Geologic Structure and Metamorphism
GENERAL STRUCTURAL FEATURES
The shaly sediments in the argillite-slate series of the Central Range are relatively plastic and
incompetent. The deformation of these sediments produced predominantly shear folding and the rocks were
broken into a complex series of fault blocks. Due to the ductility of the comparatively weak shaly rocks,
many complicated fold structures were formed. Similar folding is well-developed in the strongly deformed
metamorphic rocks, with extensive development of slaty cleavage in the more intensely metamorphosed rocks.
Fracture cleavage is developed in the sandy rocks. Low-grade regional metamorphism accompanied the
deformation and the grade of metamorphism increases eastward toward the core of the Central Range. Very
minor igneous activity resulted in small and disseminated volcanic lenses intercalated in the sediments. The
strata are locally quite imbricated and crumpled.
In the eastern Backbone Range belt of this subprovince, detailed structural mapping is difficult due to the
predominance of a monotonous and thick series of slate and phyllite. A scarcity of sandstone marker beds
makes it difficult to delineate even major structures. In northeastern Taiwan, this argillaceous series
shows a prevailing dip toward the east with an east-northeast to north-northeast regional strike. A
considerable part of the slaty beds are overturned and isoclinally folded. Southward to the high part of the
Backbone Range, the slate and phyllite are intricately folded and highly contorted and sheared. Cleavage is
more strongly developed here than elsewhere. The rocks strike north-northeast or north-south with varying
dips.
The presence of well-developed arenaceous and carbonaceous marker beds in the Hsuehshan Range belt is of
great help in subdividing the submetamorphic argillaceous sediments into a number of mappable units, thus
facilitating structural study of the region. Folding and faulting can be readily recognized, the former
being more prominent. Two kinds of folding can be distinguished: open and broad folds and small-scale folds.
The major part of the region has several large and broad folds approximately parallel with the principal
faults. Both synclinal and anticlinal structures are present, each 2 to 10 kilometers across and 10 to 40
kilometers long. The longest fold may reach 100 kilometers in strike length. In the northern part, fold axes
are mainly east-northeast. They turn to northeast and north-northeast toward the south and are nearly
north-south in the Puli-Kukuan area in central Taiwan. Dips are moderate on both limbs, ranging from
30o to
60o. Locally the dip may be
10o to
25o in the gentle part of the folds. Some folds
are symmetric with nearly vertical axial planes. Some asymmetric folds are characterized by easterly dipping
axial planes at high angles. Small-scale folds are developed on both flanks of the major folds. Their axes
are parallel to the axes of the major folds. Many of these small-scale folds are closed asymmetric folds
with their axial planes dipping mostly toward the east or southeast. Some recumbent and inverted folds are
also recorded. The major and small-scale folds can be combined into anticlinoria and synclinoria. The
Szeleng Sandstone is usually exposed in the cores of the anticlinoria and the Sulo Formation, the Aoti
Formation or the Tatungshan Formation often form the central part of the synclinoria.
Faults present are principally thrust faults with the upthrown blocks to the east or southeast. Surface dips
of the fault planes are generally steep. These faults may flatten at depth but subsurface data are not
available. A few transverse faults cut across the strata almost at right angles with nearly vertical fault
planes. They are mainly dextral strike-slip faults with offsets ranging from several meters to 1 or 2
kilometers.
BOUNDARY FAULTS
The geologic subprovince of the argillite-slate series in the Central Range is bordered on the west by the
western foothills geologic province and on the east by the metamorphic complex geologic subprovince. Both
the eastern and the western boundary are marked by a longitudinal fault which separates the two adjacent
provinces. These are the boundary faults to be discussed in this chapter. It seems that nearly all the major
geologic provinces or lithotectonic belts in Taiwan are bordered by tectonic lines.
