- Email: [email protected]
Pedogenic correlation of lateritic river terraces in central Taiwan
Geomorphology 88 (2007) 201 – 213 www.elsevier.com/locate/geomorph
Pedogenic correlation of lateritic river terraces in central Taiwan Heng Tsai a,⁎, Wen-Shu Huang a , Zeng-Yei Hseu b b
a Department of Geography, National Changhua University of Education, Changhua 50007, Taiwan Department of Environmental Science and Engineering, National Pingtung University of Science and Technology, Pingtung 91201, Taiwan
Received 12 June 2006; received in revised form 6 November 2006; accepted 6 November 2006 Available online 13 December 2006
Abstract The Pakua, Chushan and Touliu terraces are three groups of lateritic river terraces resulting from different fluvial systems in central Taiwan. Their geomorphic features do not enable regional correlations to be made. In this study, an alternative scheme based on the quantification of a pedogenic development is proposed for this correlation. Four soil pedons from the Chushan and Touliu terraces are classified as Ultisols (CS-1, CS-2 and TL-1) and an Inceptsol (TL-4) following the Soil Taxonomy. An additional six soil pedons, including an Oxisol (PK-1), Ultisols (PK-2, -3, -4 and -5), and an Inceptsol (PK-6), from the Pakua terraces were included from a previous study [Tsai, H., Huang, W.S., Hseu. Z.Y., Chen, Z.S., 2006. A river terrace soil chronosequence of the Pakua tableland in Taiwan. Soil Science 171, 167–179.]. The weighted mean profile development indices (WPDI) for the pedons suggest a chronological order of I (PK-1), II (PK-2), III (PK-3), IV (PK-4, CS-1 and TL-1), V (PK-5 and CS-2), VI (PK-6) and VII (TL-4). The correlation suggests the Chushan and Touliu terraces correspond to the same surface of an anticline, separated by the Chinshui River. Surface deformation reflects the growth of the underlying structure in which uplift folding occurred in the hinge zone and tilted to the east along the fold limb, as a result of the southern extension of the Pakua anticline. However, the different geometries on both the surface and subsurface from southern Pakua Anticline indicate a mechanism of fault-bend folding for Touliu Hill. © 2006 Elsevier B.V. All rights reserved. Keywords: River terrace; Profile development index; Pedogenesis; Soil classification; Anticline; Quaternary
1. Introduction River terraces are one of the most prominent geomorphic features in Taiwan, and constitute an excellent marker for the study of active tectonics. However, the absence of absolute datings and geochronological information means that ages are typically a problem in extracting information from deformed surfaces. Previous studies have correlated terraces mainly on the basis of ⁎ Corresponding author. No.1, Chinte Road, Changhua 50007, Taiwan. Tel.: +886 4 7232105 2819; fax: +886 4 7211186. E-mail address: [email protected] (H. Tsai). 0169-555X/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.geomorph.2006.11.004
geomorphic and pedogenic parameters such as altitude above sea level or present river channel, continuity and height of terrace scarp, surface gradient and color of topsoil (Lin, 1957; Ku, 1963; Yang, 1986). However, their results are usually inconsistent and are not satisfactory. Jenny (1941) indicated that most soil properties are time-dependent chronofunctions, and some of them have been used to correlate Quaternary successions (Morrison, 1968; Leamy et al., 1973; Mulcahy and Churchward, 1973; Birkeland, 1984; Engel et al., 1996). The activity (Feo/Fed) and crystallinity [(Fed − Feo)/Fet] ratios of free iron are good time-dependent indexes for
202
H. Tsai et al. / Geomorphology 88 (2007) 201–213
red soils (Blume and Schwertmann, 1969; Nagatsuka, 1972), and have been successfully used in northern Italy (Arduino et al., 1984, 1986), southern California (MacFadden and Hendricks, 1985) and the Kikai Island of southwestern Japan (Maejima et al., 2002). Harden's (1982) profile development index (PDI) has been proven as another successful method in quantifying field attributes of soil in order to rank soil development (Alonso et al., 1994). Moreover, Harden and Taylor (1983) have demonstrated that the index may be suitable for correlating soil chronosequences on different parent materials and in different climates. Although Vidic (1998) questioned the validity of the chronofunction, it was agreed that the parent material composition does not affect the soil development index significantly. There are many lateritic river terraces in central Taiwan. The Pakua, Chushan and Touliu terraces represent three different groups of river terraces developed in various drainage basins. No satisfactory regional correlation of terraces can be achieved based on the geomorphic parameters mentioned above. Tsai et al. (2006) reported a soil chronosequence of the Pakua terraces, and estimated the ages by the ratio of (Fed − Feo)/Fet. In this study, the weighted mean of the PDIs (hereafter termed WPDI) and the pedogenic iron are calculated and compared as a basis for regional terrace correlation. The rate
of soil development based on the WPDIs is proposed as an effective chronological indicator. 2. Study area 2.1. Geologic background Taiwan is geologically located at the convergent boundary between the Philippine Sea plate and the continental margin of the Eurasian plate (Fig. 1A). The Philippine Sea plate subducts beneath the Eurasian plate offshore eastern Taiwan, and overrides the South China Sea floor of the Eurasian plate south of Taiwan (Suppe, 1981; Tsai, 1986; Angelier, 1986; Ho, 1988). Active tectonics in Taiwan are manifested by continuing shortening and widespread seismic activity. A shortening rate of 8.5 cm/yr is revealed by GPS measurements across the 200-km-wide island (Seno et al., 1993; Yu et al., 1997). There are a series of active subparallel west-verging thrusts in the fold-and-thrust belt along the western Foothills of Taiwan, resulting from a thrust migrating westward by tectonic compression since the Plio–Pleistocene (Lee et al., 1996) (Fig. 1B). The largest inland earthquake (Mw = 7.6, ML = 7.3) of the 20th century in Taiwan was caused by the displacement of the Chelungpu Fault (Ma et al., 1999; Kao and Chen, 2000).
