-
GREATER NORTH CHINA INITIATIVE (GNCI): GREATER NORTH CHINA
INITIATIVE (GNCI): CENOZOIC GEODYNAMICS, CLIMATIC CENOZOIC
GEODYNAMICS, CLIMATIC
EVOLUTION, AND GEOLOGICAL HAZARDSEVOLUTION, AND GEOLOGICAL
HAZARDS
A whitepaper of scientific rationale and strategic plans for
cooperative research between the
IPACES and Chinese geosciences community
2005
-
GNCI Whitepaper - 2 -
List of the GNCI whitepaper drafting group
CHEN Yongshun Peking University FENG Xiahong Dartmouth College,
USA GE Shemin University of Colorado, USA LI Zhengziang University
of Western Australia, Australia LIU Mian University of
Missouri-Columbia, USA NIU Yaoling University of Durham, UK SHEN
Zhengkang State Seismological Bureau, China SONG Xiaodong
University of Illinois-Urbana Champaign, USA WANY Yang Florida
State University, USA WU Zhongliang State Seismological Bureau,
China YIN An University of California-Los Angles, USA ZHANG Youxue
University of Michigan-Ann Arbor, USA ZHAO Dapeng Ehime University,
Japan ZHAO Meixun Dartmouth College, USA Special Acknowledgement:
This document is produced from extensive discussion at a number of
NCP (North China Project) workshops. We thank the National Science
Foundation of China (NSFC) and Drs. Chai Yucheng, Yao Yupeng, and
Yu Sheng for their support and their enthusiasm in this endeavor.
Many colleagues in China contributed significantly to this document
in various aspects of its development. In particular, we thank Chen
Yong, Chen Xiaofei, Chen Bin, Gao Rui, He Jiankuan, Lu Huafu, Li
Yanxin, Liu Qiyuan, Liu Futian, Ren Jianye, Shi Yaolin, Wei Wenbao,
Wu Fuyuan, Wang Chunrong, Wang Lianshu, Wang Chengshan, Xu Yigang,
Xu Xiwei, Zan Shaoxian, Zhang Peizhen, Zheng Tianyu, Zhang Yueqiao,
Zhou Yaoqi for their valuable contributions.
-
GNCI Whitepaper - 3 -
Executive Summary The Great North China Initiative is a
multifaceted research plan aimed at a holistic understanding of the
Cenozoic evolution of the Earth systems in the Greater North China
region (GNC). The GNC is chosen because (1) it hosts the most vital
industrial, commercial, residential, and political centers of
China; (2) it is one of the most geologically active continental
regions in the world, with frequent and devastating earthquakes,
floods, draughts, dust storms and other natural hazards; (3) it has
a delicate ecosystem that is sensitive to the dynamic interplays
among lithosphere, hydrosphere, atmosphere, and biosphere; (4) it
is geologically the best studied regions in China where abundant
geological and geophysical data provide a firm foundation for the
proposed Earth system studies.
This research plan identifies some of the fundamental questions
regarding the Cenozoic geodynamics of the Earth systems in the GNC,
and outlines the required interdisciplinary approaches to address
these questions. One focus area of this research plan is active
crustal deformation and earthquakes – the GNC has the most active
intracontinental seismicity in the world. Another focus area is
asthenosphere-lithosphere interactions. In particular, effort is
called to understand the mantle processes responsible for the
thermal thinning of the GNC lithosphere since late Mesozoic, and
the diffuse intraplate igneous activity throughout the Cenozoic. A
third focus area is paleoclimate change in late Cenozoic and its
relationship with tectonics, paleoecology, and hydrology. All these
processes are integral parts of the system dynamics involving
mantle flow, lithosphere deformation, air and water circulation,
and bio-activity, hence requires cross-disciplinary approaches. In
the past few years, the IPACES (International Professionals for the
Advancement of Chinese Earth Sciences), with support from the
National Science Foundation of China (NSFC), has conducted
extensive discussions among its members and with scientists in
China through a number of workshops. We are convinced that the
Great North China Initiative (GNCI) will have a great impact on
sustainable development of China in the next decade, while the
cutting-edge integrated research in this plan will propel Chinese
Earth Sciences to an international leadership position.
-
GNCI Whitepaper - 4 -
Table of Content 1. Introduction
……………………………………………………………………….5
1.1. Why North China? …………………………………………………………...5 1.2. Integrated
Investigation of the GNC …………………………………………6
2. Lithospheric Deformation ………………………………………………………...8 2.1.
Geographic Division of the GNC …………………………………………….8 2.2. Tectonic
Development of GNC ………………………………………………8 2.3. Tectonic Boundary
Conditions of the GNC ………………………………….11 2.4. Active Deformation
…………………………………………………………..11 2.5. Major Questions………
…………..………………………………………….15 2.6. Recommendations
…….……………..……………………………………….16
3. Mantle Processes ……………………………………………………………….....17 3.1. Seismic
Velocity Structure …………………………………………………...17 3.2. Mesozoic to
Cenozoic Modification of Lithospheric Mantle ………………..19 3.3.
Cenozoic Volcanism in the Greater North China …………………………….21 3.4.
Previous Studies on Cenozoic Volcanism in the GNC ………………………24 3.5.
Major Questions ……………………………………………………………...28 3.6. Possible Research
Directions ...………………………………………………29
4. Climate, Water, and Environment ………………………………………………...34 4.1.
Introduction …………………………………………………………………..34 4.2. Paleoclimate
………………………………………………………………….35 4.3.
Paleoecology……………………………………………………………...…..37 4.4. Hydrological
Research ……………………………………………………….40
5. Integration ………………………………………………………………………...42 5.1. Climate
Change, Surface Processes, and Tectonics ………………………….42 5.2. Crustal
Stress, Earthquake Physics, and Hydrologic Processes …………...…43
6. Project Management ………………………………………………………...…….43
-
GNCI Whitepaper - 5 -
1. Introduction 1.1. Why North China?
The IPACES (International Professionals for the Advancement of
Chinese Earth Sciences – www.ipaces.org) posted two questions when
preparing this document: (1) What kind of research would have the
greatest impact on sustainable development of China in the next
decade? (2) What is the most effective research plan that would
propel Chinese Earth Sciences to an international leadership
position? After extensive discussion among the IPACES members and
with geoscientists in China, we identify the Greater North China
(GNC) region as the most promising target for addressing the above
questions. Our decision is based on the following.
(1) The GNC covers about one-quarter of China’s territory (Fig.
1) and hosts the most vital industrial, commercial, residential,
and political centers of the nation. This region is one of the most
populated areas in China. Understanding its geologic setting is a
prerequisite for making future strategic plans for the overall
sustainable development of China.
(2) The GNC is one of the most geologically dynamic settings in
the world, as indicated by the frequent occurrence of devastating
earthquakes, floods, draughts, and dust storms. These natural
hazards have enormous impacts on both the economy of China and the
quality of life in the region. Remediation of these hazards
requires a complete knowledge of geodynamic, climatic, and
land-surface processes as well as the interactions among them.
(3) The GNC has a very delicate ecosystem and is long known for
the lack of water resources. As the storitivity of any substantial
aquifers depends critically on the climatic conditions, it is
prudent to obtain paleoclimatic data that allow establishment of
quantitative models to make specific predictions on the trend of
climate variations at centennial to millennial scales. The models
can provide a better guide for future planning of large urban
centers whose survival will critically depend on the availability
of water resources.
(4) GNC is a key area for testing many prominent geological
hypotheses ranging from the nature of continental deformation to
interactions between tectonics and climate changes. Because of
this, GNC has become the focus of many international research
groups in recent years. This provides a unique opportunity to the
Chinese Earth Science community to showcase their scientific
achievements in the international arena.
(5) A large amount of geological and geophysical data in the GNC
region has accumulated over the past century. These data sets give
the Chinese Earth scientists a unique advantage as the
international geosciences community becomes increasingly more
interested in understanding diffuse continental deformation in Asia
and the coupled tectonics-climate processes.
(6) Both the IPACES and Chinese Earth Sciences community have
extensive expertise and research experience for the studies of GNC.
Many of them already have a strong track record in publications on
regional climate variations, tectonics, geochemistry, and
geophysics. With a coordinated integration of diverse research
areas and
-
GNCI Whitepaper - 6 -
approaches, the Chinese Earth Science community is poised to
take a prominent position in the regional studies of eastern Asian
geology.
Fig. 1. Geological Map of Asia. Red area outlines the Greater
North China region 1.2. Integrated Investigation of the GNC
In the past four decades, several major shifts of research
focuses have occurred in Earth Sciences. The advent of the
plate-tectonics theory in the 1960’s provides a simple kinematic
description of lithospheric deformation that is most applicable to
the oceanic regions [e.g., Morgan and McKenzie, 1965]. Research in
the 1970’s and early 1980’s had focused on testing predictions of
plate tectonics [e.g., Ernst, 1971; Dickinson, 1973] and applying
the theory to reconstructing the history of ancient mountain belts
[e.g., Dewey and Burke, 1973; Sengor, 1985]. Since the mid-1980’s,
research on continental dynamics has flourished, which is mainly
stimulated by the observation that active deformation over
continental regions of Eurasia and western United States is
distributed over broad zones (1000-2000 km wide) and located far
away (> 1000 km) from the nearest plate boundaries [Molnar and
Tapponnier, 1975; Proffett, 1977]. This new realization has led to
the on-going debate on whether continental deformation can be
described kinematically by the interaction of a few rigid blocks
following the principles of plate tectonics [e.g., Tapponnier et
al., 1986; Avouac and Tapponnier, 1993] or dynamically by continuum
thin sheets [e.g., Bird and Piper, 1980; England and Houseman,
1986; Royden, 1996]. Although some compromise was reached by the
later models that consider the role of faults or localized shear
zones in the continuum approximation [e.g., Willett and Beaumont,
1994; Kong and Bird, 1996], these early workers
-
GNCI Whitepaper - 7 -
treated the asthenosphere as passive and frictionless media
below deforming continental lithosphere. While some researchers
were well aware of the important effect of continental deformation
on regional and global climate changes [Kutzbach et al., 1989;
Harrison et al., 1992; Molnar et al., 1993], the reverse role of
climatic conditions in controlling the strain distribution and
exhumation history of major orogenic systems and continental
collision zones were not widely appreciated until very recently
[e.g., Beaumont et al., 2001; Willett et al., 2001]. In addition,
the role of asthenospheric flow in continental deformation has just
began to be explored [e.g., Liu et al., 2004]. The studies of
continental tectonics in the past two decades have resulted in a
detailed understanding of particle paths (P-T-t) within zone of
continental deformation [e.g., Harrison et al., 1998]. The nature
of the paths was mostly attributed to the kinematic nature of
deformation [e.g., thrusts vs. normal faults or extension vs.
contraction] [e.g., Spear, 1993], with little or no concerns on
surface processes that are responsible for exhuming the deep
crustal rocks [e.g., England and Houseman, 1986]. In addition, most
numerical models of continental tectonics consider only the
boundary conditions on the sides of the plates, leaving the base
and top undefined [cf. England and Houseman, 1989]. Recent advances
in Quaternary geochronology, satellite images, and digital
topography have revolutionized our views on the rates and physical
mechanisms of surface processes that shape the Earth’s landscape
via erosion and surface transport. Also, high-resolution seismic
tomography and improved numerical models incorporating realistic
features of the Earth have made the probing of deep-mantle dynamics
possible.
