Full text of DOI:10.1007/BF02959450

Vol. 43 No. 4
SCIENCE IN CHINA (Series D)
August 2000
Crustal motion of Chinese mainland monitored by GPS
ZHU Wenyao (ᅋำྒ)1, WANG Xiaoya (ฆ໌ཐ)1, CHENG Zongyi (ёᆗྪ)1,
XIONG Yongqin (༇စூ)1, ZHANG Qiang (჆ ஜ)1, YE Shuhua (ྜྷ೜ܟ)1,
MA Zongjin (৴ᆗࠫ)2, CHEN Junyong (чࢎဇ)3,
XU Houze (Houtze Hsu ༘ ܈ ႖)4, WEI Ziqing (ษᆐு)5, LAI Xi’an (ऀ๬̝)6,
LIU Jingnan (ঞ࠼઒)7, JIN Biaoren (ࠡγఅ)7, REN Jinwei (ఉࠡฯ)2
& WANG Qi (ฆ ត)6
1. Shanghai Astronomical Observatory, Chinese Academy of Sciences, Shanghai 200030, China;
2. Institute of Geology, State Seismological Bureau, Beijing 100029, China;
3. National Bureau of Surveying and Mapping, Beijing 100830, China;
4. Wuhan Institute of Geodesy and Geophysics, Chinese Academy of Sciences, Wuhan 430077, China;
5. Xi’an Institute of Surveying and Mapping, Xi’an 710054, China;
6. Institute of Seismology, State Seismological Bureau, Wuhan 430071, China;
7. Wuhan Technical University of Surveying ang Mapping, Wuhan 430070, China
Correspondence should be addressed to Zhu Wenyao (email: zhuw@center.shao.ac.cn)
Received May 25, 1999
Abstract To measure and monitor the crustal motion in China, a GPS network has been estab-
lished with an average side length of 1 000 km and with more than 20 points on the margins of
each major tectonic block and fault zone in China. Three campaigns were carried out in 1992,1994
and 1996, respectively by this network. Here we present, for the first time, the horizontal dis-
placement rates of 22 GPS monitoring stations distributed over the whole China and global IGS
stations surrounding China, based on these GPS repeated measurements. From these results by
GPS, we have obtained the sketch of crustal motion in China.
Keywords: crustal motion in China, GPS monitoring.
Chinese mainland lies among the Indian plate, the Philippines sea plate, Siberian and Mon-
golian plate. It has a peculiar tectonic feature. The inter-plate deformation and interaction between
blocks are very complicated. Especially, the Qinghai-Xizang plateau, called the world ridge, is the
youngest plateau in the world at present. It is the region where the crustal motion is the strongest
among the global continents. Therefore, the Chinese mainland is an ideal object for studying in-
traplate deformation and dynamics, and is regarded as the key to studying the lithospheric dynam-
ics of continents. Due to the limitation of observational accuracy, the quantitative survey of the
present-day crustal motion in China is basically the component of vertical motion by means of
gravimetric and leveling survey. Although there are data from geology, topology, seismology and
ü
some local fault zone monitoring[1 5], the reliable observation of the large scale crustal horizontal
motion is not available till now. Hence there is no convincing data of horizontal displacement to
study the dynamical mechanism of the crustal motion of the Chinese mainland. Now the devel-

No. 4
CRUSTAL MOTION OF CHINESE MAINLAND MONITORED BY GPS
395
opment of GPS technique provides a guarantee for directly surveying large scale crustal motion of
Chinese mainland[6,7]. A GPS network for measuring and monitoring crustal motion of China has
been completed, through which three campaigns have been made in 1992,1994 and 1996 respec-
tively. The GPS network covers the whole mainland of China with 22 points carefully selected by
Chinese geologists (see fig. 1). Through repeated measurements for several periods, we expect to
resolve the relative crustal horizontal motion with an accuracy better than 5 mm/a. These results
will be combined with gravimetric survey, leveling, geological, seismic and geomorphological
approaches to gradually establish a quantitative model about large-scale deformation. Using
modified GAMIT package[8], we have preliminarily determined the change rates of the station
coordinates by repeated measurements in three campaigns. These results may provide convincing
information related to the horizontal tectonic motion in Chinese mainland.
Fig. 1. The horizontal site velocity of China from GPS and NNR-NUVEL1A.
