QUATERNARY RESEARCH45, 254 – 262 (1996)
ARTICLE NO. 0026
Sea Level Changes in the Last Several Thousand Years,
Penghu Islands, Taiwan Strait
YUE-GAU CHEN ANDTSUNG-KWEILIU
Department of Geology, National Taiwan University, 245 Choushan Road, Taipei 106, Taiwan, Republic of China Received May 15, 1995
below 5 m, that approximate past positions of ancient mean
Holocene shore-face and beach-face deposits form plains õ5 high tide level (MHTL, Fig. 2B).
m above present sea level along Taiwan Strait. We measured The resulting age – altitude plot allows evaluation of late the14C ages of detrital mollusk shells and coral in such deposits
Holocene relative sea level change in the Penghu Islands.
at the Penghu Islands. Twelve carbonate samples — mainly from We then compare the Penghu Islands data with sea level the largest island, Makung — were dated. Age measurements for
changes elsewhere in the central and western Pacific region.
two coral samples and one mollusk sample from the same out-crop imply that the 14C ages of mollusk shells give the best
GEOLOGICAL BACKGROUND approximation of depositional age. The highest Holocene
rela-tive sea level in the Penghu Islands occurred about 4700 years
ago with a height about 2.4 m above the present sea level. There- Plate Tectonic Framework
after, relative sea level appreciably fell without detectable
fluc-During the early and middle Cenozoic the continental
mar-tuations to its recent position. Our sea level data are consistent
gin near Taiwan underwent extension and the development
with other studies from the central and western Pacific, except
of rift basins (Sun, 1982). Volcanism during the Miocene
for the timing of peak sea level position. This variation may
reflect local crustal response to hydroisostatic effects on the formed the Penghu Islands. The post-Miocene geologic
his-continental shelf. q 1996 University of Washington tory is marked by convergent tectonics due to the collision between the Asian continental margin and Philippine oceanic plate (Ho, 1982; Teng, 1987). This period of convergent deformation is known as the Penglai Orogeny. The Penglai
INTRODUCTION
orogenic belt is located 100 to 200 km east of the Penghu Islands. Several studies have shown that deformation related Local variations in late Holocene sea level change can to the Penglai Orogeny does not extend into the Penghu reflect local crustal rheology (Walcott, 1972; Kidson, 1982). Islands (Chow et al., 1991; Hsiao et al., 1991; Yang et al., To identify such variations, it is important to infer local sea 1991).
level change from water level indicators found in tectonically
stable areas. Geological and neotectonic evidence implies Long-Term Crustal Stability that the Penghu Islands have been very stable, both on long
(several million years) and short (several thousand years) Four observations imply that the Penghu Islands have been tectonically stable since the late Miocene. (1) Late Miocene time scales. Hence, the islands are an appropriate place to
evaluate local Holocene sea level change on the continental basalt and sedimentary strata (8 – 16 myr; Jaung, 1988; Lee, 1990, 1994), which formed near sea level, are located near shelf of Asia (Fig. 1A).
The only materials suitable for dating Holocene sea level modern sea level. Based on the sea level curve of Haq et al. (1987), late Miocene sea level was as much as 50 m changes in the Penghu Islands are coarse-grained detrital
mollusk shells and corals in shore-face to beach-face depos- higher than modern sea level and as much as 80 m lower than modern sea level. Consequently, there is almost no net its. Because the present positions of sampling points do not
necessarily reflect the sea level at the time the samples were vertical crustal movement since late Miocene (Liu, 1989). (2) The late Miocene sedimentary strata are not known to deposited, using the elevation of sampling points as an
indi-cator to determine the ancient sea level can sometimes be be tilted or faulted (Lee and Chen, 1992). (3) The youngest basalt found in Penghu Islands was erupted 8 myr ago (Ju-misleading (Pinot, 1979; Kidson, 1982). To avoid this
prob-lem, we used the altitude of geomorphic surfaces, at sites ang, 1988; Lee and Chen, 1992; Lee, 1994). This suggests that extensional tectonics and associated normal faulting, where the samples were collected, as indicators of past sea
level positions. The surfaces are coastal plains, at an altitude which had characterized the early and middle Miocene in 254
0033-5894/96 $18.00
Copyrightq 1996 by the University of Washington.
All rights of reproduction in any form reserved.