Fig. 1. (A) The tectonic setting of Taiwan. (B) Geological map and cross-section of the studied area in central Taiwan.
H. Tsai et al. / Geomorphology 88 (2007) 201–213
The Changhua Fault, as the active frontal thrust of the fold-and-thrust belt, results in an asymmetric anticline (also known as Pakua Tableland) with a gentler east limb (less than 10°) and a steeper west limb (up to 30°) (Lu et al., 1998; Wang et al., 2003) (Figs. 1B and 2). There is a basement high, the Peikang High, located west of the fold-and-thrust belt (Biq, 1992; Lu et al., 1998). Its obstacle effect results in the deformation of the tableland in a sigmoidal shape trending NNW–SSE in the north, and N–S in the south (Lee et al., 1996; Mouthereau et al., 1999; Sung and Chen, 2004). Leftlateral strike slip faulting associated with transtensional deformation were identified from the microstructures in the northern part of the anticline (Lee et al., 1996; Delcaillau et al., 1998; Mouthereau et al., 1999). No clear outcrops of the Changhua Fault have been found in the field (Lin et al., 2000a). Yang (1986), based on aerial photo interpretation, suggested three segments of the fault along the west of the Pakua Tableland.
203
Seismic and borehole evidence indicates the Changhua Fault as a “blind-thrust” buried underground (Hsiao, 1968; Chen, 1978; Wang et al., 2003). The balanced cross-section in Deffontaines et al. (1997) indicates that the Pakua Anticline has accommodated about 2–3 km of shortening at the southern end and 4–5 km at the northern end since 0.4–0.5 Ma. According to the available geologic maps, the Changhua Fault extends along the west of the Pakua Tableland and the Touliu Hill (CPC, 1982; CGS, 2004). Touliu Hill is an anticline that plunges to the north, where the topography does not comply with the geometry of the underground structure. The relatively symmetrical form of surface is underlain by the strata of west-limb (50–80°) and east-limb (10– 40°) (CGS, 2004; Liu, 2004). Shallow seismic reflection profiles show highly bending layers without any trace of fault offsets along the Touliu Hill (Liu, 2004). Ota et al. (2002) argued that the south segment of the fault extends along the east side of the Chinshui river valley, a
Fig. 2. (A) The geographic distribution of the Pakua, Chushan and Touliu terraces. (B) Enlarged map of the Touliu terraces. (C) The 3D relief of the study area (in 40 × 40 m resolution) and two geomorphic profiles. The geologic structure along the BB′ profile was interpreted based on geological map of the CGS (2004).
204
H. Tsai et al. / Geomorphology 88 (2007) 201–213
Table 1 Geomorphic and pedologic characteristics of the Chushan, Touliu and Pakua terraces Terrace/ pedon Chushan CS-1 CS-2 Touliu TL-1 TL-4 Pakua5 PK-1 PK-2 PK-3 PK-4 PK-5 PK-6
Slope (%)
Altitude (m)1 a.s.l.
a.c.r.
2.1 0.7
250–160 190–150
130–50 70–30
0.6 1.7
330–280 240–230
3.8 3.3 2.8 2.1 1.4 1.4
440–375 430–330 410–250 385–320 330–240 325–170
Parent material2
Classification3
Pedologic correlation
Age4 (ka)
WRB
Taxonomy
PM1 PM1
Acrisol Acrisol
Paleudult Paleudult
IV V
92 31
190–180 133–100
PM2 PM2
Acrisol Cambisol
Paleudult Dystrudept
IV VII
∼ 92a, 78 ± 34b b30a
345–275 320–180 310–140 295–225 230–130 145–45
PM1 PM1 PM1 PM1 PM1 PM1
Ferralsol Acrisol Acrisol Acrisol Acrisol Cambisol
Hapludox Paleudult Paleudult Hapludult Paleudult Dystrudept
I II III IV V VI
500–350 335 ± 77b 267 ± 65b 164 ± 48b, ~92a, ∼ 31a b30a
1: “a.s.l.” represent the altitude above sea level, “a.c.r.” represent the altitude above current river. 2: Sources of parent material from Choshui River (PM1) and Chinshui River (PM2). 3: WRB from ISSS Working Group WRB (1998), Taxonomy from Soil Survey Staff (2003). 4: Ages of CS-1, CS-2 from Ota et al. (2002), and the age of PK-1 in the range from Liew (1988), Mouthereau et al. (1999) and Tsai et al. (2006). The age with superscript “a” based on pedologic correlation, and superscript “b” interpolated from Fig. 8. 5: Soil data from Tsai et al. (2006).