The unprecedented capability available to Earth scientists to
observe Earth’s surface and deep-mantle processes provides a new
challenge and a unique opportunity in treating the
asthenosphere-lithosphere-atmosphere as a unified and interactive
system [e.g., Harrison et al., 1992; Molnar et al., 1993]. This is
arguable the most exciting frontier in Earth Sciences as it lies at
the interfaces of several seemingly unrelated disciplines in the
past. We believe that the GNC is an ideal place to explore such an
interactive system by investigating the Mesozoic and Cenozoic
asthenospheric flow pattern, lithospheric deformation history, and
surface processes via systematic and detailed geologic,
geophysical, and geochemical studies. For example, the
asthenospheric flow and its interaction with lithosphere may be
established by the studies of seismic topography and igneous
activity. Lithospheric deformation can be determined by integrated
research incorporating structural geology, geochronology,
quantitative metamorphic petrology, and sedimentology. Finally
surface processes can be investigated by examining the variation of
climatic conditions at different time scales and its relationships
to tectonics and exhumation in shaping Earth’s surface.
Some of the unique geologic features in the GNC also make the
proposed integrated system approach feasible. For example, the
Loess Plateau is a direct result of Tibetan-plateau uplift created
by Indo-Asian collision [An et al., 2001; Guo et al., 2002]. Its
late Cenozoic deposits have provided one of the most complete
records of climate change as a result of lithospheric deformation
of the Indo-Asian collision zone. Widespread Cenozoic deformation
in the GNC and their close association with igneous activity
provide tractable clues in linking lithospheric deformation with
asthenospheric processes. In the following, we propose a road map
for deriving a quantitative and more holistic understanding of the
interplay among Cenozoic lithospheric deformation, asthenospheric
flow, and atmospheric circulation in the
-
GNCI Whitepaper - 8 -
context of the Greater North China. Reaching this goal requires
closely integrated multidisciplinary studies including continental
geodynamics, igneous activity, and climatic history that we discuss
below. 2. Lithospheric Deformation 2.1. Geographic Division of the
GNC
The geographically defined GNC is a large triangular region that
has an east-west width 1500-2500 km and a north-south length of
~2000-2500 km. The area is bounded in the south by the
east-trending late Triassic Qilian-Qinling-Dabie orogenic belt
created by collision between North and South China Blocks and the
Yangtze River delta (Fig. 1). The western boundary of the GNC
follows the north-flowing Yellow River along the Yinchuan Valley in
the south and the NNE-trending Greater Hinggan Range (i.e.,
Daxing’an Ling) that straddles between Inner Mongolia and
Heilongjiang Province. The GNC is bounded in t he north by the
east-flowing Heilongjing (i.e., Amor River) along the Sino-Russia
and Sino-Mongolia border and in the east by the eastern Asian
margin off the coast of eastern China and Korea north of the
Yangtze River delta (Fig. 1). The major geographic provinces in the
region include the Ordos Plateau with an average elevation of
~2000-2500 m in the west and the low-altitude Huabei and Songliao
Basins with an average elevation of ~50-200 m in the east. The
Huabei and Songliao basins are the largest contiguous flat areas of
China and homes of some of the most vital political, agricultural
and industrial centers of the nation. The topographic boundary
between the 2-km high Ordos plateau and the low-altitude Huabei
basin is abrupt along the eastern flank of the Taihang Shan (Fig.
1). This front is expressed by the largest gravity anomaly in east
Asia [e.g., Ren et al., 1981; Ma, 1985]. The time and mechanism of
this abrupt topographic division between the Orodos and Huabei has
never been explored. It is possible that the boundary was developed
during Eocene extension of the Huabei basin in the backarc region
of the Japan trench [e.g., Yin and Chen, 2004]. It is also possible
that the front was developed during the Jurassic or Cretaceous
topographic collapse of a large plateau behind the Qinglin-Dabie
orogenic belt [Yin and Nie, 1996].
2.2. Tectonic Development of GNC
Much of the basement rocks of the GNC belong to the North China
Craton (NCC), part of the Sino-Korean craton with some of the
oldest crust on Earth [Liu et al., 1992]. The NCC is believed to
have been a coherent craton by at least ca. 1800 Ma as indicated by
the metamorphic history of the orogenic belts joining the different
Archaean blocks [e.g., Zhao, 2001; Kusky and Li, 2003; Zhai and
Liu, 2003], the ca. 1770 Ma non-metamorphosed mafic dykes that cut
the orogenic belts [Halls et al., 2000; Wang et al., 2004], and
similar aged volcanic rocks in the cover successions [S. Zhang and
Z.X. Li, unpublished SHRIMP results]. The interior of the NCC
remained tectonically stable from ca. 1800 Ma until the end of the
Paleozoic, as illustrated by the widespread, conformable or
disconformable, shallow-marine clastic and carbonate sedimentary
successions over much of the 1500 My [Wang et al., 1985].
Ordovician-age diamondiferous kimberlite pipes in eastern NCC
indicate that the lithosphere there was no less than 180 km thick
at that time [e.g., Griffin et al., 1992].
-
GNCI Whitepaper - 9 -
The lithosphere of NCC experienced significant thermal thinning
during late Mesozoic to Cenozoic, associated with widespread
rifting and volcanism. The abnormally thin lithosphere is clearly
indicated by geological and geophysical data. Petrological and
geochemical probing using xenolith from the upper mantle carried up
by volcanic extrusions indicates that the lithosphere is no more
than 80 km over much of eastern NCC, and is in places less than 60
km thick [e.g., Fan et al., 1993; Menzies et al., 1993; Griffin et
al., 1998; Xu, 2001]. Geophysical analyses also point to a much
thinner lithosphere in eastern NCC [e.g., Liu, 1987]. There has
been much debate regarding the mechanism of the thinning. These can
be summarised into three schools of thoughts: (1) orogenic related
models [e.g., Fan and Menzies, 1996; Xu, 2001; Bryant et al., 2004]
which suggest that Mesozoic orogenic root delamination/slab
break-off, or erosion of an orogenically weakened lithosphere,
caused the thinning, (2) Pacific rollback models [e.g., Ren et al.,
2002; Northrup et al., 1995] involving ocean-ward migration of the
western Pacific active plate margin, and (3) models involving the
indentation of India with Eurasia [e.g., Menzies et al., 1993; Liu
et al., 2004].
Cenozoic deformation in northern China south of latitude 40ºN is
expressed by the development of the Paleogene Huabei basin and
Neogene rift systems around the Ordos block [Ye et al., 1987; Zhang
et al., 1998]. The Paleogene Huabei Basin formed by early Tertiary
back-arc extension associated with Late Paleocene and Eocene
basaltic eruptions [Ye et al., 1987]. Due to post-rifting thermal
subsidence, the basin is largely covered by Neogene to Quaternary
sediments [Ye et al., 1987].
There are three major graben systems around the Ordos block: the
Yinchuan rift, the Hetao rift, and the Shan rift (Fig. 1). The
Yinchuan rift has been assigned to initiate in the Oligocene
because of the presence of Oligocene red beds in the rift basin [Ye
et al, 1987]. However, a close examination of seismic reflection
profiles across the rift [Ningxia BGMR, 1989] suggests that the
syn-rift sediments are late Miocene and Pliocene in age. Similarly,
the southernmost part of the Shanxi rift basin [i.e., the Weihe
graben of Ye et al., 1987] contains Oligocene and possible Late
Eocene strata. Because Paleogene strata are also widely distributed
outside the southern segment of the Shanxi rift [Wang et al.,
1996], it is possible that the inferred Paleogene initiation of
rifting by Zhang et al. [1999] was due to assigning the pre-rift
sequence to syn-rift sequence. The Yinchuan and Shanxi rifts
terminate in the south at the left-slip Haiyuan and Qinling fault
zones [Burchfiel et al., 1991; Zhang et al., 1998] that extends
eastward to the Dabei Shan region [Ratschbacher et al., 2000]. The
Hetao graben is the northern extension of the Yinchuan graben [Ye
et al., 1987]. However, how the Hetao and Shanxi rifts terminate in
the north is not clear. GPS studies in this region show a broad
left-slip shear zone trending east-west separating the Hetao and
Shanxi grabens in the south and the stable Amurian plate to the
north [Shen et al., 2000]. The left-slip Qinling fault that
terminates the Shanxi rift in the south may be linked with the
east-trending left-slip Kunlun fault in central Tibet via a series
of north-trending faults at the juncture of the westernmost Qinling
and the Kunlun Mountains [Yin, 2000; Yin and Harrison, 2000].