FIG. 1. Maps of (A) Taiwan and its vicinity, (B) study area and sampling localities. Cn, Chihkan; Co, Chinglo; L, Lintou; S, Sokang; F, Fengkuei; T, Tashihpi; K, Kupoyu.
this area, probably ceased 8 myr ago. (4) Seismic profiles that significant subsidence has not occurred in the Penghu Islands during the past several thousand years. (3) Few earth-near the Penghu Islands (Chow et al., 1991; Hsiao et al.,
1991; Yang et al., 1991) show that the strata younger than quakes have occurred historically at the Penghu Islands rela-tive to tectonically acrela-tive Taiwan (Tsai et al., 1977; Tsai, the Miocene are neither penetrated by the underlying normal
faults nor deformed by the collision event that occurred in 1986).
Evidence on both Neogene and Holocene time scales thus Taiwan.
supports our presumption that the Penghu Islands have been
Short-Term Crustal Stability tectonically stable for the past several million years. The
Penghu Islands, therefore, are well suited to record Holocene Additional evidence implies that the Penghu Islands were
sea level changes in Taiwan Strait. relatively stable during middle and late Holocene: (1) Raised
coral reefs have not been found on land (Lin, 1967; Ting,
COASTAL PLAINS AND DEPOSITS
1990), although many vast reef flats are developed under the sea surrounding the Penghu Islands. This indicates that
Geomorphic Characteristics little crustal uplift has occurred during postglacial time. (2)
Holocene shore-face deposits are widespread on the coastal Coastal plains are common in the Penghu Islands, espe-cially along marine embayments and in basaltic lowlands. plains (Lin, 1967). They have been interpreted as the product
of the Holocene high sea level stand, a general trend ob- The plains are somewhat similar to marine terraces described by Woods (1980) and Kern (1977). Elevations of the terrace served around the western Pacific (Adey, 1978). This implies
256 CHEN AND LIU
Depositional Model
Because most of the sedimentary deposits of the coastal plains probably formed on the upper shore face to beach face, and because the stratified deposits dip seaward, many of the bedding surfaces can be treated as ancient shore-face and beach-face surfaces (Fig. 2). Each such bedding surface represents an ancient shoreline profile (Fig. 2B). We propose that the deposits formed during and after the time when Holocene sea level reached its maximum (Fig. 2B). When relative sea level was highest, the shoreline was near the base of the basaltic hills and the coastal plains had not yet been developed. Then, as sea level dropped, the shore-face and beach-face sediments prograded seaward to build the low coastal plains. This process continues to the present. Where berm deposits are negligible, the coastal plain surface can be treated as a track of falling MHTL (Fig. 2B). The ancient sea level position, hence, can be easily derived by subtracting half of the mean tidal range from the altitude of coastal plain surface.
RADIOCARBON SAMPLES
Sample Position
To minimize possible tidal differences among sample
lo-FIG. 2. (A) Generalized cross-sections of the modern coast at Penghu. calities, samples for radiocarbon dating were collected as (B) Schematic cross-section showing the relationship between modern near as possible to Makung Harbor, whose tidal record was ground surface and ancient shore face at Penghu. Inclined dotted lines
used as a vertical datum. Twelve carbonate samples were schematically represent ancient shore-face profiles.
collected from seven localities (Fig. 1B). The localities are Chihkan, Chinglo, Lintou, Sokang, Fengkuei, Tashihpi, and Kupoyu, shown in Figs. 1B and 3A to 3G.