northward flowing tributary of the Choshui River, indicating two different tectonic units between Pakua Tableland and the Touliu Hill (Fig. 2). These fold structures are mainly composed of the Toukoshan Formation, which constitutes synorogenic deposits (Ho, 1988; Lee et al., 1996). The east-dipping strata of Toukoshan Formation resulting in a monocline structure inclined at less than 10° beneath the river terraces located east of the Chinshui River (Fig. 2). 2.2. River terraces and soils The Choshui River, the largest river in Taiwan, entrenched the fold-and-thrust belt in central Taiwan as an antecedent stream, and formed groups of a series of wide unpaired river terraces such as the Pakua and Chushan terraces (Lin, 1957) (Fig. 2). The Touliu terraces were developed by the Chinshui River, a south tributary of the Choshui River. The Pakua terraces are located on the southern part of the Pakua Tableland. The Touliu and Chushan terraces are separated from the Pakua terraces by the Choshui River, and are respectively located on the east and west sides of the Chinshui River (Fig. 2). The very dark sediments (Munsell 5GY 3/1) of the Choshui River are mainly eroded from the slate, phyllite and quartzite of the Tertiary metamorphic terrain in the Central Range. The sediments of the Chinshui River are considerably lighter in color (5Y 4/ 2) and are derived from a non-metamorphic rocks of sandstone and shale bedrock.
The Pakua terraces can be divided into six levels, from PK-1 (highest) to PK-6 (lowest), in the order of descending altitude, indicating a long-term northward river migration (Yang, 1986) (Table 1; Fig. 2). However, the lowering of the river by erosion is insufficient to keep pace with the rapid uplift of the anticline, preventing the development of river terraces located lower than PK-6 (Yang, 1986; Tsai et al., 2006). The defeated river diverted around the nose of the anticline at the southern end of the tableland. Delcaillau et al. (1998) indicate that the uplift of the Pakua Anticline began about 400–500 ka, which gives a maximum age estimation for the oldest river terraces developed on the surface. Liew (1988) proposed, based on palynological analyses, that the terrace sediments were probably deposited during some high sea-level intervals between 70–350 ka. The Chushan terraces consist of four levels of river terraces that have become progressively eastward-tilted. The higher the terraces, the steeper of slope on the surface. Radiocarbon dating gives an age of about 31 ka for CS-2, and indicates an age of around 92 ka for CS-1 by assuming a constant rate of tilting (Ota et al., 2002). The later age is reasonably constrained because the tilting gives a maximum uplift rate of 2.7 mm/yr (250 m/ 92,000 yr) within the range estimated by Delcaillau et al. (1998) for the Pakua terraces. The Touliu terraces also consist of four levels of river terraces distributed along the hinge of the anticlinal hill. However, they are fragmented and small in size due to erosion. Most of the river terraces mentioned above are covered by red soils and gravels, and hence are designated as
H. Tsai et al. / Geomorphology 88 (2007) 201–213
205
“lateritic river terraces” by Lin (1957). Unfortunately, material for 14C or biostratigraphic dating is poorly preserved under highly oxidized and weathering conditions. It is generally believed that the red-colored soils in Taiwan could be developed prior to 30 ka (Liew, 1988; Chen and Liu, 1991; Sung et al., 1997; Ota et al., 2002). The oldest age for red soil was reported as ≤ 900 ka based on paleomagnetic correlation (Lee et al., 1999). Tsai et al. (2006) compared the (Fed − Feo)/Fet ratios of the Kikai Island soils, and suggest ages of 40–400 ka for the soils of the Pakua terraces. Liew (1988) indicated a lateritic weathering phase occurred in latest Pleistocene time about 140 ka ago, after the major transgression associated with the Last Interglacial event. 2.3. Climate and vegetation Taiwan has a tropical climate in the southern and western plains and subtropical climate in the northern and mountainous regions. The study area is located in the tropical zone, and the climate is characterized by high temperature and humidity, substantial rainfall and summer tropical cyclones. According to the Central Weather Bureau of Taiwan (http://www.cwb.gov.tw), the mean annual temperature and precipitation for the area are 23 °C and 1642 mm/yr respectively. The seasonal oscillations in precipitation, with a dry winter and a wet summer, comprise a wet–dry tropical climate (Cwa) based on Köppen's climate classification. Subtropical broad-leaved evergreen forests are the predominant vegetation in this area. The paleoclimate is characterized by a warm–cold– warm cycle during glacial–interglacial periods of the Pleistocene. The cold and dry phase that existed during the mid-Pleistocene is considered to be younger than the Jaramillo event (about 0.7–0.9 Ma) (Tseng et al., 1992; Liew and Huang, 1994). The warm–temperate to subtropical climatic conditions continued and were separated by three very humid intervals during the early and mid-Holocene (Chen and Liew, 1990; Liew and Hsieh, 2000). Palynological evidence indicates that a warmer annual temperature of 1–2 °C with a relatively higher amount of precipitation prevailed during the warm interval than in the cool–dry interval in mid-Pleistocene times (Liew and Hsieh, 2000). 3. Materials and methods A total of 4 pedons were sampled from the Chushan and Touliu terraces, and were labelled as their terraces for easy illustration (Fig. 2). The classification and sampling of each horizon of the soils was conducted following Soil
Fig. 3. Flow chart for deriving the PDI or WPDI (redrawn and modified from Harden and Taylor, 1983).
