; and TLF, Tancheng-Lujiang fault.The triangles denote the permanent stations deployed by ChinaDigital Seismological Network. (b) Installation of the sensor. A flatmarble board is fixed horizontally at the bottom of an excavation(the depth of the excavation is about 100 cm) using fast concretemix. The sensor is then placed on this marble inside this excava-tion, with its horizontal components oriented north–south. The ex-cavation is protected by a large, 50 cm × 50 cm box without abase panel, which allows the closing of the geophone spaceand is used to thermally isolate the sensor.

115°

115°

117°

117°

119°

119°

121°

121°

29° 29°

31° 31°

33° 33°

114°

114°

116°

116°

118°

118°

120°

120°

122°

122°

28° 28°

30° 30°

32° 32°

34° 34°

AQ10

AQ04

AQ20

TLF

XGF

YCF

▴ Figure 2. The location of the middle–lower Yangtze broadbandseismic Network (MLYN), the 46 seismic stations deployed by Chi-nese Academy of Geological Sciences and the 450 km long activesource seismic experiment conducted by Chinese Academy ofGeological Sciences. The first phase of the MLYN is made upof 20 broadband seismic stations, as indicated by the red triangles.The green triangles denote the transportable broadband stationsdeployed from November 2009 to August 2011 in the passive-source seismic experiment by Chinese Academy of GeologicalSciences, the yellow triangles denote the locations of the receiv-ers in the active-source seismic experiment, and the red stars de-note the locations of the shots (the SinoProbe Program). Theinverted purple triangles denote the locations of the 35 seismicstations in the second phase of the MLYN. The black trianglesdenote the provincial stations. The black dot denotes the earth-quake that occurred on 20 July 2012 in Yangzhou, Jiangsu Prov-ince, China. XGF, Xiangfan-Guangji fault; YCF, Yangxin-Changzhoufault; and TLF, Tancheng-Lujiang fault.

942 Seismological Research Letters Volume 86, Number 3 May/June 2015

than 1 s. Because the seismometers have a sharp response fre-quency cutoff at 0.0167 Hz (60 s), the actual noise levels atfrequencies lower than 0.0167 Hz are not well constrained.At periods shorter than 0.5 s (or frequency above 2 Hz),the noise levels become slightly higher but still reasonable con-sidering that this and many other stations are located nearcities. The relatively low background noise level guarantees thathigh-quality waveforms (from local to teleseismic earthquakes)can be recorded. In comparison, the background noise level atthe station AQ20 is relatively higher, especially at periodshorter than 2 s (Fig. 4b). This is because a quarry was devel-

oped after our instrument installation at this station, raisingthe high-frequency noise levels.

PRELIMINARY RESULTS

During the first phase deployment period (May 2012–June2014), several hundreds of teleseismic earthquakes with mag-

×

×

West-East

South-North

Vertical

0.10

0.05

0.00

–0.05

–0.10

0.05

0.00

–0.05

0.10

0.05

0.00

–0.05

0 20 40 60 80 100 120 140 160 180 200Time (sec)

AQ1010

–6A

mpl

itude

(m

)

1.0

0.5

0

–0.5

–1.0

1.0

0.5

0

–0.5

0.4

0

–0.4

0 5 10 15Time (sec)

West-East

South-North

Vertical

AQ10

10–4

Am

plit

ude

(m)

(a)

(b)

▴ Figure 3. (a) The 25 July 2012 mb 5.5 earthquake recorded atstation AQ10. (b) The 20 July 2012 mb 4.9 earthquake recorded atstation AQ10 that occurred in Yangzhou, Jiangsu Province, China.The station name is shown in Figure 2.

Table 1Event List

EventDate

(yyyy/mm/dd)Origin Time(hh:mm:ss) Latitude (°) Longitude (°) Depth (km)

EpicentralDistance (°) mb

a 2012/07/25 19:01:36 −19.297 167.728 35 69.47 5.5b 2012/07/20 12:11:52 32.978 119.593 10 4.45 4.9

–110

(a)

010 110 210Period (sec)

AQ04.BHZNLNMNHNM

REFTEK 130-1

CMG-3ESPC

10–1 100 101 102

Period (sec)

PS

D 1

0*L

OG

(M**

2/S

**4/

Hz)

AQ20.BHZNLNMNHNM

REFTEC 130-1

CMG-3ESPC

(b)

–80

–110

–140

–170

–200

PS

D 1

0*L

OG

(M**

2/S

**4/

Hz)

–80

–110

–140

–170

–200

▴ Figure 4. (a) Noise spectra from Z -component of the AQ04seismic station. (b) Noise spectra from Z -component of theAQ20 seismic station. The station names are shown in Figure 2.NLNM, new low noise level models; NHNM, new high noise levelmodels. The station name, component, type of data system, andseismometer model are listed on the top right corner of each plot.The spectra are plotted as units of decibels referred to1 �m= sec2�2= Hz as a function of period.