As mentioned in the preceding chapter, the argillite-slate series unconformably covers the pre-Tertiary
metamorphic complex but .the surface of unconformity has been obscured and overprinted by a thrust fault at
least in some places. The present contact between the slate cover series and the metamorphic basement is
marked in many places by a boundary fault along which the metamorphic complex has been thrust northward or
eastward over the slate series. Due to insufficient mapping data, it is not yet certain whether this fault
is a major tectonic line traversing the entire island of Taiwan or a local structural feature restricted
only to limited areas. The boundary fault between the Tertiary slate cover and the basement is an
interesting subject for further study in Taiwan.
One major boundary fault follows the contact between the argillite slate belt and the fold-and-thrust belt
of the western foothills region. The actual fault surface is generally difficult to identify due to heavy
vegetation and soil cover on the surface. This boundary fault is named the Chuchih fault in general, but
locally, it is called the Shuichangliu fault in central Taiwan, and the Laonungchi fault or the Chaochow
fault in southern Taiwan (see Fig. 4).
This major tectonic break is interpreted as a longitudinal upthrust dipping steeply toward the east by Biq
(197l a). He further indicates that it may be a dislocation of more
than local importance. This fault may enter the island from the sea in the north and extend into the sea
again in the south. In the retrodeformable cross section of the northern Taiwan fold-and-thrust belt
prepared by Suppe (1980), this boundary fault is shown as a low-angle thrust at depth, similar to other
low-angle thrust faults in the western foothills, with a Neogene section extending back underneath the slate
belt. This boundary fault can be proved by the juxtaposition of different formations at different places
along the surface trace of the fault. It is difficult to locate this fault precisely in the Laonungchi
drainage area in Kaohsiung-hsien and Pingtung-hsien, where the strata exposed on the two sides of this fault
are not easily differentiated due to generally low metamorphic grade.
Another important tectonic break in the argillite-slate series is the boundary fault separating the
Hsuehshan Range belt from the Backbone Range belt. This longitudinal fault is called the Lishan fault
(Fig. 4). It gradually merges with the western
boundary fault of the slate series (southern extension of the Chuchih fault) north of Liukuei in southern
Taiwan. Between these two longitudinal boundary faults is the upthrust block of the Hsuehshan Range
(Fig. 4). However, the southern extension of
the Lishan fault south of Yushan is largely hypothetical as no reliable
field data are yet available.
Field recognition of the Lishan fault is extremely difficult as foliated slaty beds are exposed on both
sides of this fault at most places. Morphological evidence of this dislocation is also weak. On this ground,
some geologists (C. H. Chen and others, 1983a) still doubt the existence of this tectonic break in the slate
belt. However, geologic inference on the basis of remote sensing study and recently telemetered seismic
network data substantiate the presence of this longitudinal fault, although the exact location of the fault
line is inferred and subject to dispute in some places due to lack of concrete field evidence. This
longitudinal fault has been called the median fault in some reports due to its geographic location in the
medial zone of the island. This fault is believed to be still active on seismological evidence (F. Wu,
1978). The seismic fault discovered beneath the Ilan Plain (Tsai et al., 1975) has been interpreted as the
northern extension of this median fault in the coastal area.
On the evidence of varying stratigraphic offset and the NNW-orientation of compressive axis southeast of
Taiwan, Francis Wu (1978) believed the median fault should have a west-vergent thrust component. He inferred
a hinge action on the median fault to account for the extensional opening of the seismic fault under the
Ilan Plain. Biq (197 la) indicated that the Lishan fault is an east-dipping thrust fault largely responsible
for the strong block-uplift of the Central Range in Taiwan. The configuration of the fault trace and
tectonic synthesis suggest that the original movement of the Lishan fault is thrusting. However secondary
lateral or recent tensional movements may have locally affected this area due to relaxation of compressive
forces. The Lishan fault merges with the Chuchih fault into one single thrust fault in the north part of the
Pingtung Valley in southern Taiwan. After the coalescence of these two boundary faults, the southern section
of the median fault is named the Laonungchi fault or the Chaochow fault, and it extends along the eastern
margin of the Pingtung valley into the sea
(see Fig. 8).