Survey Staff (1993). Soil samples ≤2 mm in diameter are prepared for physical and chemical analyses. The soil properties of another 6 pedons of the Pakua terraces are reported previously in Tsai et al. (2006). Based on the guidelines of Harden (1982) and Harden and Taylor (1983) with minor modification (Fig. 3), the PDIs of the pedons were calculated based on the following six properties: (i) rubification (moist color hue and chroma from the Munsell color chart); (ii) total texture (texture class plus type of stickiness and plasticity of the wet consistence); (iii) structure (type and degree of development of soil structure); (iv) moist consistence (class); (v) dry consistence (class); and (vi) clay coatings (abundance, thickness, and location). The pH and color values (lightening or darkening) of soil used in Harden's (1982) original method were not included in our calculation. The pH was omitted because of its susceptibility to the disturbance by human activities, i.e. ploughing and fertilization in the study area. The color value of the soils has been considered an important soil property mainly in the early stages of pedogenesis because organic matter content seems to increase relatively rapid and reach a steady state faster than any other soil properties (Birkeland, 1999). The organic matter in soils will then decrease to a very low content after long-term development, leading to an insignificant variation in color value. As a result, the color value is also ignored in the calculation. The index values are calculated by comparing the soil properties of a horizon with those of the assumed parent
206
H. Tsai et al. / Geomorphology 88 (2007) 201–213
Touliu terraces (Table 1). The PDI values were further divided by the depth of the described soil profile to produce WPDI normalized values ranging between 0 (no development) and 1 (maximum development) (Birkeland, 1999). In addition to soil morphology quantified in the field, particle size distribution was determined in the laboratory using the pipette method (Gee and Bauder, 1986). Total Fe (Fet) was determined by the Reisenauer (1982) method. Free Fe (Fed) was extracted by the dithionite– citrate–bicarbonate (DCB) method (Mehra and Jackson, 1960). Amorphous Fe (Feo) was extracted by 0.2 M ammonium oxalate (pH 3.0) (McKeague and Day, 1966). All the Fe in solution was determined by atomic absorption spectrometry (Hitachi Z-8100 type, Japan). Fig. 4. Particle size distribution of the pedons and the soil texture classes (soil data of the Pakua terraces from Tsai et al., 2006).
material. The sediments of the present active floodplains were used to represent fresh and unoxidized parent material of the soils. The soils containing fragments of detrital slate or phyllites observed in the Chushan terraces indicate that the parent material was deposited by the Choshui River. In contrast, the alluvium from the Chinshui River represents the parent material of the
4. Results 4.1. Soil characteristics and classification The soils graded from yellow to reddish in hue, showing a trend that roughly agrees with the ascending altitude of terraces (the higher the terraces, the richer the red color of the soils). Consequently, the highest redness index (RI) is found in pedon TL-1, while the least
Fig. 5. The variations of soil profile distribution in clay contents (shaded region) and Fed (solid circle).