Seismological Research Letters Volume 86, Number 3 May/June 2015 943

nitude greater than 5.0 were recorded by our array. The num-ber of earthquakes is sufficient to perform several start-of-artseismic analyses that are described in the following sections.Although these methods provide constraints on different as-pects of the Earth’s structure, our ultimate goal is to integrateresults from different seismic methods to improve our under-standing of the deep geodynamic processes and the mechanismof magmatic activities contributing to the formation of themetallogenic belt. Specifically, we plan to explore whetherlithosphere detachment and asthenosphere upwelling exists,whether there is substantial evidence for lower-crustal magmainjection, and what is the pattern of lithosphere deformationand asthenosphere flow beneath the MLYMB.

Below, we briefly describe preliminary results from crosscorrelation of ambient noise and receiver functions. A detailedanalysis on each topic will be performed and reportedelsewhere.

Cross Correlation of Ambient Noise DataIn ambient noise tomography, interstation Green’s function(primarily surface waves) is extracted from cross correlationsof continuous noise recording (e.g., Shapiro et al., 2005).The interstation distances are typically required to be longerthan 3 wavelengths of interest to ensure that measureddispersion curves from cross correlations are stable and accu-rate (Bensen et al., 2007). Here, we perform ambient noisecross correlations by using the data from the first phase deploy-ment (Fig. 5a). The processing procedures generally followthose of Bensen et al. (2007) and are briefly described below.First, trend, mean value, and instrument response are removedfrom raw seismograms. Then, the seismograms are band-passfiltered at 0.2–0.01 Hz and decimated to 1 samples=s andcut into a series of one-hour-long segments. Any segment withspikes 10 times larger than the root mean square of the entiretime series is discarded. Finally, spectral whitening is applied tothe retained segments. The one-hour-long time series betweenall station pairs are cross correlated and then stacked togetherto form the stacked cross correlations. Only those stacked crosscorrelations with signal-to-noise ratio (SNR) greater than 8 andinterstation distances larger than three wavelengths of the in-terested surface waves are retained. Figure 5b shows clear sur-face-wave signals at both negative and positive time lags bystacking 14-month cross-correlation functions.

Surface-wave tomography (either based on natural earth-quakes or ambient noise) complements body-wave tomographyin two aspects: (1) surface-wave tomography can provide betterresolution in imaging the crustal and shallow uppermost-man-tle structure, and (2) it constrains absolute velocities of theEarth, which is important for directly interpreting the originsof velocity features. Considering the aperture of our deployedarray, we expect to obtain phase velocities at 5–50 s periodsfrom ambient noise tomography. On the other hand, usingteleseismic two-plane-wave tomography (Forsyth and Li, 2005;Yang and Forsyth, 2006), we are able to obtain phase velocitymaps at periods from 20 to 150 s. Hence, both phase velocity

and azimuthal anisotropy maps can be obtained by applyingthese methods to data from the two-year deployment.

In a recent study (Ouyang et al., 2014), both ambientnoise tomography and teleseismic two-plane-wave tomographyare used to obtain phase velocities, and image the 3D shearstructures of the lithosphere and underlying asthenosphere

50

100

150

200

250

300

Dis

tanc

e (k

m)

–200 –100 0 100 200Time (second)

115°

115°(a)

(b)

116°

116°

117°

117°

118°

118°

119°

119°

28° 28°

29° 29°

30° 30°

31° 31°

▴ Figure 5. (a) Station distribution and the corresponding ray-path coverage used to compute the cross correlations shownin (b). (b) 14-month cross correlations filtered between 5 and50 s periods.

944 Seismological Research Letters Volume 86, Number 3 May/June 2015

in this region. By including those stations from our first phasedeployment in the inversion (Ouyang et al., 2014), the lateralresolutions for ambient noise tomography and two-plane-wave tomography can reach ∼0:2° and 1°, respectively. Theirhigh-resolution images showed a clear low-velocity zone in theupper mantle (∼100–200 km) beneath the MLYMB, whichmay be caused by partial melting of an enriched mantle sourceassociated with subduction of ancient Pacific plate (Ouyanget al., 2014). Their results suggest that surface-wave tomogra-phy could provide useful information for better understandingof deep seismological structures and geodynamic processes be-neath the MLYMB.