From topographic expression and stratigraphic
relations, the Laonungchi fault is definitely a thrust fault with the east side upthrown. This fault is
still active, as indicated by the telemetered seismic data, although no major historical earthquake is known
to be associated with it.
OROGENIC MOVEMENTS AND STRATIGRAPHIC BREAKS
The older Tertiary argillite and slate series of the Central Range is lithologically and structurally
different form the non-metamorphosed younger Tertiary sedimentary rocks in the western foothills. This fact
led early workers to postulate an orogenic movement between the deposition of these two major geological
entities. The older Tertiary rocks were believed to have been deformed and metamorphosed before the younger
Tertiary rocks were deposited on the west. As the youngest strata in the slate belt are middle Miocene, the
orogeny could be middle Miocene. It was named the Puli orogeny in early geologic literature after the town
of Puli in central Taiwan.
The postulated middle Miocene Puli orogeny is no longer acceptable to geologists who are in strong support
of the plate tectonic theory. They argue that the older Tertiary slates and the younger Tertiary clastic
sediments in western Taiwan represent continuous deposition in one Tertiary subsiding trough and that there
was no mid-Tertiary orogeny. Rocks of overlapping ages have been found in these two litho- tectonic belts.
There is no evidence of angular discordance and coarse clastic deposits to support the hypothesis of a
mid-Tertiary orogeny. The most important tectonic movement in the western basin must have taken place during
the collision between the Eurasian plate and the Philippine Sea plate in Plio-Pleistocene time. In this
orogeny, both the argillite-slate belt in the Central Range and the Neogene clastic rocks in the western
foothills were subjected to major diastrophism over the entire island. The comparatively high degree of
metamorphism of the slate series merely reflects the greater uplift and depth of tectonic burial in the east
due to proximity to the collision center.
The study of orogenic movements in the argillite-slate series is most controversial and challenging. This
argillaceous series is generally undifferentiated structurally due to its apparent continuity without
distinct breaks or convincing unconformable contacts. Furthermore, a complete stratigraphic sequence of the
argillaceous strata with definite age assignments can hardly be established because fossiliferous horizons
are usually separated by intervening barren strata totally lacking fossils. Dating of orogenic movements is
difficult without precise knowledge of the stratigraphy of the rocks. As uncertainty of both structure and
stratigraphy of the slate series is the rule, it defies any attempt to unravel in detail the orogenic
history of these argillaceous strata.
Some geologists consider the conglomerate lenses and patches intercalated at several horizons in the
argillaceous sequence as evidence of tectonic and orogenic disturbance. Paleontologists are inclined to
propose stratigraphic breaks on the evidence of faunal gaps in the studied sections. However, fossil
zonation in the slate series is not continuous. Barren intervals are found in many stratigraphic sections,
some reaching a considerable thickness. Therefore, it is very questionable to use missing fossil zones as
evidence of stratigraphic or structural breaks.
The coarse conglomerates in the slate sequence certainly indicate crustal instability and differential
movement including uplift, erosion, and deposition. But whether all these conglomerates should be classified
as orogenic is questionable; they may reflect merely the deposits of canyons and channels cut into the slope
of the passive mid-Tertiary Asian continental margin. The conglomerates usually occur as several thin or
bulging lenses or interbeds intercalated at different horizons in the slates. The conglomerates exposed in
widely separated areas in the slate series could be of different stratigraphic horizons. It is, therefore,
not wise to assume that all these conglomerates are of the same age; and there is no fossil evidence to show
the age of the slates overlying and underlying the conglomerates. It is not justified to interpret all these
conglomerates as basal conglomerate without further proof. Conglomerates may be formed in many ways. It could
be intraformational, resulting from local crustal unrest; or it could be redeposited gravel carried from the
shelf or shore into submarine fan channels by turbidity currents. Detailed study of the sedimentology and
mode of occurrence of the conglomerate in the argillite-slate series and more precise dating of the
surrounding strata are essential before any conglomerate layer is defined as "orogenic."