surfaces and shoreline angles are mainly 2 – 4 and 5 – 6 m, Using a tripod-mounted level, we measured coastal plain respectively. Terrace widths range from several meters to heights where we collected14
C samples, assuming tide level several hundred meters. The widest terraces are located be- at the time of leveling as reference zero. Then we adjusted side modern marine embayments, such as the embayment at the measured heights to a common datum by using the mean
Chihkan (Fig. 3A). water line of Keelung Harbor, northern Taiwan. Each
sur-veyed level represents an ancient MHTL, which corresponds to the sample age that will be reported later. The leveling Sedimentary Environment
precision is about{10 cm according to our tentative surement. The local tidal record used in this study was mea-The coastal plains are composed of marine deposits of
sured at Makung Harbor, by the Central Bureau of the Minis-variable thickness. The thickness can exceed 10 m but more
try of Transportation and Communication, Republic of commonly is less than 5 m. The major constituents of the
China. Holocene deposits are detrital bioclastics, including corals,
mollusks, and foraminifers. This sediment composition
indi-Age Data cates predominantly marine sources. In outcrop most of these
marine deposits are well sorted. Planar beds and large-scale The conventional ages, displayed in Table 1, were deter-(i.e., several meters) trough cross-beds are the dominant sed- mined by the 14
C dating laboratory of the Precision Instru-imentary structures (Fig. 2A). Stratification dips gently sea- ment Development Center, National Science Council, Re-ward (Fig. 2A) and is laterally continuous. Lenses of poorly public of China, located in the Department of Geology, Na-sorted bioclastic debris and basaltic gravel are present but tional Taiwan University. Sample qualities were all checked less common. The sedimentary environment of the stratified by X-ray diffraction analysis to assure the purity, and the deposits is best interpreted as upper shore face to beach face. aragonite content is expressed in Table 1 asú95% if calcite peaks were not found. Corrections for reservoir and initial Storm berms may be represented by poorly sorted deposits.
258 CHEN AND LIU
FIG. 3—Continued
TABLE 1
14
C Ages from the Penghu Islands
Sample Depth 14
C Age Calibrated Age composition
Sample NTU Laboratory Locality Longitude/latitude (m) (yr B.P.)a
(cal yr B.P.)b
Sample typec
(Aragonite %) PH770710A NTU-1148 Chinglo 119738*499/ 0.8 4190{50 4110 – 4400 Coral (D) ú95
23736*119
PH-BKA NTU-1178 Paikeng 119738*579/ 0.5 4510{50 4530 – 4820 Mollusks (b, c) ú95 23735*379
PH-CKA NTU-1175 Chihkan 119734*389/ 2.8 4340{50 4340 – 4570 Mollusks (b) ú95 23740*069
PH-CKC NTU-1176 Chihkan 119734*409/ 2.8 4490{50 4510 – 4810 Mollusks (b) ú95 23740*059
PH790404A NTU-1342 Chihkan 119734*469/ 1.7 1940{40 1380 – 1560 Mollusks (b, c) ú95 23740*259
PH-FKA NTU-1180 Fengkuei 119732*179/ 0.7 1890{50 1310 – 1530 Mollusks (b, c, l) ú95 23732*269
PH-GBP NTU-1182 Kupoyu 119732*479/ 0.8 900{50 440 – 560 Mollusks (b) ú95 23742*499
PH800831A NTU-1501 Tashihpi 119734*329/ 0.2 2020{40 1480 – 1680 Mollusks (b, c, l) ú95 23735*459
PH800905A1 NTU-1498 Lintou 119737*479/ 2.1 3630{30 3440 – 3590 Mollusks (b) ú95 23733*389
PH800905A2 NTU-1499 Lintou 119737*479/ 2.0 3510{40 3310 – 3460 Coral (B) ú95 23733*389
PH800905A3 NTU-1502 Lintou 119737*479/ 2.0 4720{40 4830 – 5040 Coral (D) ú95 23733*389
PH800908B NTU-1510 Sokang 119735*469/ 1.8 1810{40 1270 – 1410 Mollusks (b,c) ú95 23731*459
aAll the conventional ages were calculated using the14C half-life of 5568 yr and were corrected for mass fractionation of carbon isotopes by normalizing
the d13C values of the samples to025 ‰ relative to PDB, an international standard.
bAge data were calibrated according to Stuiver and Braziunas (1993). We assume a standard reservoir correction (DRÅ0) and an error multiplier of one (kÅ1). The calibrated age ranges represent two standard deviations under phase assumptions.
c(D) represents dome-type coral and (B) branching-type. Mollusks used in this study are Barbatia bicolorata (b), Cardita variegata (c) and Lioconcha
sp. (I).
value effects, proposed by Stuiver and Braziunas (1993), low porosity should be more subject to contamination than were also applied to the 14
C ages of the samples analyzed. those of high porosity.