Receiver Function AnalysisThe two-year recordings make it possible to obtain reliablereceiver functions and quantify lateral variations of crustal

thickness and VP=V S ratio in the MLYMB and the surroundingregions. These parameters in turn would help to understand themechanism of magmatic activities and rock compositions in thestudy area.

We adopt a modified receiver function method (Zhu andKanamori, 2000) to measure crustal thickness and VP=V Sratio in this region. To obtain receiver functions, we first vis-ually examine all seismograms from 80 earthquakes occurredbetween June 2012 and August 2013 with magnitude greaterthan 5.4 and epicentral distances between 30° and 90°. Next,we only picked 72 earthquakes with high SNR. The earthquakedistribution generally shows a reasonable distance and azimu-thal coverage (Fig. 6a). A 65 s time window (5 s before and 60 safter the P arrival) is then used to isolate the interested bodywaves and calculate receiver functions. Figure 6b shows anexample of receiver functions recorded at station AQ10. The

(a) (b)

(c)

▴ Figure 6. (a) Teleseismic events used in the receiver function analysis. The blue star denotes the position of the MLYN. (b) An exampleof the individual receiver functions sorted by epicentral distance recorded at station AQ10 located in the center of the whole network. Thered, black, and green dashed lines indicate the phases of Ps, PpPs, and PpSs � PsPs. (c) H�κ stacking results correspondingly. The redcircle denotes the best estimate of the crustal thickness and VP = V S ratio, and the ellipse is the uncertainty estimation.

Seismological Research Letters Volume 86, Number 3 May/June 2015 945

primary P to S conversions and the Moho multiples are clearlyvisible in the receiver functions as seen in Figure 6b. With theobtained receiver functions, we then perform a joint inversionto measure the Moho depth and VP=V S ratio at this station(Fig. 6c). We plan to apply the same procedure to the rest ofthe stations and obtain a spatial variation of Moho depth andVP=V S ratio in this region. These results can be used to testthe existence of lower-crustal delamination and partial melting.

SUMMARY

A temporary seismic network was deployed to record seismicdata for studying the crustal and upper-mantle structure in theMLYMB. The first phase of the network consists of 20 broad-band stations covering the south part of the MLYMB, and thesecond phase covers the northeast part of the MLYMB with atotal of 35 stations including the redeployed 20 stations fromthe first phase. The installation method adopted for theseseismic stations generally follows that for the USArray deploy-ment. This network significantly increases the station coveragein the MLYMB and has successfully recorded continuous datasince 2012. Many seismological methods can be applied (e.g.,Ouyang et al., 2014) to obtain high-resolution structures in thecrust and upper mantle and their relationships with generationof deep geodynamic processes and shallow ore formation.

DATA AND RESOURCES

The project is expected to finish in December 2015. The ex-pected total volume of data is ∼900 GB. The data will be an-alyzed by the Principal Investigators (PI)s of this project(Qingtian Lü, Guibin Zhang, Xinfu Li, Hongyi Li, andGuoming Jiang) and are also available for collaborative workat this stage. Our data will be open to the public after two yearsof this project and also will be available from the data man-agement center for the Deep Exploration in China (http://www.sinoprobe.org/DataShare.aspx; last accessed December2014, currently under construction).

ACKNOWLEDGMENTS

We are grateful to Editor Zhigang Peng, Managing EditorMary George, and an anonymous reviewer for their helpfulcomments and critical reviews. We thank Xiaoming Xu (work-ing in the Institute of Geophysics, China Earthquake Admin-istration) for his helpful discussions. This work is funded bythe Ministry of Land and Resources of China under theProject SinoProbe-03, the National Science Foundation ofChina (Grant Numbers 41374057 and 41474045), and theProgram for New Century Excellent Talents in University(NCET).

REFERENCES

Bendat, J. S., and A. G. Piersol (1971). Random Data: Analysis andMeasurement Procedures, John Wiley and Sons, New York, NewYork, 407 pp.

Bensen, G. D., M. H. Ritzwoller, M. P. Barmin, A. L. Levshin, F. Lin, M.P. Moschetti, N. M. Shapiro, and Y. Yang (2007). Processing seismicambient noise data to obtain reliable broad-band surface wavedispersion measurements, Geophys. J. Int. 169, 1239–1260.