H. Tsai et al. / Geomorphology 88 (2007) 201–213
207
Table 2 The morphological characteristics and the quantification of the studied pedons Pedon Horizon Depth (cm) CS-1
CS-2
TL-1
TL-4
PM1 PM2
A Bt1 Bt2 Bt3 Bt4 A Btv1 Btv2
0–20 20–45 45–85 85–120 120–150 0–10 10–25 25–70
Texture a Structure b Consistence c
Munsell color Moist
10YR 4/4 10YR 4/6 7.5YR 6/8 7.5YR 6/8 7.5YR 7/8 10YR 6/6 10YR 5/6 10YR 5/5 (95%) 2.5YR 6/4 (5%) Btv3 70–95 10YR 5/6 (90%) 2.5YR 6/4 (10%) Btv4 95–125 10YR 5/6 (85%) 2.5YR 6/4 (15%) A 0–35 7.5YR 5/6 Bt1 35–65 7.5YR 4/6 Bt2 65–100 7.5YR 4/4 Bt3 100–125 5YR 4/6 Bt4 125–155 5YR 4/4 Bt5 155–180 5YR 4/6 A 0–10 10YR 3/4 AB 10–20 10YR 3/4 Bw1 20–35 7.5YR 3/4 Bw2 35–55 7.5YR 3/4 (Choshui river) 5GY 3/1 (Chinshui river) 5Y 4/2
Dry
Dry Moist Wet
Clay RI e coatings d
HI f
0.39 0.56 (PM1) 0.47 0.61 0.61 0.61 0.44 0.49 (PM1) 0.44
10YR 5/4 10YR 6/6 7.5YR 7/6 7.5YR 7/6 7.5YR 7/6 10YR 6/6 10YR 7/6 10YR 6/6
CL SiCL SiC C C CL CL CL
2f&msbk 2f&mabk 2m&cabk 2m&cabk 2m&abk 2f&msbk 2f&mabk 2f&mabk
sh h h h h sh sh h
fr fir fir fir fir fir fir fir
ssps ssps sp vsp vsp ssps ssps ssps
–h – 1dpf 1dpf 1dpf – – 1fpf
0 0 3.33 3.33 2.86 0 0 0
10YR 6/6
SiC
2m&cabk
sh
fr
sp
1fpf
0
7.5YR 6/6 SiC
2m&cabk
sh
fr
sp
1fpf
7.5YR 5/4 7.5YR 6/4 7.5YR 6/6 5YR 5/6 5YR 5/6 5YR 5/6 10YR 4/4 10YR 4/4 7.5YR 4/6 7.5YR 4/4 5GY 4/1 5Y 5/2
1f&msbk 2f&mabk 3f&mabk 3f&mabk 3m&cabk 3f&mabk 1fgr 1f&msbk 1f&msbk 2f&msbk
sh sh vh h vh vh lo sh sh sh
fr fr fir fir vfir vfir lo fr fr fr
sp sp sp sp vsvp vsvp sopo sopo sp sp
– 1fpf 1fpf 1dpf 2dpf 2dpf – – – –
WPDI g
0.48 0.53
CL SiCL CL C C C SL SCL SCL SC
3 3.75 2.5 7.5 5 7.5 0 0 3.33 3.33
0.53 0.39 0.53 (PM2) 0.46 0.52 0.57 0.64 0.65 0.18 0.32 (PM2) 0.27 0.36 0.39
a
Using the US system; SL = sandy loam, C = clay, CL = clay loam, SiC = silty clay, SiCL = silty clay loam. sg = single grain, 3 = strong, 2 = moderate, 1 = weak, f = fine, vf = very fine, m = medium, c = coarse, gr = granular, abk = angular blocky, sbk = subangular blocky, vc = very coarse. c h = hard, lo = loose, fir = firm, fr = friable, v = very, w = weak, o = none, s = sticky, p = plastic. d 1 = few; 2 = common; f = faint; d = distinct; pf = ped face. e RI; Redness index = ((10 − hue) × chroma) / value (Torrent et al., 1980, 1983). f HI; Horizon index, PDI = The sum of HI multiplied by horizon thickness (Appendix 2 in Birkeland, 1999). g WPDI; Weighted mean of profile development index calculated based on parent material from the Choshui River (PM1) and the Chinshui River (PM2), WPDI = PDI divided by the pedon thickness (Appendix 2 in Birkeland, 1999). h No feature. b
reddish hue or lower RI values were found in pedons TL-4 and CS-2. Except for TL-4, all the soils exhibit fine soil texture with a sand fraction lower than 40% (Table 3). The wide range from sandy clay loam to clay in texture classes exhibits a decreasing trend of sand content with the altitude of the terraces (Fig. 4). Fine and very fine sands are the dominant subdivisions of the sand fraction divided into five classes, reflecting the fine and highly weatherable parent materials of the soils (Table 3). As a result, a high rate of soil development is expected. Pedons CS-1 and TL-1 show a similar particle size distribution, low contents of each sand fraction. This suggests an equally long period of weathering for the parent materials. The fine texture of the soils reflects the morphologies of structure and consistence. The considerable develop-
ment in structure and degree of stickiness and plasticity are in accordance with the abundant clay coatings on the pedsurface observed in the field. Strongly coarse angular and subangular blocky structures are generally found in CS-1, CS-2 and TL-1 pedons, while moderately finer ones are observed in TL-4 pedons. Clear accumulation of illuvial clay is identified from the faint or distinct clay coatings on the pedsurface of subsurface horizons in each pedon, except for pedon TL-4 (Fig. 5; Table 2). Additionally, high clay contents were found in all the soils related to strong chemical weathering (Table 3). The vertical variations of clay and silt along the profiles also indicate a strong degree of leaching and weathering within the soils (Fig. 5). However, the clay contents of the Chushan terraces are generally higher than those of the Touliu terraces. Based on the observed
208
H. Tsai et al. / Geomorphology 88 (2007) 201–213
Table 3 Particle size distribution of the studied pedons Pedon Horizon Size distribution (%)
Ultisols with the Subgroup of Typic Paleudults following the Soil Taxonomy (Soil Survey Staff, 2003). The soil developed on TL-4 is as Typic Dystrudepts because of the absence of clay coatings in its B horizons.