In the early geologic reports, three different "basal" conglomerate horizons were proposed within and below
the slate belt, each designated by a Roman letter. They are M conglomerate, E conglomerate, and N
conglomerate. The M conglomerate has already been discussed in the previous section. The E conglomerate,
named by Yen and others (1956), was proposed as a basal conglomerate separating, the "Cretaceous" Pihou
Formation from the Eocene slate in northeastern Taiwan and has already been discussed in the section "age
and fossils". No Cretaceous fossil has yet been found in the Pihou Formation to substantiate its age
assignment. In addition, the mode of occurrence of the E conglomerate at the type locality is also
incompatible with a basal conglomerate origin. The conglomerate occurs in several discrete bulging patches
or lenses interbedded at varying horizons in the slate and is not restricted to a definite horizon
separating two different rock units. The occurrence looks more like a fan gravel deposit than an erosional
residue. Recent age dating analysis by Eric W. Law (personal communication) also suggests that the slates
overlying and underlying the E conglomerate at the type locality are generally of similar ages. An appraisal
of all the geologic and paleontologic features shows that the E conglomerate could not be a basal
conglomerate separating the Eocene slates from "Cretaceous" slates. L. S. Chang (1966) considered the E
conglomerate to be in part an intraformational conglomerate. The above conclusion, however, is not intended
to rule out the possibility of a tectonic hiatus between Eocene and "Cretaceous" in other parts of the slate
series, but convincing evidence has yet to be found.
In the Backbone Range belt, a stratigraphic break between the Miocene Lushan Formation and the Eocene
Pilushan Formation is represented by the lack of Oligocene strata between these two formations. L, S.
Chang (1963 a) reported earlier that a faunal gap could have been present between the Lushanian stage and
the Pilushanian stage in his studies. This gap may represent a possible orogenic disturbance between the
deposition of Miocene and Eocene rocks in the Central Range, but no direct evidence of this disturbance was
found at that time. Later work by L. S. Chang (1974) claimed that conglomerate blocks are distributed at the
basal part of the Miocene slates on the western slope of the Nantawushan in Pingtung-hsien, southern
Taiwan. This conglomerate was named the N conglomerate by L. S. Chang (1972) to mark a possible Oligocene
hiatus or unconformity between Miocene and Eocene in the slate belt. At the type locality Nantawushan on the
western side of the southern Central Range, Eocene Discocyclina-limestone succeeds Miocene slates in
an inferred inverted sequence, with no intervening Oligocene fossils. At the type section, however, no
conglomerate outcrop has been found in place at the contact, although limestone conglomeratic blocks
containing Discocyclina species are scattered on the mountain slopes. Thus physical evidence of the N
conglomerate and of the Miocene/Eocene hiatus is very weak at the type section, where overturning of the
strata must first be established to substantiate this unconformity.
On the other hand, the study of calcareous nannofossils by T. C. Huang (1980) along the south cross-island
highway yielded biostratigraphic data to support an unconformity between the upper Oligocene and Eocene
rocks in the slate series. He suggested that the upper Oligocene transgressive strata (Likuan Formation) may
unconformably overlie the Eocene strata (Pilushan Formation). According to Huang, this unconformity has also
been substantiated by the structural mapping of C. T. Lee (1977) in the area concerned. This unconformity,
if present, could be related to the Miocene/Eocene hiatus reported by L. S. Chang (1972) and be grouped into
the same orogenic event. However, neither Huang nor Lee mention the occurrence of conglomerate at the upper
Oligocene (?)/Ecocene hiatus in their study area. Lee (oral communication) is of the opinion that this
surface of unconformity could be disturbed by later faulting and has become indistinct. Furthermore, the
limestone at the basal part of the upper Oligocene (?) Likuan Formation is generally conglomeratic and could
be an autobreccia resulting from compression and fracturing. It is to be determined if this conglomeratic
limestone is the equivalent of the N conglomerate proposed by L. S. Chang (1972; 1974).