The final results are presented in Table 1 with the age unit Samples from three different types of organisms (PH8-of calibrated years before 1950 (cal yr B.P.). 00905A1, mollusk; PH800905A2, dome-coral; PH80090-Samples collected at the shoreline angle gave greater ages 5A3, branching coral; see Fig. 3C), all from one site at (i.e., PH-CKA, PH-CKC, and PH-BKA; see Figs. 3A and Lintou, were dated to test what type of carbonate sample 3B) than those samples collected at the outer edge (i.e., yields the most reliable estimate of depositional time. The PH790404A, PH800908B, PH-FKA, and PH-GBP; see Figs. dome-type coral sample gave an age older than the other two 3A, 3D, 3E, and 3G). This age distribution pattern supports types, probably because of its long lifespan and its possible our model of deposition during shoreline recession. recycling from older deposits. Sometimes dome-type corals can grow, either continuously or intermittently, for 1000
Sample Selection years. Some shore-face clasts may consist of the inner
por-tion of such long-lived corals. If the dated dome-coral sample The major sample type used in this study is aragonitic
came from the core of the coral, the sample could be centu-shell of mollusks that have relatively short lifespan and
ries older than its time of deposition. On the other hand, low porosity. We conservatively estimate a typical
life-larger debris can survive after several depositional cycles. span shorter than five years because our dated samples
Even if measured on a short-lived organism, the sample age are all less than 5 cm in diameter (Krantz et al., 1984).
can be much older than its time of deposition. Besides, it is Samples from organisms with short lifespans should
pro-vide a closer range of 14
260 CHEN AND LIU
TABLE 2
Modern Altitudes of Ancient Sea Level
Ground surface Calibrated14
C Age altitude (m) at Altitude Modern altitudes of
Locality (cal yr B.P.) sample positiona
correction (m) Paleo sea level (m)
Paikeng 4530 – 4820 3.3{0.1 0.9 2.4{0.1 Chihkan 4510 – 4810 3.1{0.1 0.9 2.2{0.1 Chinglo 4110 – 4400 3.0{0.1 0.9 2.1{0.1 Lintou 3440 – 3590 3.0{0.1 0.9 2.1{0.1 Tashihpi 1480 – 1680 1.3{0.1 0 1.3{0.1 Chihkan 1380 – 1560 1.4{0.1 0.9 1.5{0.1 Fengkuei 1310 – 1530 2.5{0.1 0.9/(0.2{0.1) 1.4{0.2 Sokang 1270 – 1410 2.2{0.1 0.9 1.3{0.1 Kupoyu 440 – 560 2.3{0.1 0.9/(0.7{0.1) 0.7{0.2
aUncertainties from measurement of modern ground altitudes do not exceed 10 cm.
Samples collected from mollusks and branching corals, ent heights of modern berms at Fengkuei and Kupoyu several times during 1991 and 1992 and obtained their average berm which both have short lifespans, recorded a younger and
more realistic deposition age. Due to their original small size heights of 0.2 { 0.1 and 0.7 { 0.1 m, respectively. The difference in berm heights is probably related to local wave and irregular shape, these organisms sustain fewer cycles of
erosion before being completely destroyed. The branching energy during deposition.
A thin lenticular unit at Tashihpi resembles modern depos-coral sample shows a slightly younger age than that of the
mollusk shell, which may be due to pores containing impuri- its along the coast of Penghu Bay, which has very low wave energy (Fig. 3F). The modern deposits contain only a small ties that went undetected by X-ray diffraction.
Only complete shells were analyzed to minimize dating amount of detrital carbonate that forms a lens near MTL. Based on this analogy, we interpret the height of the lenticu-of reworked, old material. Two complete mollusk samples
(i.e., PH-CKA, PH-CKC; see Fig. 3A), separated several lar unit at Tashipi as indicating ancient MTL. Therefore, no correction would be needed to its leveled altitude (Table 2). meters from one another along the same horizon, were
col-lected at Chihkan to check for repeatability. The 14
C ages The modern mean tidal range measured from Makung determined for these two samples are statistically
compara-ble (Tacompara-ble 1).
To further reduce possible errors from dating of trans-ported shells, we dated only three species of mollusks: Bar-batia bicolorata, Cardita variegata and Lioconcha sp. Sam-ples of Cardita variegata and Lioconcha sp. were used only to complement Barbatia bicolorata where specimens of the latter species were too small to yield a reliable 14
C age. Because all three mollusk species belong to the same reef environment (Kira, 1965), which is adjacent to the upper shore face in Penghu Islands (Fig. 2A), dated shells need not have traveled far or long before deposition. In addition, each species has a large, thick shell that provides enough material for dating.