Sand-size class a
Sand Silt Clay VC% C%
M% F%
VF%
4.2. WPDI and pedogenic Fe
Chushan terraces CS-1 A 25 Bt1 18 Bt2 14 Bt3 16 Bt4 17 CS-2 A 38 Btv1 35 Btv2 28 Btv3 18 Btv4 14 B4 14
47 46 42 36 39 30 29 32 42 46 50
28 36 44 48 44 32 36 40 40 40 36
0.3 0.1 0 0.2 0.6 0 0 0 0 0 0
1.8 1.8 1.5 1.9 2.8 0 0.1 0.1 0 0 0.1
2.1 1.9 1.7 2.1 2.3 0.5 0.4 0.3 0.1 0.1 0.1
7.7 4.6 4.9 6.4 5.2 25.9 20.0 20.0 8.6 7.3 1.6
13.3 9.7 6.3 5.0 6.5 11.2 14.5 7.9 9.4 6.6 12.5
Touliu terraces TL-1 A Bt1 Bt2 Bt3 Bt4 Bt5 TL-4 A AB Bw1 Bw2
50 47 42 35 33 32 17 20 21 12
28 36 32 40 44 48 20 24 32 36
0.2 0.1 0 0.1 0.6 0 2.1 2.4 1.9 6.8
0.7 0.4 1.1 1.0 1.1 0.9 11.5 8.4 7.3 11.4
0.8 0.6 1.6 1.6 1.8 1.7 7.6 6.5 5.7 5.5
7.6 3.6 9.2 11.4 7.4 7.4 19.4 24.2 22.3 14.7
12.8 12.0 14.0 11.3 11.8 10.1 21.9 14.0 10.0 13.6
22 17 26 25 23 20 63 56 47 52
a VC = very coarse (2.00–1.00 mm); C = coarse (1.00–0.50 mm); M = medium (0.50–0.25 mm); F = fine (0.25–0.10 mm); VF = very fine (0.10–0.05 mm).
trend of clay accumulation in the subsurface horizons as well as on the evidence of clay coating and clay contents, the soils of CS-1, CS-2 and TL-1 are classified as
The field morphology of the soils was quantified following the procedure mentioned above (Fig. 3). The degree of differentiation between each soil horizon is revealed by the vertical variation of Horizon Indices (HI). In general, the HI value increases with soil depth with transition to a sharp increase in subsurface B horizons along the profile (Fig. 6). Such differentiation becomes progressively greater for pedons located at higher altitudes, such as CS-1 and TL-1. The WPDI values of the soils range from 0.32 to 0.56 (Tables 1 and 2). The higher the WPDI value, the older the age is indicated for the soil. All the pedons show a different degree of rubification. This is attributed to different degrees of chemical weathering associated with Fe released from silicate minerals. In addition, the soils bearing Ferromagnesianrich minerals from detrital slates or phyllites possess high Fet contents. Furthermore, the free Fe (Fed) content of soils reached up to 30 g/kg (Table 4). However, the amorphous Fe (Feo) is relatively low, such that the ratio of oxalate to DCB extractable Fe (Feo/Fed value) is always b 0.5, indicating a high degree of Fe crystallization (Blume and Schwertmann, 1969). The accumulation of Fed in subsurface horizons indicates the migration of free Fe from the eluvial to the illuvial horizons, particularly in the soils on the terraces at high
Fig. 6. The profile variation of horizontal index (HI) of the soils from (A) Chushan and Touliu terraces, and (B) Pakua terraces (from Tsai et al., 2006).