Based on the evidence of N conglomerate and other features, L. S. Chang (1972 and 1974) believed that the E
conglomerate proposed by Yen and others could be redefined as the N conglomerate. He also staled that where
the Miocene Lushan Formation is in direct contact with the Tananao Schist on the eastern slope of the
Central Range, any conglomerate discovered between these two units should be named N conglomerate instead of
M conglomerate.
From the paleontologic studies discussed above, a Miocene or late Oligocene/ Eocene break possibly exists in
the argillite-slate series of the Central Range. This hypothetical break represents the only diastrophism
recorded during deposition of the slate series. L. S. Chang (1963 and 1972) suggested that this movement be
named the Puli Movement mentioned previously. However, the definition of the Puli Movement is quite
different from the hiatus discussed here, and this hiatus does not exist in the Puli area. A more
appropriate name should be proposed for this hiatus and hypothesized diastrophic event.
METAMORPHISM AND METAMORPHIC FACIES
The argillaceous rocks in this geologic province have been weakly metamorphosed into an argillite-slate
series during Plio-Pleistocene orogeny, with metamorphic grade increasing from west to east. A general
comment on metamorphic grades in northern Taiwan is shown in a map by Liou (1981b). He indicated a
progressive metamorphic sequence from zeolite facies through prehnite-pumpellyite facies to greenschist
facies in a west to east direction. This sequence is believed to be a direct consequence of the
continent-arc collision in Plio-Pleistocene time. In Liou's map, zeolite facies metamorphism is 'delineated
on the border zone between the western foothills and the Hsuehshan Range belt of the slate series. The
prehnite-pumpellyite facies assemblages are developed toward the eastern part of the Hsuehshan Range belt. A
transition to low greenschist facies is indicated eastward in the Backbone Range belt of the slate series.
The characteristic mineral assemblages suggest that metamorphism of the slates and phyllites in the Backbone
Ranges took place under chlorite zone conditions of the greenschist facies.
The progressive metamorphic sequence of the slate series in central and southern Taiwan is little known due
to lack of systematic petrographic studies. Liou (1981b) is of the opinion that similar but less intense
metamorphic zonation should be anticipated in southern Taiwan. In the explanatory notes of the metamorphic
facies map of Taiwan (C. H. Chen and others, 1983b), the metamorphic grade of the Tertiary rocks in Taiwan
decreases progressively from north to south and from east to west. This is a direct result of the collision
of the Eurasian continental margin with the Luzon island arc. The tectonic compression of collision
increases toward the east where thick and deep stratigraphic loading occurs. On the other hand, because
plate collision in Taiwan began in the north and is gradually proceeding to the south; southern Taiwan,
where collision is still active, exposes at the surface rocks that have been subjected only to low-grade
regional metamorphism.
From the analysis of metamorphic fades in Taiwan (C. H. Chen and others, 1983b), the Lushan Formation on the
west side of the Backbone Range belt and to the north of the Hsilochi stream is mostly of
prehnite-pumpellyite fades. To the south of the Willing Farm in the upper course of the Tachiachi, the
Chiayang Formation, the Tachien Sandstone and the Shihpachungchi Formation in the Hsuehshan Range belt are
of greenschist facies. However, the metamorphic grade is higher than that of the outboard areas. This
manifests clearly the relation of metamorphic grade to the age of rocks and to the depth of burial. The
Lushan Formation east of Tayuling on the central cross-island highway is higher in metamorphic grade than
the Lushan Formation on the west. On the other hand, the metamorphic grade of the Lushan Formation south of
the Hsilochi (greenschist facies) is slightly higher than that of the same formation north of the Hsilochi
(prehnite-pumpellyite facies). This is also related to burial depth. The metamorphic grade of the Eocene
rocks to the south of Yushan in the areas of Kuanshan, Pinanchushan, Wutoushan and Tawushan is comparatively
low, in the category of prehnite-pumpellyite facies only; these strata have not been buried very deeply.
|