The Tanlu fault zone bounds the eastern edge of the modern
Huabei Basin and is a first order Cenozoic tectonic feature in East
Asia (Fig. 1). Its Cenozoic development may have been related to
opening of Bohai Bay during Paleogene back-arc extension [e.g.,
Allen et al., 1997; Ren et al., 2002a]. Zhang et al. [1999] showed
that the southern segment of the Tanlu fault experienced three
phases of deformation: (1) normal-slip, and (2) left-slip with a
normal
-
GNCI Whitepaper - 10 -
component, and (3) right-slip with a minor normal component. GPS
survey shows that the fault is extensional accommodating east-west
extension [Shen et al., 2000].
To the east of the continental GNC are several large back-arc
basins (Sea of Okhotsk, Japan Sea) and highly extended continental
margins (Bohai Bay, East China Sea) (Fig. 1). Their tectonic
development was closely related to that in the continental GNC, and
the opening of the marginal seas off the east coast of the GNC and
along other part of east Asian continent may have played important
roles in change of ocean currents and consequently the climate.
The Japan Sea (also known as East Korea Sea) opened in the Late
Paleocene and Eocene [Lallemand and Jolivet, 1986; Celaya and
McCabe, 1987]. However, the oceanic crust in the basin was not
created until Late Oligocene and lasted to the end of the Early
Miocene (30-15 Ma). The Japan Sea is bounded in the northeast by
the right-slip Sakhalin fault that may have accommodated about 400
km of displacement during the opening of the basin [Jolivet et al.,
1994]. There are three models for the origin of the Japan Sea.
Lallemand and Jolivet [1986] suggest that the Japan Sea was
developed as a pull-apart basin between two right-slip faults. Yue
and Liou [1999] proposed that the opening of the Japan Sea was
related to the development of the left-slip Altyn Tagh fault in
northern Tibet. Both models consider the development of the Japan
Sea was associated with the Indo-Asian collision. Alternatively,
the opening of the Japan Sea was attributed to back-arc extension
[Jurdy, 1979; Celaya and McCabe, 1987] caused by slow convergence
between Eurasia and Pacific plate [Northrup et al., 1995].
The Bohai Bay extensional domain is separated from Japan Sea by
the Korea peninsular that appears to have experienced little
Cenozoic extension (Fig. 1). This extensional system extends
southward to the East China Sea and westward to the Huabei Basin
[Zhao and Windley, 1990; Allen et al., 1997; Ren et al., 2002a].
Extension in the Bohai Bay region started in the Paleogene and was
most active between the Eocene and latest Oligocene [Allen et al.,
1997; Ren et al., 2002a]. Rift-related structures and sedimentary
sequences are overprinted by Quaternary dextral transpressional
deformation, causing inversion of some earlier normal faults [Allen
et al., 1997].
The continental margin of East China Sea consists of three
tectonic zones: (1) the East China Sea extended continental shelf,
(2) the Taiwan-Sinzi folded zone, and (3) the Okinawa Trough (Fig.
1). The tectonic evolution of the East China Sea has been
summarized by Zhou et al. [1989] and Kong et al. [2000]. Between
the latest Cretaceous and earliest Paleocene, extension occurred in
the East China Sea as expressed by the development of detachment
faults. This extensional event is part of widely distributed
extension in east Asia [Ren et al., 2002a]. Between the Late
Paleocene and Early Oligocene, extension was focused in the East
China Sea region. Significant crustal thinning during this period
was manifested by the development of normal faults, development of
a narrow basin, and rapid subsidence associated with the basin
formation [Kong et al., 2000]. Contraction began in the central and
northern Taiwan-Sinzi folded zone in the middle Oligocene [Kong et
al., 2000] and was significantly intensified in the late Middle
Miocene [Ren et al., 2002b]. This event may have been associated
with the subduction of the Palau-Kyushu ridge on the Philippine
plate [Kong et al., 2000]. Due to very thick sequence of syn- and
post-rift sediments, the age of the southern Taiwan-Sinzi folded
zone is poorly constrained. Kong et al. [2000] suggest that the
-
GNCI Whitepaper - 11 -
southern Taiwan-Sinzi folded zone initiated in the Late Miocene,
possibly related to collision between the Luzon arc and Eurasia. In
contrast, Sibuet et al. [2002] propose that the Taiwan-Sinzi folded
belt terminated its development at ~15 Ma during a major plate
reorganization at the junction of the Philippine Sea plate and
South China Sea. The development of the Okinawa Trough is the
youngest deformation event in the East China Sea. Its opening may
have started in the Late Miocene associated with clockwise rotation
of the Ryukyu arc [Sibuet et al., 1998]. 2.3. Tectonic Boundary
Conditions of the GNC
Understanding the Cenozoic deformation history of the GNC
requires the knowledge of the history of boundary conditions around
the region. The area north of the GNC is an active zone of
distributed deformation between the southern edge of the Siberia
Craton as marked by the Baikal rift and the Sino-Mongolia border.
This zone is dominated by extensional and strike-slip faults (Fig.
1). South of the GNC is the relatively stable South China Craton
and the northern margin of the Tibetan plateau. The latter is
marked by east-trending left-slip and thrust faults (Fig. 1).
Because the diffuse nature of Cenozoic deformation north and south
of the GNC, the exact kinematic history of the boundaries can only
be established by systematic investigation of the deformation
histories in the two regions.
The evolution of the margin seas along the eastern margin of
Asia is also poorly understood. The key issue is the history of the
Philippine plate motion over the Cenozoic. It remains uncertain
about the past position of various plates over the western Pacific
in the Cenozoic. This uncertainty presents a great challenge in
using plate-tectonic boundary conditions to investigate the
deformation history of the GNC region in a forward fashion, but at
the same time it offers an opportunity to use land-based geology to
inversely determine the possible plate boundary history such as
strike-slip vs. subduction, Kula vs. Philippine plate
reconstructions in the western Pacific [Yin and Chen, 2004].
2.4. Active Deformation Active tectonics of GNC is expressed by
the frequent occurrence of large earthquakes. Seismicity in GNC
tends to concentrate along the rims of the Huabei basin such as in
the Shanxi rift system in the west, the Shanhaiguan uplift in the
north, and the Tanlu fault zone in the east (Fig. 2). The GNC
region has recorded some of the deadliest earthquakes in human
history including the 1556 M8 Huaxian in Shanxi with a death toll
of 830,000 and the 1976 M7.8 Tangshan in Hebei with a death toll of
~250,000. During the last major burst of seismicity in GNC, eight
earthquakes with magnitude >M6.5 occurred between 1966 and 1976.
Because GNC is the home of nation’s capital Beijing, which is the
site of the 2008 Summer Olympic Games, it is urgently important to
have a systematic assessment of earthquake potentials and their
possible impacts in the major urban areas of northern China. Before
discussing this issue in detail, we first outline the general
progress in Chinese earthquake studies in recent decades.
Since 1960s, Chinese seismologists have done extensive work in
GNC on lithosphere structure, earthquake processes, seismic stress
field, as well as studies on earthquake precursors and earthquake
prediction. During the past 20 years, Chinese geologists have
compiled a “Map of Active Tectonics of China” at a scale of 1:4
million, in which more than
-
GNCI Whitepaper - 12 -
200 active tectonic zones have been identified [Deng et al.,
2003]. This map delineates the active tectonic belts of China that
bound relatively aseismic blocks by their slip rates that range
from
-
GNCI Whitepaper - 13 -
is an order of magnitude lower than that along the
block-bounding faults and that M=7 earthquakes generally occurred
along the block boundaries.
Paleoseismic studies of the GNC region have been mostly
concentrated on the rim of the Ordos plateau. Because the plateau
is relatively aseismic, many Chinese geologists refer to it as the
Ordos block. Along the northern edge of the Ordos block, Ran et al.
[2003a] found that 62 paleo-earthquakes have occurred in the late
Quaternary, of which 33 occurred in the Holocene. They also
discovered that the recurrence intervals of major earthquakes
differ for individual fault segments, individual faults, and
composite fault zones. One of the major active faults along the
northern edge of the Ordos block is the 220-km long Daqingshan
normal fault zone. It bounds the NW-trending Hetao rift system
along the northwestern edge of the Ordos plateau [Ran et al.,
2003b]. This fault initiated in the Eocene with a total slip >
2.4 km since the Quaternary. A paleoseismic study shows that 7
major paleoseismic events occurred on the fault since 19 ka BP.
They occurred at 18.75 ± 0.75 ka, 16.97 ± 0.96 ka, 14.65 ± 0.67 ka,
11.82 ± 0.69 ka, 9.45 ± 0.26 ka, 6.83 ± 0.26 ka, and 4.50 ± 0.23 ka
BP, with an average recurrence interval at 2.375 +/- 0.432 ka.
The NNE-trending Helan Shan-Yinchuan fault zone marks the
western edge of the Ordos Plateau and GNC. Deng and Liao [1996]
show that this is a normal-right-slip fault cutting late
Pleistocene and Holocene alluvial fans and offseting the Great Wall
of the Ming Dynasty built at ~400 years B.P. right-laterally for
1.45 m and vertically for 0.95 m. The offset event may result from
the M = 8 Yinchuan-Pingluo earthquake of 1739. On the basis of
terrace offset, scarp morphology, and paleoseismic trenching across
the fault, Deng and Liao [1996] conclude that four large
earthquakes with M = 8 occurred along the fault at 8400, 4600-6300
(or 5700), 2600, and 256 years B.P., with a recurrence interval of
these earthquakes is 2300-3000 years. Paleoseimic studies were also
conducted along the southern edge of the GNC region. Along the
Luoshan Fault zone at the boundary between the GNC and NE margin of
the Tibetan plateau [Wei et al., 2003]. The fault strikes N-S and
is reverse right-slip and has a minimum slip rate of 2.15 ± 0.2
mm/yr. Four recent events have occurred on the fault zone: after
8200 +/- 600 years BP, between 3130 ± 240 years BP, at 4150 ± 120
year BP, and before 2230 ± 170 years BP.
Fig. 2. GPS site velocities relative to stable Eurasia. Data are
from the CMONOC network [Wang et al., 2003]
-
GNCI Whitepaper - 14 -
GPS. The application of GPS technology has made it possible to
observe continuous deformation of the crust, providing
unprecedented constraints for understanding the deformation of
continents and the genesis and rupture processes of earthquakes.