Chang, Y., X. Liu, and Y. Wu (1991). The Copper-Iron Belt of the Lowerand Middle Reaches of the Yangtze River, Geological PublishingHouse, Beijing, China.

Forsyth, D. W., and A. Li (2005). Array analysis of two-dimensionalvariations in surface wave phase velocity and azimuthal anisotropyin the presence of multipathing interference, Geophys. Monog. Ser.157, 81–97.

Hou, Z., X. Pan, Z. Yang, and X. Qu (2007). Porphyry Cu–(Mo–Au)deposits not related to oceanic-slab subduction: Examples fromChinese porphyry deposits in continental settings, Geoscience 21,332–351 (in Chinese with English abstract).

Huang, Z., H. Li, Y. Zheng, and Y. Peng (2009). The lithosphere ofNorth China Craton from surface wave tomography, Earth Planet.Sci. Lett. 288, 164–173.

Huang, Z., W. Su, Y. Peng, Y. Zheng, and H. Li (2003). Rayleigh wavetomography of China and adjacent regions, J. Geophys. Res. 108,no. B2, 2073, doi: 10.1029/2001JB001696.

Jiang, G., G. Zhang, Q. Lü, D. Shi, and Y. Xu (2013). 3-D velocity modelbeneath the middle–lower Yangtze River and its implication to thedeep geodynamics, Tectonophysics 606, 36–47.

Ling, M., F. Wang, X. Ding, Y. Hu, J. Zhou, R. Zartman, X. Yang, andW. Sun (2009). Cretaceous ridge subduction along thelower Yangtze River belt, Eastern China, Econ. Geol. 104,303–321.

Liu, J., C. Kuang, R. Gao, A. Xue, Z. Lu, H. Hou, and H. Wang (2010).Data processing test and research on the deep seismic reflectionprofile in polymetallic deposits area: Taking an example of Luzongore concentrated area, Acta Petrol. Sin. 26, no. 9, 2561–2576.

Lü, Q., L. Han, J. Yan, Y. Lian, D. Shi, and T. Yan (2010). Seismic im-aging of volcanic hydrothermal iron-sulfur deposits and its hostingstructure in Luzong ore district, Acta Petrol. Sin. 26, no. 9, 2598–2612 (in Chinese with English abstract).

Lü, Q., Z. Hou, D. Shi, J. Zhao, M. Xu, and M. Chai (2004). Tentativeseismic reflection study of Shizishan Orefield in Tongling and itssignificance in regional exploration, Mineral Deposits 23, no. 3,390–398 (in Chinese with English abstract).

Lü, Q., Z. Hou, J. Zhao, D. Shi, X. Wu,Y. Chang, R. Pei, D. Huang, andC. Kuang (2003). Deep seismic reflection profiling reveals complexcrustal structure of Tongling ore district, Sci. China Earth Sci. 47,443–449.

Lü, Q., J. Yan, D. Shi, S. Dong, J. Tang, M. Wu, and Y. Chang (2013).Reflection seismic imaging of the Lujiang-Zongyang volcanicbasin, Yangtze Metallogenic belt: An insight into the crustalstructure and geodynamics of an ore district, Tectonophysics 606,60–77.

Luo, Y., Y. Xu, and Y. Yang (2012). Crustal structure beneath the Dabieorogenic belt from ambient noise tomography, Earth Planet. Sci.Lett. 313/314, 12–22.

Otnes, R. K., and L. Enochson (1972). Digital Time Series Analysis, JohnWiley and Sons, New York, New York, 467 pp.

Ouyang, L., H. Li, Q. Lü, Y. Yang, X. Li, G. Jiang, G. Zhang, D. Shi, D.Zheng, S. Sun, J. Tan, and M. Zhou (2014). Crustal and uppermostmantle velocity structure and its relationship to the formation of oredistricts in the middle–lower Yangtze River region, Earth Planet.Sci. Lett. 408, 378–389, doi: 10.1016/j.epsl.2014.10.017.

Peterson, J. (1993). Observation and modeling of seismic backgroundnoise, U.S. Geol. Surv. Technical Rept. 93-312, 1–95.

Ren, J., B. Niu, Z. He, G. Xie, and Z. Liu (1997). Tectonic frame andgeodynamic evolution of eastern China,Geol. Res. 29, no. 30, 43–55(in Chinese with English abstract).

Shapiro, N. M., M. Campillo, L. Stehly, and M. H. Ritzwoller (2005).High-resolution surface-wave tomography from ambient seismicnoise, Science 307, no. 5715, 1615–1618.