1
Campaigns and data processing
In October, 1991, a group, in charge of the task of “Investigation on Present Crustal Motion
and Geodynamics in China” under a key basic research program, invited specialists concerned in
geology, geodynamics and geodesy all over the country to discuss a plan for establishing a GPS
crustal motion monitoring network in China. It has been proposed to establish a GPS network with
an average side length of 1 000 km and more than 20 points on the margins of each major tectonic
block and fault zone in the mainland. Subsequently, three campaigns by the Chinese GPS network
were carried out in 1992,1994 and 1996 respectively. Owing to the conditions at that time, the first
observation campaign was only performed in the some areas of the Qinghai-Xizang Plateau with 4

396
SCIENCE IN CHINA (Series D)
Vol. 43
sets of Leica 200 GPS receivers during Oct. 8üOct. 12, and Oct. 21üOct. 24. The last two cam-
paigns spread over the whole mainland. The second campaign, employing 15 sets of Leica 200
GPS receivers, was divided into two sections: eastern section from Aug. 15 to Aug. 20 and west-
ern section from Aug. 28 to Sept. 2. These two sections, where Urumqi, Golmud and Xiaguan sta-
tions were regarded as three common points, consist of 15 and 10 stations respectively. The third
campaign was operated on the 22 stations in the same period from July. 25 to Aug. 3 with such
three kinds of receivers as Ashtechz12, Rogue SNR 8000 and Leica 200. There were seven com-
mon baselines in the Qinghai-Xizang area to be repeatedly measured in the three campaigns. They
are Lhasa-Wenquan, Golmud-Wenquan, Golmud-Xining, Lhasa-Xiaguan, Golmud-Xialatuo, Wen-
quan-Xiaguan and Xining-Xialatuo.
In order to maintain the unity and self-consistency of the terrestrial reference frame during
the three campaign observations and to study the background of crustal motion of Chinese
mainland, we chose the following international terrestrial reference frame (ITRF) sites as fiducial
ones whose data were processed simultaneously together with our network data: ALGO, ALBH,
TAIW, TSKB, GOLD, FAIR, KOKB, ONSA, MASP, HART, SANT, WETT, HERS, PAMA,
KWJ1, YAR1, YELL, MAC1 and so on. In addition, five IGS stations surrounding China, TAEJ,
GUAM, KIT3, IRKT and IISC, were added in the 1996 data processing.
The observations in integrated adjustment for the whole network are daily baseline solutions.
The adjustments for multi-day data were combined by using their variance-covariance matrices
with the Kalman filter program GLOBK package in ITRF96 with more than 20 global GPS sites
being fiducial.
The method of data processing for daily baseline solution is summarized as follows:
Software used
Modified GAMIT
Measurement model
Basic observables
L1, L2 two frequency carrier phases; P code only for receiver clock
and ambiguity estimation; elevation angle cutoff is 15 degrees;
sampling interval is 60 s; weighting is uniform.
Modeled observables
Troposphere
Double difference ionosphere-free linear combination
A priori Saastamoinen atmospheric correction model
Displacements due to solid earth tidal and ocean loading come from
IERS standard (1995); atmospheric load and polar tides are not
considered.
Tidal displacements
Satellite orbits
EOP
IGS precise ephemerides for 1994,1996, and GFZ for 1992
IGS EOP series for 1994 and 1996, and GFZ EOP series for 1992
Estimated parameters
Station coordinates
To select IGS stations as the fiducial points, its daily coordinate
updates to be from its coordinates and velocities in ITRF96 were
assigned the a priori constraint 1ꢀ (from ITRF96) in the estimation.

No. 4
CRUSTAL MOTION OF CHINESE MAINLAND MONITORED BY GPS
397
Other station coordinates were estimated without a priori con-
straints
Satellite orbits
To adopt loose constraints 10-7 for the initial state vectors of satel-
lite
orbits and 10% for three scale factors of solar radiation pressure
Zenith delay estimated once two hours for each station
Estimated as real values without a priori
Troposphere
Ambiguity
Station clock bias
Station clock biases including epoch, rate and acceleration are
estimated during the preprocessing using P code measurements.
Averaged repeatabilities in daily solution of site coordinates for latitudinal, longitudinal and
radial components are listed below.
Latitude /10ꢁ4 s
Longitude/10ꢁ4 s
Radial/10ꢁ3 m
1992
1994
1996
2.33.0
1.6
1.5
24.3
2.1
2.0
22.1
18.4
2 Results and analysis
The results of horizontal site velocities for GPS crustal motion monitoring network in China
are shown in table 1.
Table 1 Horizontal site velocities for GPS crustal motion monitoring network in China
Lati.
/ (°)
Long.