The combination of these observations with physics-based dynamic
models and the ever advancing computational technology now allows
us to quantitatively interpret earthquake nucleation processes and
earthquake rupture processes and predict propagation of seismic
waves in complex media, simulation of strong ground motions, and
earthquake hazards caused by strong ground motions. These
breakthroughs not only represent major scientific achievements but
also provide vital information for engineers, policy makers, and
the general public. At the moment, we cannot predict earthquakes in
a classical sense and we cannot predict major breakthroughs leading
us to that goal. However, these new advances in seismology and
crustal deformation have led us to a new stage of seismic hazard
mitigation and will continue to provide us scientific bases for
understanding earthquake phenomena and predictability.
Contemporary crustal deformation in North China has been
measured since the 1950s. Early horizontal measurements were made
mainly using triangulation, which however could not be used for
precise surveying of tectonic deformation except for measuring
coseismic displacements of large earthquakes [Huang, 1980]. Until
the emergence of GPS technique in the early 1990s, for several
decades leveling was the only effective means to detect tectonic
deformation. The very first GPS network in North China was
established in 1992, and expanded in 1995 to become the North
China/Capital Circle Network [Li et al., 1995]. Using data from
repeated surveys of this network, Shen et al. [2000] determined
that the region moved about 3-8 mm/yr ESE with respect to the
Eurasia plate. These results were confirmed by subsequent studies
[Wang et al., 2001; Yang et al., 2002]. Monitoring of crustal
deformation in North China was significantly improved after the
founding of the Crustal Motion Observation Network of China
(CMONOC) in 1998. About 300 CMONOC survey mode GPS stations are
located in North China covering effectively all the known regional
active faults, and have been surveyed in 1999, 2001, and 2004
respectively. Using GPS data from the North China/Capital Circle
and the CMONOC networks many models have been proposed to quantify
tectonic block motions in North China [Xu et al., 2002; Li et al.,
2003; Yang et al., 2003; Huang et al., 2003]. Wang et al. [2003]
analyzed the CMONOC data and found that the deformation field
across the Zhangjiakou-Penglai seismic zone for the 1999-2001 time
period was quite consistent with that of the 1992-1996 period.
Deformation across the northern Shanxi Rift zone, however, showed
insignificant slip for the 1999-2001 time period, different from
the result obtained earlier for the 1992-1996 time period across
the same segment of the fault. Such a discrepancy raises a
question: is the rifting process across the Shanxi Rift varying
with time [He et al., 2003], and if so, what is the cause of it?
Interpolation of the CMONOC velocity field revealed dextral shear
motion trending NNE in North China, at a rate of 2-4×10-8/yr for
most part of the region except the area southwest of the Ordos
plateau where the shear strain rate is up to 5-7×10-8/yr [Shen et
al., 2003]. Liu and Yang [2005] derived the strain rates in north
China using the updated Chinese GPS data. The high strains are
found in the North China plain and around the Ordos block. Using
the observed crustal kinematics as boundary conditions, they have
shown in a 3D finite element model that the long-term distribution
of high strain energy in the North
-
GNCI Whitepaper - 15 -
China plain and circum-Ordos rifts are consistent with seismic
moment release in the past ~2000 years. Thus the intense seismicity
in North China is not a transient phenomenon but reflect long-term
strain energy accumulation and release resulting from crustal
tectonics (Fig. 3).
Fig. 3. Left panel: GPS velocity (relative to stable Eurasia)
and strain rates (background). Middle panel: predicted long-term
strain energy (background) and stress states represented in
sterographic lower-hemisphere projections. Right panel: seismic
strain energy released in the past ~2000 years [Liu and Yang,
2005]. 2.5. Major Questions To better understand active tectonics
and its control on earthquakes in the GNC region, multidisciplinary
efforts are needed to address the following scientific issues. Some
of the possible research directions and important questions are
outlined below. (1) Structural and geological factors that give
rise to intraplate earthquakes in north China.
• How are the intraplate earthquakes controlled by the far-field
boundaries conditions such as India-Eurasia collision in the south
Pacific subduction in the east? What dolithospheric structures
control the spatial pattern of seismicity?
• Are the Cenozoic volcanism and active tectonics related and
how? What is the role of igneous activity in creating lithospheric
stress?
(2) Characterization of GNC earthquake sources, faults, and
structure of the source region
• What are the characteristics of the NC earthquakes? How are
the earthquakes distributed with spatially and with depth? Do the
focal mechanisms change with depth? Are the earthquakes in NC
fundamentally different from those in other regions, such as in the
stress drops? What are the important features of the rupture
processes of major NC earthquakes? What are the frictional features
of the NC earthquake faulting?
Major tasks in this endeavor include (1) precise determination
of earthquake locations, mechanisms, and rupture process; (2)
mapping of active fault; (3) 3-D crustal structure around the
active faults; and (4) basic understanding of the physics of
faulting.
-
GNCI Whitepaper - 16 -
(3) Present-day crustal deformation and strain field • What is
the present-day deformation (strain field) in NC? How is the
deformation
related to geological structures? Why the present day
deformation is mostly NNE dextral shear? How does it change from
extension in the past to simple shear? Does deformation change with
depth? How does the crustal deformation evolve with time? Why is
the present-day deformation concentrated mainly in the area around
the Ordos block and along the Zhang Jia Kou – Peng Lai seismic
belt?
(4) Models of crustal deformation • How does NC continental
lithosphere deform? Can the model of “Active Tectonic
Blocks” be applied to NC? What is the rheology profile of the
lithosphere and how does it change laterally? How does the rheology
depends on composition and thermal structure? What's the thermal
profiles of the lithosphere of NC? How does it relate to reology of
the lithosphere? What's the visco-elastic deformation style? Can we
build a consistent dynamic model of deformation of the region?
• What is the role of mid crust and crust-mantle coupling in
controlling the crustal deformation and earthquake genesis? What is
the lithospheric strength profile? What's the nature of low
velocity in midcrust? Does it relate to earthquake genesis?
• What is the nature and degree of coupling between crust and
mantle? (5) Genesis of intraplate earthquakes
• What controls the seismic genesis of NC and Ordos? What
controls the limit of the depth of the earthquakes? Is there
systematic difference between earthquakes in the grabens and in the
shear zones? How are earthquakes controlled by the properties of
faults, such as strength, segmentation, curvature, and shape?
• How does stress evolve with time and space? How do faults
interact with one another? What are the main factors triggering
earthquakes (static or dynamic Coloumb stress change or other
factors)? Why do present-day major earthquake concentrate on the
Hetao–Zhangjiakou–Peng Lai seismic belt, and why is there no modern
major earthquakes in the Fei-Wei seismic zone? Why in the Zhang Jia
Kou – Peng Lai seismic belt, which is in the ESE direction, the
earthquake faulting is usually along NNE direction?
(6) Earthquake hazard mitigation and earthquake prediction • A
major application of this initiatives is the seismic hazard
assessment and strong
ground motion simulation in NC. Given our understanding of the
Earth structure in NC and earthquake sources and crustal
deformation, can we achieve realistic strong ground motion
simulation? How do the NC basins affect strong ground motions? Can
the seismic zoning be made more accurate? Can we gain better
understanding of long- and short- term earthquake predictions (one
of the most fundamental goals of seismology)?
2.6. Recommendations (1) Establishing a National Quaternary
Geochronological Center. An important
aspect of paleoseismic study is to determine the temporal
evolution of major earthquakes and their evolution over space. For
example, recent clustering of major earthquakes in eastern
California of the United States, Turkey, the Kunlun Range of the
central Qinghai-Xizang (Tibet) plateau, and Sumatra of Indonesia
show
-
GNCI Whitepaper - 17 -
remarkable spatial migration of earthquakes over a few years.
Determining the pattern of such clustering require more systematic
studies of fault zones. However, as urban development has covered
more and more land surfaces, trenching across some of the major
faults may not be possible. This will require other means to
determine the reoccurrence interval of major earthquakes such as
determining the morphologic surfaces. This would require the
application of a suite of analytical techniques such as the U-Th,
thermal and optical luminescence, cosmogenic, and the traditional
14C dating methods.
(2) Blind structures in the urban areas of the GNC. A major
change in the focus of paleoseismic studies occurred in the United
States in the late 1980’s. Previously, research was mostly
concentrated on faults that are capable of generating large
earthquakes such as the San Andreas fault in southern California.
However, the mid-80’s and early 90’s, two moderate earthquakes
struck the Los Angeles region, the 1987 Whittier Narrow earthquake
(Mw 5.9) and 1994 Northridge (Mw 6.7) earthquake. The latter caused
a total economic damage of ~ US$40 billion. From our brief review
above, we note that the paleoseimic studies in the GNC region have
mostly focused on major block-bounding faults. However, some of the
minor faults within the “stable” blocks may be capable of
generating moderate size earthquakes. With a favorable condition,
the possible moderate-sized earthquakes could produce more
devastating effects on human lives and regional economy than the
large faults located remotely from the major urban centers.
3. Mantle Processes 3.1. Seismic Velocity Structure Seismology
is fundamentally an observational science. The Earth's major
internal layers (the solid inner core, the fluid outer core, the
lower and upper mantle, and the crust) were rapidly discovered in
the first half of the 20th century following the development of
modern seismometers in late 19th century. Systematic earthquake
location and determination of earthquake mechanisms played a
central role in formation of the plate tectonic theory in the late
1960s, following the deployment of the World-Wide Seismic Network
in early 1960s. Accumulation of seismic data made it possible to
conduct seismic tomography that started in 1970s and 1980s, which
has formed a key to the understanding of the dynamics of Earth’s
interior. The availability of modern high precision broadband
digital seismometers is now making it possible to image delicate,
yet vital features, such as, the plume structure, ultra low
velocity structure in the lowermost mantle (D” region), and the
layering of the solid inner core, and to detect temporal changes of
earth’s structure or sources, such as the rotation of the inner
core, the healing of the ruptured fault zones, and the plumbing of
magma chambers. The availability of modern broadband portable
digital seismic stations since the early 1990s has turned the
nature into great field laboratories for seismologists and
tectonicists. The observation and interpretation of these “field
seismological laboratory” works have greatly increased seismic
resolution of lithosphere structure and earthquake rupture
processes and are playing increasingly important role in solving
key geological and geodynamic problems.