H. Tsai et al. / Geomorphology 88 (2007) 201–213
209
altitude (Fig. 5). Except for TL-4, all pedons are within Nagatsuka's (1972) red soil region [Feo/Fed ≤ 0.4, (Fed − Feo)/Fet ≥ 0.5] (Fig. 7). 4.3. Regional terrace correlation The geomorphologic correlation of river terraces is commonly based on parameters like altitude above sea level or the present river channel, gradient, longitudinal profiles etc. However, these geomorphic features are often inadequate to correlating river terraces deformed by active tectonics. The HI and WPDI indexes obtained from various morphological characteristics of the studied soils, in addition to soil color, are applied for this purpose. The river terraces are correlated in a chronological order of I, II, III, IV to n. Based on the degree of soil development, the chronosequence of the Pakua terraces are designated with roman numerals from I (oldest) to VI (youngest) accordingly (Tsai et al., 2006) (Table 1). The soils with similar WPDI values can be correlated. In fact, the correlation is best provided from the vertical variation of HI index along the soil profiles (Fig. 6). In spite of the difference in WPDI values between CS-1 and TL-1, the similar pattern in HI along the profiles strongly suggests an equivalent degree of pedogenesis. Table 4 Free Fe oxides of the studied pedons Pedon
Horizon
Free Fe oxides (g/kg)
Feo/ Fed
(Fed − Feo)/ Fet
Feo
Fed
Fet
Chushan terraces CS-1 A Bt1 Bt2 Bt3 Bt4 CS-2 A Btv1 Btv2 Btv3 Btv4 B4
3.8 4.6 4.7 4.2 3.3 1.5 1.1 1.3 0.8 1.1 0.9
26.0 28.5 33.7 36.5 33.3 22.2 26.3 27.4 31.0 25.0 30.8
41.1 42.0 41.6 42.7 47.0 33.2 37.1 38.8 37.1 40.7 36.3
0.15 0.16 0.14 0.11 0.10 0.07 0.04 0.05 0.03 0.04 0.03
0.54 0.57 0.70 0.76 0.64 0.62 0.68 0.67 0.81 0.59 0.83
Touliu terraces TL-1 A Bt1 Bt2 Bt3 Bt4 Bt5 TL-4 A AB Bw1 Bw2
3.92 5.37 4.80 4.09 3.24 3.76 2.99 3.27 3.42 4.43
15.9 14.5 14.3 16.6 15.8 18.8 8.24 10.2 10.5 9.86
16.0 16.1 17.8 15.4 24.3 26.8 19.2 11.6 25.9 30.6
0.25 0.37 0.34 0.25 0.21 0.20 0.36 0.32 0.33 0.45
0.75 0.57 0.53 0.81 0.52 0.56 0.27 0.60 0.27 0.18
Fig. 7. Plot of Fe (Feo/Fed) versus (Fed − Feo)/Fet for soils of the Chushan, Touliu and Pakua terraces. The solid line represents the regressive power curve of all data. The soils with known ages of the Kikai Island (Maejima et al., 2002) are plotted for comparison.
As a result, a sequence of CS-1 = TL-1 N CS-2 N TL-4 from high to low in the degree of soil development, is proposed as basis for regional correlation of the river terraces. The WPDI and HI values indicate that the pedogenesis of CS-1 and CS-2 are respectively equivalent to PK-4 and PK-5, which are correlated as stages IV and V. TL-4, the youngest soil, is correlated as stage VII in the sequence. Regional correlation is established in Table 1. The resultant correlation suggests that the terraces CS-1 and TL-1 correspond to the same geomorphic surface that was divided by incision of the Chinshui River when PK-4 of the Pakua river terrace was developed. The soils of the terraces based on the correlation from the youngest to oldest agree with the development sequence of Inceptsol → Paleudult → Oxisol indicated in the Soil Taxonomy (Table 1). 5. Discussion 5.1. Pedologic chronometry Soil age can be estimated by its positive relationship to Fe activity (Feo/Fed) and/or negative relationship to Fe crystallinity (Fed − Feo)/Fet (Arduino et al., 1984; MacFadden and Hendricks, 1985; Arduino et al., 1986). In the example from Japan, Maejima et al. (2002) showed a highly positive correlation between the absolute ages (Nagatsuka and Maejima, 2001) and crystallinity ratios of free iron oxides for the lateritic red soils on raised coral reef terraces of Kikai Island, and successfully estimated ages of the red-colored soils developed on Minamidaito Island. Fig. 7 shows the relationship of (Feo/Fed) versus (Fed − Feo)/Fet for the study soils mentioned above. The data of Chushan and
210
H. Tsai et al. / Geomorphology 88 (2007) 201–213
Fig. 8. Relationship between WPDIs and soil ages indicating the rate of soil development.
Touliu terraces roughly agree with those of Kikai Island and Pakua terraces. The wide scattering of data from each pedon overlaps with other pedons, which makes it difficult to estimate a correct age. Moreover, the higher (Feo/Fed) and lower (Fed − Feo)/Fet ratios of CS-1 compared with CS-2 in the observed relationship suggest CS-1 to be younger than CS-2, which does not conform to the likely sequence in terrace formation. There are studies showing the increase in the values of soil development index with time, which is suitable for chronometric application (Harden and Taylor, 1983; Rodbell, 1990; Vidic and Lobnik, 1997). In this study, the pedologic chronometry is established from three soils with known ages by linear regression. The soils of Chushan terraces give WPDI values of 0.51 (CS-2) and 0.56 (CS-1) for ages of 30 ka and 92 ka, respectively (Tables 1 and 2). The highest WPDI value of 0.73 represents the oldest soils (PK-1) of Pakua terraces. The maximum age of the Pakua terraces lies between 350– 500 ka (Liew, 1988; Delcaillau et al., 1998; Tsai et al., 2006). The mean value and the range of ages from PK-1 introduce uncertainties in age estimation of other undated soils or river terraces from regression (Fig. 8; Table 1). Although the extrapolation of the linear trend does not yield proper ages for TL-4 and PK-6, their relative ages can be obtained by pedogenic correlation based on the WPDI and HI values.