/ (°)
Nrate
/mmgaꢁ1
( by GPS)
15.5ꢂ1.7
Erate
Nrate
/mmgaꢁ1
(by NNR-NUVEL1A) (WRT Eurasian plate )
Erate
Nrate
Erate
Sites
/mmgaꢁ1
Lhasa
29.50
33.05
36.25
36.48
39.57
31.13
25.46
39.33
36.93
38.85
32.33
41.50
43.62
39.68
26.57
43.60
39.42
30.50
23.03
24.63
31.10
91.36
91.86
94.88
101.65
98.50
100.75
100.25
75.92
79.93
88.02
80.10
86.17
50.4ꢂ2.3
48.9ꢂ2.4
36.8ꢂ2.2
35.6ꢂ2.5
34.3ꢂ2.7
49.0ꢂ2.4
33.1ꢂ2.2
23.4ꢂ2.7
24.1ꢂ2.7
25.5ꢂ2.7
30.1ꢂ3.9
30.7ꢂ2.7
31.2ꢂ2.3
28.5ꢂ2.8
34.4ꢂ2.7
19.2ꢂ2.8
26.6ꢂ3.0
34.5ꢂ2.5
29.7ꢂ2.9
34.1ꢂ3.1
32.5ꢂ2.0
ꢁ6.6
ꢁ6.7
25.1
25.1
24.9
24.3
24.5
24.4
24.1
25.9
25.7
25.3
25.6
25.4
25.2
23.3
23.7
20.6
22.5
22.8
23.0
22.3
22.3
22.1
17.1
14.2
10.3
11.0
0.0
25.3
23.8
11.9
11.3
10.8
24.6
9.0
Wenquan
Golmud
Xining
Jiuquan
Xialatuo
Xiaguan
Kashi
10.4ꢂ1.8
6.7ꢂ1.5
1.1ꢂ1.8
ꢁ7.5
ꢁ9.2
ꢁ8.4
ꢁ9.0
2.6ꢂ2.0
ꢁ9.0ꢂ1.8
ꢁ11.1ꢂ1.5
16.5ꢂ2.1
18.5ꢂ2.1
7.7ꢂ2.2
10.5ꢂ3.4
8.8ꢂ2.3
2.3ꢂ1.7
ꢁ13.9ꢂ2.1
ꢁ10.8ꢂ2.1
ꢁ13.9ꢂ2.1
ꢁ14.9ꢂ2.3
ꢁ8.3ꢂ1.8
ꢁ13.4ꢂ2.2
ꢁ10.3ꢂ2.4
ꢁ13.4ꢂ1.4
ꢁ9.2
ꢁ2.4
ꢁ3.5
ꢁ5.8
ꢁ3.5
ꢁ5.2
ꢁ5.6
ꢁ1.9
18.9
22.0
13.5
14.0
14.0
7.9
ꢁ2.5
ꢁ1.6
0.2
4.5
5.3
Hotan
Ruoqiang
Shiquanhe
Korla
Urumqi
Dongsheng
Guiyang
Changchun
Beijing
87.71
6.0
109.96
106.68
125.44
115.89
114.25
113.34
118.60
121.20
ꢁ11.2
ꢁ10.4
ꢁ13.9
ꢁ12.3
ꢁ12.4
ꢁ11.8
ꢁ13.3
ꢁ13.3
ꢁ2.7
ꢁ0.4
0.0
ꢁ2.6
4.1
ꢁ1.6
3.0
0.1
5.2
10.7
ꢁ1.4
4.1
11.7
6.7
Wuhan
Guangzhou
Quanzhou
Shanghai
11.8
10.2

398
SCIENCE IN CHINA (Series D)
Vol. 43
In order to study the background field of Chinese crustal motion, from May 1995 to June
1998, we also carried out successive measurements of displacement velocities of 12 IGS sites dis-
tributed in China and its neighborhood, including Shanghai, Lhasa, Taibei, Wuhan, Xi’an, Irkutsk,
Kitab, Taejon, Tsukuba, Usuda, Bangalore and Guam. The results are shown in table 2 [9], where
corresponding results from ITRF96 are also listed for comparison. For the sake of intuitions, we
also give fig. 1 showing these horizontal site velocities estimated by GPS and NNR-NUVEL1A.