-
GNCI Whitepaper - 18 -
Another major effort in the GNC region is deep seismic sounding
(DSS) using active sources to image the crustal structure [e.g.
Teng et al., 1979; Li and Mooney, 1998; Zhang et al., 1999; Li et
al., 2001]. The coverage of the DSS profiles is particularly dense
in North China [e.g., Duan et al., 2002]. These results show that
the crust of North China thickens gradually from east to west, with
its averaged thickness of about 35 km.
In recent year, a number of surface wave dispersion and
inversion studies have been done, covering a large area of east
Asia, using data from global stations of the Chinese Digital
Seismic Network (CDSN) and other stations in adjacent regions
[e.g., Wu et al., 1997; Ritzwoller and Levshin, 1998; Curtis et
al., 1998; Xu et al., 2000; Zhu et al., 2002, Huang et al., 2003a;
Lebedev and Nolet, 2003]. Body-wave studies on NC area from the
western countries have mostly been limited, including body wave
modeling [Beckers et al., 1994], propagation of Pn, Sn and Lg waves
[Xie, 2002; Rapine and Ni, 2003], receiver functions [Mangino et
al, 1999]. However, some high-resolution tomographic studies using
regional and local travel-time data have been carried out for the
whole country [e.g., Liu et al., 1990; Liu and Jin, 1993] and for a
local area in the north and east China [Liu et al., 1986; Xu et
al., 2001; Huang and Zhao, 2004], with most published in the
Chinese literature.
Another significant effort in last few years, particularly in
the last two years, is the use of the travel times of Pn waves to
invert for the velocity and anisotropy distribution in the
uppermost mantle and the crustal thickness of China [Song, 2004]
(Fig. 4). Tomographic inversions of Pn waves haven been conducted
in the whole country [Wang et al., 2002; Sun et al., 2004; Liang et
al., 2004; Hearn et al., 2004] as well as in local regions of
China, including the Tibetan plateau [Zhao and Xie, 1993; McNamara
et al., 1997], Xinjiang and the western China [Pei et al., 2002],
Southwest China (Sichuan-Yunnan region) [Huang et al., 2003b], the
northeastern margin of Qinghai-Tibetan plateau [Xu et al., 2003],
and the eastern and northern China Pn [Wang et al., 2003] and Sn
[Pei et al., 2004]. These studies have revealed significant
features of thin crustal thickness, lower upper mantle velocity,
consistent with upper upwelling and lithospheric thinning, which
may have controlled the genesis of the rich oil and gold deposits
of the region [Song et al., 2004].
Pei et al. [2005] perform inversion of P-wave arrival times of
both regional and
teleseismic earthquakes to obtain 3-dimensional P-wave seismic
velocity variations within the upper mantle below the GNC region.
The most important findings of their study are as follows. (1) No
fast P-wave velocity anomalies can be related to subducted oceanic
slabs
Fig. 4. Inversion results for Pn velocity (color) and anisotropy
(bars) in China (from Liang et al., 2004). The bar indicates the
fast Pn direction, and the length is proportional to the anisotropy
amplitude, saturated at 4%.
-
GNCI Whitepaper - 19 -
beneath the 660-km discontinuity; instead the subducted oceanic
slabs become flattened and stagnant along the transition zone. (2)
The western end of the stagnant oceanic slabs lying along the
660-km transition zone can be correlated with the prominent surface
topographic break in the GNC between the Ordos Plateau to the west
and the Hubei plain to the east along the NNE-trending Taihang Shan
Range (~105°E). The western end of the flat stagnant slabs is
located ~ 1500 km west of the active trench in the western Pacific.
(3) Slow P-wave velocity anomalies are present at depths of 100-250
km below the active volcanic arc and the stagnant slabs along the
660-km transition zone. A simple tectonic model is proposed to
explain our observations and their potential correlation to the
complex tectonic history of east Asia. In the model, vigorous
convection is operating within this horizontally expanded “mantle
wedge” above both the active subducting slab in the western Pacific
and the ceased stagnant slabs beneath much of the North China
plain. This horizontally expanded convection was probably resulted
from both rapid eastward migration of the western Pacific trench
system and the sinking of the Mesozoic and Cenozoic slabs now
trapped at the 660-km transition zone. Both the widespread Cenozoic
volcanism and associated extensional basins in east Asia may have
been the manifestation of this vigorous upper mantle convection
beneath the continental lithosphere. Finally the negative thermal
anomaly associated with the stagnant slabs along the 660-km
discontinuity has not only caused a broad depression of the
boundary due to its negative Clapeyron slope but also effectively
shielded the above asthenosphere and continental lithosphere from
any possible influence of mantle plumes originated from the lower
mantle.
Fig. 5 Map view of the Vp variations at depths of 120, 300, and
500 km [Pei et al., 2005].
3.2. Mesozoic to Cenozoic Modification of Lithospheric
Mantle
The lithosphere of the NCC is no longer as thick as 180 km or
more. Petrological and geochemical probing using xenolith from the
upper mantle carried up by volcanic extrusions indicates that the
lithosphere is no more than 80 km over much of eastern NCC, and is
in places less than 60 km thick [e.g., Fan et al., 1993; Menzies et
al., 1993; Griffin et al., 1998; Xu, 2001]. Geophysical analyses
also point to a much thinner lithosphere in eastern NCC [e.g., Liu,
1987]. Such a thin lithosphere is highly unusual for an old
continental craton like the NCC, and would have significant
bearings on modern seismicity.
Apart from the dramatic lithospheric thinning, the NCC also
suffered two episodes of orogenic modifications during the
Mesozoic. Major intro-cratonic structures produced by those events,
like the Tan-Lu fault, could directly influence the distribution of
epicentres. The first orogenic episode occurred during the Triassic
to mid-Jurassic (Indosinian), corresponding to the collision of the
NCC with the South China Block. The consumption of
-
GNCI Whitepaper - 20 -
over 200,000 km2 of continental crust had led to
ultra-high-pressure (UHP) metamorphism. The collisional event also,
for the first time since the formation of the NCC, caused major
tectonism and relative elevations in the interior of the NCC. It
modified the crustal and lithospheric architecture of the NCC in
three ways. First, as shown in some tectonic models, major
thick-skinned crustal thrusts of up to 15 km thick and hundreds of
kilometres in dimensions may have developed in south-eastern NCC
[Li, 1994, 1998]. This N to NNW-verging thrust system may extend to
western Shandong to the west and southern Liaoning to the north.
Minor, south-verging thick-skinned thrusting may also have occurred
along northern NCC. Second, major crustal/lithospheric faults that
cut cross the entire craton, e.g., the Tan-Lu Fault, were
developed. Third, the continental collision event may have started
the mantle erosion (lithospheric thinning) of the NCC [e.g.,
Menzies et al., 1993].
The second orogenic event was the so-called Yanshanian Orogeny
developed along northern NCC. It was probably related to the
closure of the Mongol-Okhotsk Ocean to the north. Although
thin-skinned thrusting was widespread along northern NCC [e.g.,
Davis et al. 1998], the shallow nature of such structures may not
have significant bearings on modern tectonics and seismicity.
Fig. 6. History of lithospheric thinning and regrowth in eastern
NCC [Xu, 2001].
The lithospheric thinning occurred during late Mesozoic to
Cenozoic. Presently active is occurring along the Shanxi rift
system only whereas the thinned lithosphere is growing back in
eastern NCC [for a review see Xu, 2001]. There has been much debate
regarding the mechanism of the thinning. These can be summarised
into three schools of thoughts: (1) orogenic related models [e.g.,
Fan and Menzies, 1996; Xu, 2001; Bryant et al., 2004] which suggest
that Mesozoic orogenic root delamination/slab break-off, or erosion
of an orogenically weakened lithosphere, caused the thinning, (2)
Pacific rollback models [e.g., Ren et al., 2002; Northrup et al.,
1995] involving ocean-ward migration of the western Pacific active
plate margin, and (3) models involving the indentation of India
with Eurasia [e.g., Menzies et al., 1993; Liu et al., 2004].
-
GNCI Whitepaper - 21 -
3.3. Cenozoic Volcanism in the Greater North China The plate
tectonics theory has provided a solid framework for understanding
the
distribution of volcanism along plate boundaries: (1) at
seafloor spreading centers where two plates pull apart, the ocean
crust is being continuously created by volcanism, and (2) at
convergent boundaries where the oceanic plate returns into the
Earth’s deep interior through subduction zones, volcanic arcs such
as the “Pacific ring fires” are being built. However, plate
tectonics theory, by its original definition, cannot explain
earthquakes and volcanic activities occurring within plate
interiors. Hotspots or deep-rooted mantle plumes have been widely
invoked to be responsible for “intra-plate” volcanism. Intraplate
volcanism is indeed widespread and is thus an important mode of
mantle melting. Many of the intraplate volcanic activities are
apparently associated with widely perceived mantle plumes/hotspots
such as the Hawaii, Samoa, Tahiti volcanic islands, but such
association is not clear in many other cases. For example, Cenozoic
volcanic activities widespread in eastern Australia [e.g., Johnson
ed., 1989], eastern China [e.g., Deng et al., 1998; Zhang et al.,
1998; Liu, 1999], western and central Europe [e.g., Wilson &
Patterson, 2001], the well-known Cameroon volcanic line straddling
the Atlantic-African passive continental margin [e.g., Fitton and
Dunlop, 1985; Halliday et al., 1988] and numerous seamounts
scattered throughout much of the Earth’s ocean floor [e.g., Batiza,
1982] away from plate boundaries cannot be readily explained by
either plate tectonics theory or mantle plume hypothesis. Mantle
source “wet spots” [e.g., Green & Falloon, 1998] may explain
some of the “intraplate” melting “anomalies”, but such mechanism
alone cannot account for the aforementioned widespread and large
scale volcanic activities.