HOLOCENE SEA LEVEL CHANGE
FIG. 4. Age – altitude plot of Holocene sea level in the Penghu Islands. As inferred above, the ground surface of the coastal plain
Eight shaded rectangular boxes represent ancient sea level positions, whose approximates ancient positions of MHTL at Paikeng,
Chih-ages are derived from mollusk shells. Arrow shows age correction of dome-kan, Chinglo, Lintou, and Sokang; berm deposits are not
type coral from Chinglo (see text). All boxes cover the total error range of developed in these areas. However, at Fengkuei and Kupoyu, each sample. The total error is composed of age error (2s) and altitude the ground surface probably represents an ancient berm crest, measuring error. The smooth curve, drawn to intersect all the sample boxes,
represents a best-guess Holocene sea level trace. comparable to modern berms nearly. We measured the
Harbor is 1.8 m (1991 and 1992 tidal record, reported by sea level maximum. These results support the hypothesis of Walcott (1972) that postglacial sea levels may differ with the Central Bureau of the Ministry of Transportation and
Communication, Republic of China). If the tidal range geographical setting. around Makung Harbor has been relatively constant for the
past several thousand years, the approximate difference be- CONCLUSION
tween MHTL and MTL has been close to 0.9 m. We use
Based on geologic and neotectonic evidence, the Penghu this value to correct the ancient MHTL to ancient MTL in
Islands have been tectonically stable, both since late Mio-the oMio-ther places that lack berm deposits. At Mio-the Fengkuei
cene and during Holocene. Thus, it is an appropriate location and Kupoyu localities, we further corrected for the berm
to evaluate Holocene sea level change in this part of the height based upon their modern values, 0.2 {0.1 and 0.7
western Pacific. Based on a progradational deposition model,
{0.1 m respectively (Figs. 3E and 3G; Table 2). Based on
the altitude and age of nine ancient sea levels have been these parameters, the modern altitudes of ancient sea levels
reconstructed by leveling the ground surface of the coastal have been calculated (Table 2).
plains and by determining the ages of the coarse-grained The age of 4110 – 4400 cal yr B.P. for a dome-type coral
detrital bioclastics found within the coastal plains. The high-located near Chinglo is probably older than its depositional
est sea level in the Penghu Islands occurred about 4700 years time. We apply a correction of 1100 years (Fig. 4), based
ago at roughly 2.4 m above present sea level. Afterward, on the age difference observed between the dome-type coral
sea level appears to have fallen to its present position without sample (PH800509A1) and the mollusk sample
(PH80050-large fluctuations. The different peak times of Holocene sea 9A3) at Lintou.
levels between the Penghu Islands and nearby areas demon-Nine ancient sea level records — two from Chihkan and
strate that Holocene sea level changes were variable among one from each of seven other localities — are plotted in Fig.
different geographic localities. 4. The records show that relative sea level reached a middle
Holocene maximum about 2.4 m above that of the present
day and that it subsequently fell to modern level without ACKNOWLEDGMENTS
any fluctuation large enough for us to detect.
We thank Dr. Jack C. L. Liu and Dr. Kenneth Ridgway for their valuable comments and help in polishing the draft. We also thank Dr. Arthur L.
DISCUSSION Bloom and an anonymous referee for suggestions in content and writing
of this paper, and Miss Hwa-Wen Chen for fossil identification.
Holocene Sea Level Maximum
REFERENCES
Because the most landward of the shore-face deposits
gives the oldest ages, of about 4700 cal yr B.P., the highest Adey, W. H. (1978). Coral reef morphogenesis: A multi-dimensional model. Holocene sea level in Penghu Islands probably occurred Science 202, 831 – 837.
about 4700 years ago. This age agrees with previous findings Chow, J., Chen, H. M., Chang, T. Y., Kou, C. L., and Tsai, S. F. (1991). by Nakada (1986), who reported a high sea level from 4000 Preliminary study on hydrocarbon plays around Nanjitao basin. Taiwan
Strait. Petroleum Geology of Taiwan 26, 45 – 56. to 6000 yr B.P. in the central and western Pacific regions.