Table 2 Horizontal velocities for some IGS sites in China and its neighbourhood by crustal
motion monitoring network in China
Nrate
Erate
Nrate
Erate
Nrate
Erate
Sites
/mmgaꢁ1
( by GPS )
/mmgaꢁ1
/mmgaꢁ1
( From ITRF96 )
(WRT Eurasian plate )
37.3ꢂ0.6
5.6ꢂ0.4
ꢁ7.6ꢂ0.5
3.6ꢂ0.3
18.1ꢂ0.3
ꢁ11.8ꢂ0.3
ꢁ14.6ꢂ0.2
ꢁ9.6ꢂ0.3
ꢁ9.5ꢂ0.3
ꢁ8.0ꢂ0.2
ꢁ8.3ꢂ0.3
ꢁ8.1ꢂ0.3
41.0ꢂ1.2
ꢁ7.5ꢂ0.6
24.6ꢂ0.7
31.0ꢂ1.0
46.6ꢂ0.9
33.1ꢂ0.8
28.9ꢂ0.6
36.4ꢂ0.8
ꢁ5.6ꢂ0.8
ꢁ0.1ꢂ0.7
31.3ꢂ0.8
29.4ꢂ0.8
36.4
4.4
ꢁ6.7
4.1
17.3
ꢁ13.4
ꢁ15.4
ꢁ11.2
ꢁ10.7
ꢁ6.9
ꢁ7.4
ꢁ12.1
18.2
40.1
38.7
ꢁ9.4
25.1
29.9
44.4
32.5
29.8
36.5
ꢁ2.0
2.6
IISC
GUAM
IRKT
21.5
2.2
ꢁ29.9
1.8
1.2
5.1
KIT3
24.7
0.5
20.5
10.8
7.8
LHAS
SHAO
TAEJ
TAIW
TSKB
USUD
WUHN
XIAN
ꢁ0.4
3.7
6.2
14.1
ꢁ22.8
ꢁ19.4
8.4
7.6
4.1
3.7
36.6
23.8
6.2
The station LHAS in table 2 is different from that in table 1, they are 20 km apart.
Measurement precision for horizontal site velocity (north component and east component) in
the tables is estimated by
ꢀ r
T
12(n ꢁ1)
n(n ꢃ1)
ꢀ v ꢄ
,
where ꢀ r is the standard deviation of single solution of site coordinate components, n the num-
ber of single solutions, and T the span of the whole observations in unit of year[10]
.
From the results in tables 1, 2 and fig. 1, we may see the sketch of crustal motion of Chinese
mainland in the Eurasian plate.
1) The consistency of the size and the direction of the horizontal velocities from GPS data
and NNR-NUVEL1A for three IGS stations IISC, KIT3 and IRKT, which are situated to the south,
west and north of China respectively, suggests that the effects of the Indian plate, Eurasian plate
and the Siberian blocks on Chinese mainland have been stable in the past one million years.
2) Due to the complicated tectonic structure of the mainland crust, its horizontal motion ap-
pears to be apparently inhomogeneous. Observational results indicate that the crustal motion be-
tween the east and west of China has an explicit difference. Taking Longmen fault zone and An-
ning river-Xiaojiang zone (N-S seismic zone of China) on the eastern fringe of the Qing-
hai-Xizang plateau as a boundary, the effect of the Indian plate on the crustal deformation in the
east of China is by far weaker than that in the west of China. The relative crustal motion in the
whole west of China under the effect of the strong northward propulsion by the Indian plate ap-

No. 4
CRUSTAL MOTION OF CHINESE MAINLAND MONITORED BY GPS
399
pears to shorten in north-south direction and lengthen in east-west direction, and what is more, the
latter is larger than the former. Stations located in the east part of the west of China, such as
Lhasa,Wenquan, Golmud, Xining and Jiuquan, have an explicit north-eastward motion with re-
spect to Eurasian plate and their moving rates gradually reduce from south to north. However,
Kashi and Hotan stations in the west part have only a northward motion and no explicit eastward
motion. For the first time, these results have provided a direct observational proof in support of
such an important geological inference that “the northward push-pressing of the Indian plate has
two prominent sections: the Pamir wedge in the west and the Asama wedge in the east. These two
wedges advance northward at different velocities. The Pamir wedge deeply penetrates into the
Eurasian plate at a northwestward velocity of 43 mm/a; while the Asama wedge dives deeply into
the Eurasian plate at a northeastward velocity of 64 mm/a”[1, 11]
3) It is indicated by these GPS meas-
.
urement results that there are obvious block
characteristics for the crustal motion in the
west of China. Fig.2 shows the locations of
the major tectonic blocks and horizontal ve-
locities of observing sites in the west of
China. There are the largest northward and
eastward movements for both Lhasa and
Wenquan sites located in the Xizang block of
Qinghai-Xizang subplate. That the moving
rate of Wenquan site is less than that of Lhasa
site indicates that the effect of the
Fig. 2. The major tectonic blocks and horizontal site velocity in
west China. I. Himalaya block; II, Xizang block; III. Gan-Qing
block; III, Chuan-Dian block; V. Tarim block.
push-pressing coming from the Indian plate
gradually reduces from south to north. In fact,
ref. [1] indicates that taking Bangongcuo-Nujiang suture zone as a boundary by dashed line in fig.