It is possible that the melting anomalies of the kind may
reflect a mode of mantle thermal anomalies that are yet to be
established, or simply reflect mantle compositional anomalies. An
understanding of the origin of these melting anomalies is
fundamentally important because it will represent an advancement in
our knowledge on how the Earth works within or outside the
framework of the mature and widely accepted plate tectonics theory
and mantle plume hypothesis. Furthermore, intraplate volcanic
activities on land present a severe threat to human activities.
Hence, such studies have both scientific significance and practical
importance.
The GNC region is an ideal natural laboratory for examining
intra-plate volcanism. For example, the documented lithosphere
thinning [Menzies et al., 1993, Deng et al., 1996, 2004; Griffin et
al., 1998; Zheng, 1999; Wu and Sun, 1999; Xu, 2001; Gao et al.,
2002; Zhang et al., 2002; Yan et al., 2003] in the Mesozoic for an
otherwise stable craton is inexplicable with existing theories. The
widespread Mesozoic, and in particular, Cenozoic intraplate
volcanism requires mechanisms that are beyond the scope of plate
tectonics and mantle plumes hypothesis [Niu, 2005]. Only with
combined multidisciplinary expertise, experienced field geologists,
meticulous analysts, and creative thinking and modelling skills, is
it possible to achieve the objectives with success and to lead to
new advances towards understanding the working of the Earth within
the interior of the plates and eastern China tectonics in
particular.
While basaltic volcanism is ultimately caused by some sort of
thermal or compositional (including volatiles) anomalies in the
mantle, it is the volcanic products – the volcanic rocks that carry
the messages about their source materials, histories and detailed
processes. Hence,
-
GNCI Whitepaper - 22 -
detailed petrology and geochemistry of representative volcanic
rocks is the first step for more comprehensive studies in the
context of regional tectonic histories. Before we discuss this
aspect of research, we briefly outline the igneous history of the
GNC region below. Northeast China lies in Northeast mainland Asia
and straddles two tectonic units: (i) North China-Korea Craton and
[Powell and II] Mongol-Okhotsk Tectonic Belt. During late Jurassic
and early Cretaceous, there was widespread volcanism of evolved
magmas along Daxing’anling, mostly the calcalkaline series
andesite-dacite-rhyolite, with some trachyandesite-trachyte. The
widespread magmatism was part of the Yanshanian Orogeny in East
Asia. The exact nature of the Yanshanian Orogeny is still being
investigated, and one explanation is that it is due to subduction
of both the Pacific plate and the Mongolo-Okhotsk plate during the
early stage (160-130 Ma) and widespread intra-arc extension during
the late stage (130-110 Ma) [e.g., Yin and Harrison, 1996]. More
recent volcanism in Northeast China since about 80 Ma distinguishes
from earlier volcanic activities in composition: Recent volcanic
rocks are mostly basaltic (tholeiite, basanite and alkali olivine
basalt) with minor evolved trachytes and rhyolite. In late
Cretaceous, basaltic magmatism began to appear sporadically in
Shandong and Liaoning. Most Cenozoic volcanos line up in linear
belts on the sides of Northeast China Plain (Fig. 7). These belts
will be referred to as “East”, “West” and “North” Volcanic Belts
although they do not lie exactly on the east side, etc. Because
many volcanoes along a belt are roughly synchronous (within the
last 1.8 Ma), the belts do not seem to be hotspot tracks. Instead,
they are more likely rifts or “hot lines”. At present, the nearest
subduction zone (Japan Trench) is more than 1000 km away. Recent
tomographic studies indicate that the subducting Pacific slab
becomes stagnant in the mantle transition zone under Northeast Asia
[e.g., Fukao et al., 1992; Kárason and van der Hilst, 2000; Zhao,
2001a, 2004; Chen et al., 2005; Pei et al., 2005].
-
GNCI Whitepaper - 23 -
Fig. 7. Known Cenozoic volcanoes in Northeast China,
distinguished by their age. Overlapping symbols of different ages
mean that the volcano was active over these periods. Small black
dots mean volcanos of unknown age (but likely Cenozoic). Data for
China are from Liu [1999], Ma et al. [2002], and references
therein. For areas outside China, only Holocene volcanos are shown
[Simkin and Siebert, 1994]. Names of volcanic fields (clockwise
then center) are: PL: Penglai; QX: Qixia; LQ: Linqu; DT: Datong;
HNB: Hannuoba; JN: Jining; ABG: Abaga; KSKT: Keshiketeng; HLH:
Halaha; NMH: Nuominhe; KL: Keluo; WDLC: Wudalianchi; XK: Xunke; GS:
Geshan; GDS: Gedashan; SYS: Shuangyashan; JD: Jidong; FZ: Fangzhen;
SZ: Shangzhi; MDJ: Mudanjiang; ShuL: Shulan; JBH: Jingbohu; YT:
Yitong; ZFS: Zengfengshan; TC: Tianchi; WTE: Wangtian’e; LG:
Longgang; QY: Qingyuan; KD: Kuandian; SL: Shuangliao. For
convenience, the volcanos from SYS to MDJ to JBH to KD to PL are
referred to as the “East” Volcanic Belt. Those on the opposite side
along Daxing’anling are referred to as the “West” Volcanic Belt.
Those along Xiaoxing’anling are referred to as the “North” Volcanic
Belt. Before 80 Ma, volcanic activities were calcalkaline and
silicic. Basaltic volcanism occurred since about 80 Ma in the late
Cretaceous, and it becomes more alkaline with time. There was
sporadic basaltic volcanism in Shandong and Liaoning in the late
Cretaceous. Paleogene volcanism mostly occurred in Xialiaohe,
Bohai, as well as Huabei and Subei Plains. Often the basaltic flows
have been covered by more recent sedimentation, but boreholes show
that tholeiitic basaltic flows may be over 1000 m thick in places.
Volcanic activities peaked in Neogene. In the west of Huabei Plain,
Zhangjiakou-Weichang-Chifeng-Jining volcanic fields cover more than
20,000 km2, with both tholeiites and alkali basalts. In the east,
along Tanlu Fault and Yilan-Yitong fault systems, there was
widespread alkali basaltic eruption, especially Changbaishan
volcano group (including Tianchi, Zengfengshan, Wangtian’e), which
has been active for over 20 million years. Quaternary volcanism is
distributed around Dongbei Plain. Individual volcanic centers are
usually small and a volcanic field typically consists of several to
tens of cinder cones. The most famous volcano is the Tianchi
volcano, a typical stratovolcano with a caldera and also being the
most recent. Its last major eruption in 1200 AD was explosive with
a dense-rock equivalent volume of 32 km3 [Guo et al., 2002]. At
32km3, this is one of the largest historic eruptions, about twice
the volume of the 1883 eruption of Krakatau that killed 36,000
people. Other volcanic fields consist of mostly cinder cones, such
as Wudalianchi in Heilongjiang Province, Jinbohu and Longgang in
Jilin Province, Kuandian in Lianing Province, and Datong in Shanxi
Province. Neogene alkali basalts often contain mantle and crustal
xenoliths. Both volcanic rocks and xenoliths can be used to study
the interior of this region. Note that there are also volcanoes in
Southeast China (and Southeast Asia), but there is a significant
gap in between. Furthermore, some volcanos further south in SE
China may be genetically associated with the India-Eurasia
collision. These volcanoes are also important to investigate. In
this Greater North China Initiative, the focus will be on volcanoes
in NE China. Understanding the origin of the young volcanism in NE
China is critical not only for understanding the tectonic evolution
of the Greater North China, but also for the development of China.
Volcanic eruptions are main geologic hazards. Tianchi volcano, for
example, has the potential to produce large-scale explosive
eruptions in the future. Tianchi is already a
-
GNCI Whitepaper - 24 -
tourist attraction (especially for Chinese and Koreans) and it
will become more so as the economy of China develops. Understanding
the past and monitoring the volcano will be the key in forecasting
future eruptions and mitigation of life and property losses. 3.4.
Previous Studies on Cenozoic Volcanism in the GNC
There have been numerous investigations on the volcanic rocks
and mantle xenoliths in them in this vast area. Some are published
in international journals [e.g., Zhou and Armstrong, 1982; Chen et
al., 1984; Peng et al., 1986; Fan and Hooper, 1989, 1991; Song and
Frey, 1989; Song et al., 1990; Zhi et al., 1990; Basu et al., 1991;
Tatsumoto et al., 1992; Liu et al., 1994; Snyder et al., 1997;
Griffin et al., 1998; Hsu et al., 1998; Xu et al., 1998a,b; Menzies
and Xu, 1998; Chen et al., 2001; Liu et al., 2001; Ren et al.,
2002; Xu, 2002; Wilde et al., 2003; Wu et al., 2003; Xu et al.,
2003]. Most are understandably published in Chinese literature,
especially in the 1980’s [e.g., Liu et al., 1979, 1981; Deng et
al., 1980, 1987a,b; Lu et al., 1981, 1983; Wang et al., 1981, 1983,
1985; 1988; E et al., 1982, 1983; Feng et al., 1982; Liu and Wang,
1982; Du and Du, 1983; Hu et al., 1983; Sheng et al., 1983; Chen an
Peng, 1985, 1986; Wu et al., 1985; Zhu et al., 1985, 1988; Chen et
al., 1986, 1986a,b; Qiu et al., 1986a,b, 1988; E and Zhao, 1987;
Liu, 1987, 1988, 1989; Xie et al., 1988, 1989a,b,c; Liu et al.,
1989a,b; Wu, 1989; Zhi, 1989, 1990; Luo and Chen, 1990; Tang, 1990;
Tang and Tian, 1990; Mu et al., 1992; Liu et al., 1993; Zhi et al.,
1994; Liu and Xie, 1995; Deng et al., 1998; Fan et al., 1998,
1999a,b,c, 2000, 2001, 2002; Zhang et al., 2002]. There are some
books in Chinese, one expounding a single theme [Deng et al.,
1996], some timely collections of papers [e.g., Chi, 1988; Liu,
1992, 1995], and one single-authored book summarizing information
on volcanoes in China [Liu, 1999]. These studies have provided a
large amount of information, including petrographic and petrologic
descriptions, geochemical studies, isotopic data and synthesis,
inference of mantle compositions and conditions, and a basic
picture of volcano distribution in space and time.