946 Seismological Research Letters Volume 86, Number 3 May/June 2015

Shi, D., Q. Lü, W. Xu, J. Yan, J. Zhao, S. Dong, and Y. Chang (2012).Crustal structures beneath the mid-lower Yangtze Metallogenic beltand its adjacent regions in eastern China—Evidences from P-wavereceiver function imaging for a MASH metallization process? ActaGeol. Sin. 86, no. 3, 389–399 (in Chinese with English abstract).

Shi, D., Q. Lü,W. Xu, J. Yan, J. Zhao, S. Dong,Y. Chang, and SinoProbe-03-02 Team (2013). Crustal structure beneath the middle–lowerYangtze metallogenic belt in East China: Constraints from passivesource seismic experiment on the Mesozoic intra-continentalmineralization, Tectonophysics 606, 48–59.

Song, Z., G. Chen, C. An, L. Chen, Z. Zhuang, Z. Fu, Z. Lv, and J. Hu(1993). 3-D velocity structure of crust and upper mantle beneathcontinental China and the sea areas, Sci. China (Series B) 23, no. 2,180–188.

Wang, Q., J. Xu, Z. Zhao, Z. Bao, W. Xu, and X. Xiong (2004).Cretaceous high-potassium intrusive rocks in the Yueshan–Hongzhen area of east China: Adakites in an extensional tectonicregime within a continent, Geochem. J. 38, 417–434.

Wang, Q., Z. Zhao, X. Xiong, and J. Xu (2001). Melting of the underplat-ing basaltic lower crust: Evidence from the Shaxi adakite-like sodium-rich quartz diorite–porphyrite in Anhui Province, Geochimica 30,no. 4, 353–362 (in Chinese with English abstract).

Xu, G., G. Li, S. Wang, H. Chen, and H. Zhou (2000). The 3-D structureof shear waves in the crust and mantle of east continental Chinainverted by Rayleigh wave data, Chin. J. Geophys. 43, no. 3,366–376 (in Chinese with English abstract).

Xu, T., Z. Zhang, X. Tian, B. Liu, Z. Bai, Q. Lü, and J. Teng (2014).Crustal structure beneath the middle–lower Yangtze metallogenicbelt and its surrounding areas: Constraints from active sourceseismic experiment along the Lixin to Yixing profile in East China,Acta Petrol. Sin. 30, no. 4, 918–930 (in Chinese with Englishabstract).

Yang, Y., and D. W. Forsyth (2006). Regional tomographic inversion ofamplitude and phase of Rayleigh waves with 2-D sensitivity kernels,Geophys. J. Int. 166, 1148–1160.

Zhang, Q., Y. Wang,W. Liu, and Y. Wang (2002). Adakite: Its character-istics and implications, Geol. Bull. China 21, 431–435 (in Chinesewith English abstract).

Zhang, Q., Y. Wang, Q. Qian, J. Yang, Y. Wang, T. Zhao, and G. Guo(2001). The characteristics and tectonic–metallogenic significancesof the adakites in Yanshan period from eastern China, Acta Petrol.Sin. 17, 236–244 (in Chinese with English abstract).

Zhou, L., J. Xia,W. Shen,Y. Zheng,Y. Yang, H. Shi, and M. H. Ritzwoller(2012). The structure of the crust and uppermost mantle beneathSouth China from ambient noise and earthquake tomography,Geophys. J. Int. 189, 1565–1583.

Zhu, L., and H. Kanamori (2000). Moho depth variation in southernCalifornia from teleseismic receiver functions, J. Geophys. Res.105, 2969–2980.

Xinfu Li1

Key Laboratory of Geo-detectionChina University of Geosciences, Beijing

Ministry of EducationBeijing 100083, China

xinfuli@cugb.edu.cn

Longbin OuyangHongyi LiDan ZhengMing ZhouJing Tan

Sanjian SunGuibin Zhang

School of Geophysics and Information TechnologyChina University of Geosciences

Beijing 100083, China

Yingjie YangCCFS, GEMOC ARC National Key CentreDepartment of Earth and Planetary Sciences

Macquarie UniversityNorth Ryde, New South Wales 2109, Australia

Qingtian LüMLR Key Laboratory of Metallogeny and Mineral Assessment

Institute of Mineral ResourcesChinese Academy of Geological Sciences

Beijing 100037, China

Published Online 1 April 2015

1 Also at School of Geophysics and Information Technology, China Uni-versity of Geosciences, Beijing 100083, China.

Seismological Research Letters Volume 86, Number 3 May/June 2015 947

View publication statsView publication stats