2, the Xizang block can be divided into two tectonic subblocks. Lhasa and Wenquan sites are situ-
ated respectively on these blocks in the south and north. Golmud, Xining and Jiuquan stations are
located in the Ganqing block of the Qinghai-Xizang subplate. They possess the northward and
eastward movement at almost the same velocity, which is weaker than that of Lhasa and Wenquan
sites. The small differences of the velocities among these sites also appear to gradually diminish
northwards and eastwards, and may reveal that the Ganqing block is not a rigid one. In the west,
there are two blocks whose crustal motion is special. One is located in the west of Yunnan Prov-
ince, which is a rhombic block enclosed by Jinshajiang River, Red River fault zone and a series of
north-south directional fault zone in the south part of north-south seismic zone. Xialatuo and
Xiaguan stations belong to this block. With small southward slip, the north part of the block (Xia-
latuo station) has an explicit eastward motion due to the push of the west Xizang block. This re-
sults in an apparent dextral motion. The other block, which includes Kashi, Hotan and Ruoqiang

400
SCIENCE IN CHINA (Series D)
Vol. 43
stations, is Tarim block in the south of Xinjiang block. In this block, on one hand, northward mo-
tion of its west part (Kashi and Hotan stations) is much greater than that of its east part (Ruoqiang
station) caused by the Pamirs’ northward push in the Indian plate; on the other hand, this block
penetrates into the neighboring blocks like a wedge. Its motion is different from the eastward mo-
tion of the surrounding blocks. It is the only block without eastward motion in the mainland of
China.
Acknowledgements This work was supported by the National Climbing Project of China (Grant No. 970231003), the
National Important Basic Research Project (Grant No. G1998040703) and the Major Project for Basic Research of the Chinese
Academy of Sciences (Grant No. KJ951-1-304).
References
1. Editorial Board of “Atlas of Lithospheric Dynamics in China” (eds.), Introduction of Lithospheric Dynamics in China (in
Chinese), Beijing: Seismological Press, 1991.
2. Avouac, J. P., Tapponnier, P., Bai, M. et al., Active thrusting and folding along the Northern Tien Shan and Late Cenzoic
rotation of Tarim relative to Dzungaria and Kazakhotan, J. Geophys. Res., 1993: 6755.
3. Bilham, R., Larson, K., Freymueller, J., GPS measurements of present-day convergence across the Nepal Himalaya, Na-
ture, 1997, 386: 61.
4. Holt, W. E., Ni, J. F., Wallace, T. C. et al., The active tectonics of the Eastern Himalayan syntaxis and surrounding regions,
J. Geophys. Res., 1991, 96 (B9): 14595.
5. Molnar, P., Caen, H. L., Fault plane solutions of earthquakes and active tectonics of the Tibetan Plateau and its margins,
Geophys. J. Int., 1989, 123.
6. Dong, D.N., Bock, Y., GPS network analysis with phase ambiguity resolutions applied to crustal deformation studies in
California, J. Geophys. Res., 1989, 94: 3949.
7. Zhu Wenyao, Cheng Zongyi, Xiong Yongqing et al., Preliminary results of measuring the crustal deformation in Qing-
hai-Xizang area using GPS technique, Science in China, Ser. D, 1998, 41(3): 314.
8. King, R. W., Bock, Y. (eds.), Documentation for the GAMIT GPS analysis software, release 9.6, Cambridge: Mass. Instit.
Tech., 1997.
9. Zhu Wenyao, Cheng Zongyi, Wang Xiaoya et al., Background of crustal movement of Chinese mainland, Chinese Science
Bulletin, 2000, 45(6): 557.
10. Cruttal Dynamics Project Preprint No.51, NASA/GSFC 1985.
11. Ye, S. H. , ed., Earth in MotionüüPresent-day Crustal Motion and Geodynamical Researches and Applications (in Chi-
nese), Changsha: Hunan Science and Technology Press, 1996.