Fig. 8. North-south (a) and east-west (b) vertical cross
sections of P-wave velocity images under the Changbai intraplate
volcano in NE Asia [Zhao et al., 2004]. Red and blue colors denote
slow and fast velocities, respectively. The velocity perturbation
scale is shown below the cross sections. Black triangles in (a) and
(b) denote the intraplate volcanoes. White dots denote earthquakes
that occurred within 100 km of the profiles. The two dashed lines
denote the 410 and 660 km discontinuities. (c) Locations of the
cross sections in (a) and (b). Black and red triangles denote
seismic stations and volcanoes, respectively. The contour lines
show the depths of the Wadati-Benioff deep seismic zone.
The numerous studies resulted in various proposed origins for
volcanism in NE China, which includes almost all possibilities of
intraplate volcanism (plume/hotspots, rifting,
-
GNCI Whitepaper - 25 -
back-arc extension, and lithosphere delamination and thinning),
reflecting a lack of understanding of the region. (1) Deng et al.
[1996, 1998] proposed that there are a number of mantle plumes
responsible for diffuse volcanism over the large area. (2) Liu et
al. [2001] suggested that back-arc extension related to the opening
of Japan Sea contributed to the volcanism. Zhao et al. [2004], Lei
and Zhao [2005], and Chen et al. [2005] based on imaged the
presence of a flat slab at 400 to 600 km depth within the
transition zone, also suggested a close relation between the
stagnant slab and alkali basaltic volcanism (Fig. 8). (3) Ren et
al. [2002] attributed the volcanism to a series of rift basins. (4)
A fourth hypothesis that is gaining more support is that the
volcanism is related to the delamination or thinning of the
lithosphere beneath Northeast mainland Asia [e.g., Griffin et al.,
1998; Menzies and Xu, 1998; Wilde et al., 2003]. (5) Niu [2005]
suggested that because of a sudden change in the lithosphere
thickness at the Great Gradient Line (at the boundary between East
China and West China), eastward mantle flow crossing the Line
experiences rapid decompression, leading to mantle partial melting
and volcanism in North and NE China. Until now, the consequences of
these various hypotheses have received only minimal discussion.
Many Neogene and Quaternary alkali basalts contain mantle
xenoliths, mainly spinel lherzolite and harzburgite. Xenoliths have
been investigated extensively to understand the geotherm and mantle
conditions [e.g., Deng et al., 1980; Liu et al., 1981; E et al.,
1982; Feng et al., 1982; Du and Du, 1983; Lu et al., 1983; Sheng et
al., 1983; E and Zhao, 1987; Xu et al., 1998b, 2003; Zhang et al.,
1998, 2000; Zhi, 1989; Chen et al., 2001; Zheng, 2001; Xu, 2002;
Zhou et al., 2002; Wilde et al., 2003], as well as He isotopes.
Although there is some difficulty in obtaining pressure because the
mantle xenoliths do not contain garnet, one major conclusion based
on mantle xenolith and other studies is that the lithosphere has
been thinned in the Mesozoic [e.g., Griffin et al., 1998; Xu et
al., 1998a,b; Menzies and Xu, 1998]. With a large body of
literature on NE China volcanoes, a large database on Northeast and
North China volcanoes exists. Recently, Chen et al. [personal
communication] have made an effort to compile the database.
Examination of the compilation reveals that the reliability of the
data is difficult to assess and there are inconsistencies.
Preliminary comparison seems to show that Sr-Nd-Pb isotopic data
are consistent among different groups and may be considered
reliable [e.g., Zhou and Armstrong, 1982; Basu et al., 1991;
Tatsumoto et al., 1992; Tu et al., 1992; Zhang et al., 1998; Chen
et al., 2001; Wu et al., 2003]. On the other hand, the quality of
trace element data, which has the potential to indicate whether
subducted or subducting slabs may be involved in their
petrogenesis, as well as other mantle characteristics, is not
always high. For example, Figure 2 compares literature Ta versus Nb
data in Quaternary NE China basalts [various authors, and names
intentionally withheld] and recent data from the University of
Michigan [Chen et al., personal communication]. There is a lot of
scatter in the literature data, with Nb/Ta ratio varying by a
factor of 10 (Fig. 9). More recent data show much less scatter with
Nb/Ta ratio varying by only about 20%. It is almost certain that
the large scatter in the literature data is due to poor analyses
although one has to analyze exactly the same rock to make sure this
is the case. Low-quality data do not allow unambiguous assessment
of various models, such as the involvement of slab component in the
derivation of primary magma.
-
GNCI Whitepaper - 26 -
0
1
2
3
4
5
6
0 20 40 60 80 100
Literature dataNew data
Ta
(ppm
)
Nb (ppm)
NE China
Fig. 9. Comparison of literature data and new and yet
unpublished data by Chen et al. [personal communication]. There is
large scatter in literature data.
Another example of questionable data quality is helium isotopic
ratios. Helium isotopes can be diagnostic of a deep mantle
reservoirs [e.g., Graham, 2002; Porcelli and Ballentine, 2002]
although the exact depth (e.g., upper versus lower mantle) or
source, or the uniqueness is still debated. For example, mid-ocean
ridge basalts have a roughly uniform 3He/4He ratio (about 8 times
the atmospheric ratio, or 8Ra; the atmospheric 3He/4He ratio is
1.4x10-6). Some ocean island basalts have high 3He/4He ratios (up
to 35 times Ra), often interpreted to indicate a plume component.
Continental crust usually has a low ratio of less than 0.1Ra. He
isotope ratios in mantle xenoliths and megacrysts brought to the
surface by basalts in NE China have only been measured by two
groups and are highly variable [Xu et al., 1998; Li et al., 2002;
He isotopic data in Xu and Liu, 2002 are the same as those in Xu et
al., 1998]. The reported ratios vary widely. There are some
intermediately high 3He/4He ratios, which, if verified, may suggest
a mantle plume contribution for NE China volcanics. There are also
some extremely high ratios, even higher than the ratio in the solar
wind, as well as extremely low ratios. The very high ratios likely
reflect cosmogenic 3He addition [Porcelli et al., 1987], especially
the extremely high 3He/4He ratio (up to 700 Ra) reported by Li et
al. [2002], but the authors did not carefully assess this
possibility. Because He isotopes might provide key evidence for the
involvement of plume component, it is critical to have reliable He
isotope data. In a recent (and unpublished) study, Chen et al.
[personal communication] obtained some He isotopic data, which
showed helium isotopic ratio in mantle xenoliths in NE China
volcanics is relatively uniform. Hence the high variability of He
isotopic ratios in NE China volcanics and related rocks is almost
certainly due to either sampling problems or analytical errors. By
careful sampling, these problems can be avoided so that the true
mantle component may be revealed by He isotopes. With regard to
geochronology, although many K-Ar ages have been obtained before,
the reliability of the ages is again difficult to assess and some
ages have recently been found to be in error [Fan Qicheng, personal
communications].
-
GNCI Whitepaper - 27 -
0.1
1
10
100
1000U MichiganCited in Gao 2004Xu and Liu 2002Xu et al. 2003Li
et al. 2002MORB range
3 He/
4 He
(Ra)
MORB range
Highest OIB
Solar
Cosmogenic
Chondrite, planetary
Earth's atmosphere
NE China, U of Michigan
Crustal contamination
Fig. 10. Comparison of literature data and unpublished six data
points on three volcanic fields by Chen et al. [personal
communication]. On the left-hand side of the data, typical 3He/4He
ratios of various reservoirs are given. There is large scatter in
literature data. Hence it is difficult to decide which data can be
reliably used and which ones should not be used if one needs to use
the data to test a hypothesis.
In recent years several geophysical studies were conducted to
investigate the structure of the crust and upper mantle under the
active volcanoes in NE China. Magnetotelluric soundings revealed
low-resistivity anomalies in the crust under the Changbai volcano
[Tang et al., 1997, 2001]. Seismic explosion experiments revealed
low-velocity anomalies in the crust and upper mantle down to a
depth of 40 km, suggesting the existence of magma chambers under
the Changbai volcano [Zhang et al., 2002]. With the recent
installation of 19 portable seismic stations in NE China [Wu and
Hetland, 1999], a few studies have been made to determine the
three-dimensional (3-D) structure of the crust and upper mantle
beneath the Changbai volcano. For example, receiver function
techniques were applied to the teleseismic waveforms recorded by
the portable seismic network to map the geometry of the seismic
discontinuities in the crust and upper mantle (the Moho, 410 and
670 km discontinuities) [Ai et al., 2003; Li and Yuan, 2003;
Hetland et al., 2004]. These studies showed that the crust is
thicker and contains low-velocity bodies beneath the Changbai
volcano, and that the 670 km discontinuity is depressed under NE
China, suggesting that the subducting Pacific slab is stagnant in
the mantle transition zone. High-resolution seismic images of the
mantle down to 800 km depth are determined beneath the Changbai
volcano by applying a teleseismic tomography method to relative
travel time residuals recorded by the portable seismic network
[Zhao et al., 2004; Lei and Zhao, 2005]. The results show a
columnar low-velocity anomaly extending to 400 km depth under the
Changbai volcano. High-velocity anomalies are visible in the mantle
transition zone, and deep earthquakes occur at depths of 500-600 km
under the region, suggesting that the subducting Pacific slab is
stagnant in the transition zone, as imaged clearly also by both
global tomography [Zhao, 2004] and regional tomography [Pei et al.,
2005; Chen et al., 2005]. These seismological results from
tomographic and receiver function analyses suggest that the
Changbai volcano is not a hotspot like Hawaii but a kind of
back-arc intraplate volcano related to the upwelling of hot
asthenospheric materials associated with the deep subduction,
dehydration, and stagnancy of the Pacific slab under NE Asia [Zhao,
2004; Zhao et al., 2004; Lei and Zhao, 2005; Chen et al., 2005]. In
summary, there has been extensive work on Cenozoic igneous
activities and associated
-
GNCI Whitepaper - 28 -
mantle xenoliths in the GNC region. These works provide
background information on which to build new studies. Nevertheless,
many volcanic fields have not been investigated, or have not been
investigated in any detail. Furthermore, for previous studies, the
quality of the data is sometimes difficult to assess, and for some
the quality is low, reflecting either analytical difficulties in
the early years, or sloppiness in sampling and analyses. New,
careful, systematic, and coordinated studies are necessary to
improve our understanding of the regional volcanism in the GNC
region. 3.5. Major Questions
Major questions remain regarding Cenozoic volcanism in NE China,
including (the questions go from the more concrete to the more
elusive):
(1) Structure of the deep mantle crust
• Exactly how thick is the NCC lithosphere? How did the thinning
occur in late Mesozoic [e.g., Gao et al., 2004]?