Fairbridge, R. W. (1961). Eustatic changes in sea level. Physics and Chemis-The 2000-year variation in peak time may reflect the local
try of the Earth 5, 99 – 185. properties of the Earth’s crust (Walcott, 1972; Nakada,
Fujii, S., and Fuji, N. (1967). Postglacial sea level in the Japanese islands. 1986).
Journal of Geoscience, Osaka City University 10, 43 – 50. After calibrating the ages reported by previous workers
Haq, B. U., Hardenbol, J., and Vail, R. P. (1987). Chronology of fluctuating in the same way as has been done in this study, we find two
sea levels since the Triassic. Science 235, 1156 – 1167. end members for the timing of the sea level maximum. One
Ho, C. S. (1982). ‘‘Tectonic Evolution of Taiwan.’’ Ministry of Economic end member is from the neighboring areas on the continental Affairs, Taipei, Taiwan, Republic of China.
side of Penghu, such as eastern China coast (Huang et al., Hsiao, P. T., Hu, C. C., Lin, K. A., Hsu, S. H., Fuh, S. C., Chang, T. Y., 1987; Pirazzoli, 1991), Japan (Fujii and Fuji, 1967; Sugi- Hsiuan, T. H., Sheen, H. C., Kuo, C. L., and Lee, C. J. (1991). Hydrocar-bon potential evaluation of the Penghu Basin. Petrol. Geol. Taiwan 26, mura, 1977), and Vietnam (Morner, 1983). In these areas,
215 – 229. [In Chinese] the Holocene sea level maximum occurred at around 6000
Huang, Z. G., Li, P. G., Zhang, Z. G., and Zong, Y. Q. (1987). Sea level years ago. The other end member is from the areas on the
changes along the coastal area of south China since late Pleistocene. In oceanic side of Penghu — mid-ocean islands (Kayanne et al.,
‘‘Late Quaternary Sea level Changes’’ (Y. Qin and S. Zhao, Eds.), pp. 1988; Miyata et al., 1988; Sugimura et al., 1988; Pirazzoli, 142 – 154. China Ocean Press.
1991) and the southern Ryukyu Islands (Ota et al., 1985). Juang, W. S. (1988). Geochronology and chemical variations of late Ceno-In these areas, the Holocene sea level maximum occurred zoic volcanic rocks in Taiwan. Ph.D. dissertation, Institute of
Oceanogra-phy, National Taiwan University. [In Chinese] at around 4000 years ago. The Penghu Islands are
Geomor-262 CHEN AND LIU
phic and geologic development of Holocene emerged reefs on Rota and Ota, Y., Pirazzoli, P. A., Kawana, T., and Moriwaki, H. (1985). Late Holo-cene coastal morphology and sea level records on three small islands, Guam, Mariana Islands. In ‘‘Sea Level Changes and Tectonics in the
Middle Pacific,’’ Report of the HIPAC Projection 1986 and 1987, pp. the South Ryukyus, Japan. Geographical Review of Japan 58(B), 2, 185 – 194.
35 – 57. Department of Geography, University of Tokyo.
Peltier, W. R., Farrell, W. E., and Clark, J. A. (1978). Glacial isostasy and Kern, J. P. (1977). Origin and history of upper Pleistocene marine terraces,
relative sea level: A global finite element model. Tectonophysics 50, 81 – San Diego, California. Geological Society of America Bulletin 88, 1553 –
110. 1566.
Pinot, J.-P. (1979). Les indicateurs deceles par des methodes oceanograph-Kidson, C. (1982). Sea level changes in the Holocene. Quaternary Science
iques, geophysiques ou geochimiques. Oceanis 5, 335 – 355. Reviews 1, 121 – 151.
Pirazzoli, P. A. (1991). ‘‘World Atlas of Holocene Sea Level Changes.’’ Kira, P. (1965). ‘‘Shells of the Western Pacific in Color,’’ Vol. I. Hoikusha,
Elsevier Oceanography Series, Vol. 58. Elsevier, Amsterdam. Japan.
Shepard (1963). Thirty-five thousand years of sea level. In ‘‘Essays in Krantz, D. E., Jones, D. S., and Williams, D. F. (1984). Growth rates of
Honor of K. O. Emery,’’ pp. 1 – 10. Univ. of Southern California Press, the sea scallop, Placopecten maggellanicus, determined from the18O/16O
Los Angles. record in shell calcite. Biological Bulletin 167, 186 – 199.