• What’s the 3D velocity structure of the lithosphere and upper
mantle in GNC and adjacent areas? What is the lithosphere
thickness? What is the seismic anisotropy of the region? What is
the structure in the deeper Earth beneath GNC? How is the structure
in GNC related to regional structure and tectonics of the
India-Eurasian collision and the subduction of the western
Pacific?
• One of the most fundamental questions is how the intraplate
earthquakes and tectonics in GNC are related to the deeper
structure and dynamics. Is there correlation of earthquake
locations with lithosphere velocity structure? How well can we map
the velocity structure into temperature, composition, partial melt,
and the rheology of the region? What is the origin of the seismic
anisotropy and does it have any relation to stress field and
deformation?
(2) Spatial and temporal variation of igneous activity • What is
the spatial and temporal distribution of Cenozoic igneous activity
GNC and
what control them? • What is the role of Pacific subduction vs.
possible hot-spot activities in producing
igneous activities in GNC. • What are the spatial-temporal
geochemical variations? Are there large-scale spatial
patterns, and if so, what is their significance? Are there
general evolutionary trends for single, long-lived volcanoes, and
for whole volcanic fields or regions? If so, what do they
imply?
• How do mantle source composition and conditions of magma
formation vary spatially and through time?
(3) Relation between seismic velocity structure and igneous
activity • What is the distribution of the igneous activity in
space and time? • What is the relation between deep structures and
the distribution of igneous activities?
What is the lateral and depth extent of the low-velocity bodies
under each of the active volcanoes such as the Changbai and
Wudalianchi?
• Are the low-velocity bodies due to high temperature or high
water content (or some other compositional differences)?
-
GNCI Whitepaper - 29 -
• Is the subducting Pacific slab really stagnant in the mantle
transition zone under NE Asia? If so, how far does it penetrate
westward to the interior of the Asian continent, and how does it
affect the surface tectonics in addition to volcanism?
• Are there mantle plumes in the upper and/or lower mantle under
NE Asia? Is there a mantle plume component characterized by high
3He/4He ratios? Is there contribution from subducted slabs to the
composition of the igneous rocks in GNC (which would support the
back-arc hypothesis)?
• Could there be interactions between a mantle plume and the
subducting slab beneath GNC?
• What happened at about 80 Ma that led to the transition from
calcalkaline silicic volcanism to basaltic volcanism? What led to
the thinning of the lithosphere? What is the deep process and
tectonic significance? What are the roles of the subducting Pacific
plate, or back-arc extension, or India-Asia collision? Are there
mantle plumes? What is the accurate timing of the transition? Does
the transition age progress from west to east?
• What can we infer about mantle compositions, conditions and
processes that contribute to the widespread volcanism?
3.6. Possible Research Directions Seismology. Permanent seismic
stations in North China are very sparse. To determine the 3-D crust
and mantle structure of this region, at least tens of portable
seismic stations should be installed for a period of 6 months to 2
years. Waveforms from local, regional and teleseismic events can be
recorded by the seismic network using state-of-the-art
seismological methods. One method is seismic tomography for
determining the 3-D P and S wave velocity variations in the crust
and mantle. From the obtained P and S wave velocities (Vp, Vs),
Vp/Vs ratio or Poisson’s ratio tomography can be estimated.
Amplitudes of seismic waves can be used to determine seismic
attenuation (seismic quality factor, Q) tomography. Hot magma
chambers and mantle plumes would exhibit low-Vp, low-Vs, low-Q and
high Poisson’s ratio, while cold subducting slabs exhibit high-Vp,
high-Vs, high-Q and low Poisson’s ratio. From the seismic velocity
and attenuation tomographic images, we can estimate the size and
spatial extent of magma chambers, mantle plumes and subducting
slabs [Zhao, 2001b; Zhao et al., 1997]. Poisson’s ratio (or Vp/Vs)
is a key parameter in studying petrologic properties of crustal and
mantle rocks [Christensen, 1996; Zhao et al., 1996] because it can
provide tighter constraints on the composition than Vp or Vs alone.
Its value in common rock types ranges from 0.20 to 0.35. Poisson’s
ratio has proved to be very effective for detecting magma and
fluids in the crust and mantle [Zhao et al., 1996, 2002]. From Vp,
Vs, Q and Poisson’s ratio, temperature and content of melts and
fluids can also be estimated. Another useful seismological method
is teleseismic receiver functions which can be used to determine
geometry and sharpness of seismic discontinuities such as the Moho,
and the so-called “410” and “670” km discontinuities which
represent the upper and lower boundaries of the mantle transition
zone [e.g., Ai et al., 2003]. In and around the cold subducting
slab, the 410-discontinuity is elevated, while the
670-discontinuity is depressed, thus the mantle transition zone
would become thicker. In contrast, if a hot mantle plume exists,
the 410-discontinuity will be depressed, while the
670-discontinuity will be elevated, and the
-
GNCI Whitepaper - 30 -
mantle transition zone will become thinner. Thus, receiver
function analyses can also provide information on mantle plumes and
subducting slabs, complementing tomographic imagings.
Field studies of Cenozoic igneous rocks. Careful and
well-documented fieldwork is a prerequisite for all the subsequent
laboratory studies. If detailed mapping has not been carried out
yet for some major volcanic fields, mapping will be the first step.
In the five years of the Greater North China Initiative, it is
hoped that all major volcanic fields will be mapped. To collect
samples for the purpose of laboratory analyses, for every sample,
the longitude, latitude and elevation of the sample collection site
will be recorded, plus other relevant information on the sample
(such as quarry, road cut, valley, cliff, etc.) If the flow or
eruption units have been mapped, such unit should be indicated for
the sample. It is important to follow some sampling protocols,
which may depend on the purposes of samples. For example, for He
isotopic analysis, to avoid cosmogenic helium, it is important to
sample fresh exposures, meaning rock interior many meters inside
was recently (within the last tens of years) exposed. Recent
quarries and road cuts are excellent sampling sites. If one simply
picks up a megacryst on the ground, or if one hammers out a sample
from a naturally exposed surface, the megacryst might have been
exposed to cosmic ray bombardment for thousands or more years, and
its helium signature may be largely cosmogenic. To avoid crustal
contaminations, it is best to sample young fresh unaltered samples
(such as Holocene eruptions), and samples away from any crustal
xenoliths. To further coordinate the studies, it might help to
choose a suite of samples from various volcanic fields as
“reference” samples, for which all possible geochemical analyses
and experimental work will be carried out on them (this was done in
the “Basaltic Magmatism Project” in the 1970’s in the US). For
geochemical modeling and for deeper understanding, very often it is
critical to have all analyses done on a selected number of samples.
However, the literature data often have major element and some
trace element analyses for one rock, REE analyses for another,
Sr-Nd isotopic analyses for another, Pb isotope analyses for
another, etc.
Geochronology. A number of authors published dates on volcanic
rocks in greater NE China [Chen and Peng, 1985; Liu, 1987; Wang et
al., 1988; Luo and Chen, 1990; Liu et al., 1992; Chen et al.,
1992a,b]. The data are extensive and cover many volcanic fields.
However, there are important limitations. First, for some volcanic
fields, there are no age data. Secondly, for most volcanic fields,
the age data are insufficient to examine the full duration of
volcanic activity, and therefore, how petrochemistry evolved with
time. For example, there are about 20 eruption centers in Kuandian
volcanic field [Liu et al., 1992], but only two samples have been
dated to be 0.12 Ma for Liujia and 0.27 Ma for Huangyishan.
Therefore, the age data are not enough to investigate the space-age
relation of volcanic activities in the volcanic field and whether
geochemical characteristics vary with time. Trace element
characteristics in three eruption centers (Qingyishan, Huangyishan,
and Liuja) of Kuandian volcanic field are quite different. Even in
one eruption center (Huangyishan), there is significant variation
in trace element geochemistry. Do these differences reflect time
progression? Did the three eruption centers (only several km away
from one another) erupt similar kind of magma at the same time? Or
are the three centers fed by different magma chambers? More dating
results will be necessary to address these questions.
-
GNCI Whitepaper - 31 -
To provide the spatio-temporal distribution of volcanic
activities in the greater NE China, most or all volcanic centers
need to be dated. For larger volcanos with prolonged eruption
history, all major eruptions need to be dated. These studies will
provide understanding of the spatio-temporal distribution of
volcanic activity, as well as possible secular trend of primary
magma composition in a single volcano and in the whole NE China
volcanic province that might be related to change in mantle
conditions. By the end of the Greater North China Initiative, it is
hoped that there will be hundreds to thousands of new and accurate
dates. With such data, a movie will be made at 1-Myr interval to
show how the volcanic activity in the field area of Figure 1
evolved with time. That is, for every one million year interval, a
movie frame will be created to show the distribution of the
volcanic activities at that time period. For the 80-Ma interval,
there will be 80 frames (roughly 3 seconds). This 3-second movie
would show the spatial-temporal distribution of volcanic
activities, the regional variation and pattern in eruptive
activity, and would help to recognize possible hotspot tracks,
rifts or hot lines, and other