Stuiver, M., and Braziunas, T. F. (1993). Modeling atmospheric14
C influ-Lee, C. Y. (1990). Late Miocene paleogeography reconstruction of Penghu
ences and14
C ages of marine samples to 10,000 BC. Radiocarbon 35(1), Islands. Presented at the 1990 Annual Meeting of the Geological Society
137 – 190. of China; abstracts with Program, p. 24. [In Chinese]
Sugimura, A. (1977). Ice sheets, continents and oceans. Kagaku 47, 749 – Lee, C. Y., and Chen, C.-H. (1992). Temporal and spatial variation of 755. [In Japanese]
basalt in Penghu Islands. Presented at the 1992 Annual Meeting of the
Sugimura, A., Maeda, Y., Matsushima, Y., Rodda, P., and Matsumoto, E. Geological Society of China, abstracts with Program, p. 92. [In Chinese]
(1988). Lobau lowland, Viti Levu, Fiji. In ‘‘Sea Level Changes and Lee, C. Y. (1994). Chronology and geochemistry of basaltic rocks from Tectonics in the Middle Pacific,’’ Report of the HIPAC Project in 1986 Penghu Islands and mafic dikes from east Fujian: Implications for the and 1987, pp. 59 – 65. Department of Geography, University of Tokyo. mantle evolution of SE China since late Mesozoic, p. 226. Ph.D. disserta- Sun, S. C. (1982). The Tertiary basins of offshore Taiwan. In (A. Salivar-tion, National Taiwan University. [In Chinese] Sali, Ed.), pp. 125 – 135. Technical Programme Committee, ASCOPE. Lin, C. C. (1967). Geology of the offshore islands of Taiwan. Quar. Jour. 1981,
Bank of Taiwan 18(4), 229 – 256. [In Chinese] Teng, L. S. (1987). Stratigraphic records of the late Cenozoic Penglai Orog-eny of Taiwan. Acta Geol. Taiwanica 25, 205 – 224.
Liu, T. K. (1989). Radiometric dates of rocks and sediments from Penghu
Islands and Lutao, and their implication of vertical crustal movement. Ting, F. T. C. (1990). Subsidence in an uplifting environment: Case studies Technical report, Energy & Resources Laboratories of Industrial Technol- in Taiwan. Proceedings of the Geological Society of China 33(1), 1 – 7. ogy Research Institute. Tsai, Y. P., Teng, T.-L., Chu, J. M., and Liu, H. L. (1977). Tectonic implications of the seismicity in the Taiwan region. Memoir of the Geo-Miyata, T., Maeda, Y., Matsumoto, E., Matsushima, Y., Rodda, P., and
logical Society of China 2, 13 – 41. Sugimura, A. (1988). Emerged notches and micro-atolls on Vanua Levu,
Fiji. In ‘‘Sea Level Changes and Tectonics in the Middle Pacific,’’ Report Tsai, Y. P. (1986). Seismotectonics of Taiwan. Tectonophysics 125, 17 – of the HIPAC Project in 1986 and 1987, pp. 67 – 76. Department of 37.
Geography, University of Tokyo. Walcott, R. I. (1972). Past sea levels, eustasy and deformation of the Earth. Quaternary Research 2, 1 – 14.
Morner, N. A. (1983). Differential Holocene sea level change over the
global evidence for glacial eustasy, geoidal eustasy and crustal move- Woods, A. J. (1980). Geomorphology, deformation, and chronology of ments. In ‘‘Abstracts of International Symposium on the Coastal Evolu- marine terraces along the Pacific coast of central Baja California, Mexico. tion in the Holocene,’’ pp. 93 – 96. Japan Society for the Promotion of Quaternary Research 13, 346 – 364.
Science, Tokyo. Yang, K. M., Ting, H. H., and Yuan, J. (1991). Structural styles and tectonic modes of Neogene extensional tectonics in southwestern Taiwan: Impli-Nakada, M. (1986). Holocene sea levels in oceanic islands: Implications
for the rheological structure of the Earth’s mantle. Tectonophysics 121, cations for Hydrocarbon exploration. Petroleum Geology of Taiwan 26, 1 – 31.
263 – 276.