Touliu terraces (Lin, 1957; Ota et al., 2002). On the contrary, the pedogenic correlation of the river terraces indicates the Chushan terraces and the Touliu terraces are associated with same growing fold structure that has become uplifted and tilted. The smooth and flat surfaces of the Touliu terraces are uplifted above the hinge of the anticline, and the tilted surfaces of the Chushan terraces represent the east flank of the anticline (Fig. 2). They appear to be the southern extension of the Pakua Anticline folded by the slip of the fault dipping eastward, which favors the argument that the Changhua Fault extends along the west of the Touliu Hill (CPC, 1982; CGS, 2004). Although Suppe and Namson (1979) suggested the geological structure of Touliu Hill fits the fault–bend folding, Lin et al. (2000b) applied the model of faultpropagation-fold to depict the geologic evolution of this area. The geomorphic profiles show that the extension along the CS-1 surface toward the Touliu Hill reaches an altitude much higher above the TL-1 surface, indicating an anticline with a long, flat top, bordered by shorter limbs (Fig. 9A). The flat-topped geometry of anticline presented by the Touliu terraces and the Chushan terraces, in contrast to the narrow and sharp crest of the Pakua anticline, indicates a mechanism by fault–bend folding (Fig. 9B). According to Delcaillau (2001), the
5.2. Geomorphologic implications Previous studies suggested the Chushan Tableland and the Touliu Hill belong to different tectonic units based on their different patterns of terrace deformation, which forms the extension of the Changhua fault along the Chinshui River between the Chushan terraces and
Fig. 9. (A) Geomorphic profiles of the river terraces along the Choshui River. Shaded block indicates the hanging-wall of the Changhua Fault. (Geomorphic profiles of Pakua terraces from Tsai et al., 2006.) (B) Models of fold growth associated with thrust fault. Dotted lines are axial surfaces of the growing fold (Modified after Suppe, 1983, 1985).
H. Tsai et al. / Geomorphology 88 (2007) 201–213
Pakua terraces grew as a lateral propagation (westward and southward) of an active anticline, which had greater deformation in the south where a fault-propagation-fold may have formed. However, seismic and borehole evidence proved another fault–bend–fold system under the northern Pakua Tableland (Hsiao, 1968; Chen, 1978; Wang et al., 2003). The transitions between folding mechanisms along the Changhua Fault indicate the fault growth from segment linkage, which agrees with the geomorphologic observations (Yang, 1986; Delcaillau, 2001). The present water gap between Pakua Tableland and Touliu Hill probably marks the boundary between two fault segments beneath the surface (Fig 2). 5.3. Rate of soil development The red soils of the Chushan, Touliu and Pakua terraces form a sequence of Inceptsol, Ultisol and Oxisol following the Soil Taxonomy (Soil Survey Staff, 2003). Their pedogenic development increases with their WPDI values as well as their ages (Table 1). These relationships suggest that the soil WPDIs may assist pedologists not only in soil classification but also in assessing the age of the soils. By relating the WPDI values to soil ages, an approximation of the development rate of Pleistocene soils in Taiwan is revealed (Fig. 8). However, the validity of the linear regression is limited by the number and quality of soil ages. The age (or time) curve of soil development is usually approached by power, exponential or logarithmic chronofunctions (Bockheim, 1980; Birkeland, 1984, 1999). They usually flatten markedly by 10 ka (Rodbell, 1990; Alonso et al., 1994). The linear trend in Fig. 8 represents the rates of soil development of Pleistocene terraces in Taiwan. This rate in a tropical climate is about 2 and 3 times faster than those estimated in Ljubljana Basin, Slovenia (temperate climate) and Salamanca, Spain (Mediterranean type) respectively. Note that the soils of Slovenia (Vidic and Lobnik, 1997) and Spain (Alonso et al., 1994) are compared by recalculating WPDI values with identical soil properties to those used in this study. The increasing trend of the rates toward warmer climate and higher precipitation agrees with the common opinion of climatic effect on soil development (Vidic, 1998). However, their difference is much lower than that reported by Harden (1990), who indicates that the rates in Pleistocene times are more than 10 times slower in semi-arid than in the moister regions. The comparison in this study minimized the bias introduced by soil properties, profile thickness and other factors, thus providing an objective view on soil development.
211
6. Conclusions The soil pedons studied from the Chushan and Touliu terraces in central Taiwan are identified as Inceptsol (TL4) and Ultisol (CS-1, CS-2, TL-1). The WPDI values range from 0.31 to 0.58 for these soils. Based on the degree of soil development, the river terraces (Papua, Chushan and Touliu terraces) of different fluvial systems are chronologically arranged and correlated as I (PK-1), II (PK-2), III (PK-3), IV (PK-4, CS-1 & TL-1), V (PK-5 & CS-2), VI (PK-6) and VII (TL-4). Based on this correlation, the terraces CS-1 and TL-1, separated by the Chinshui River, are considered to correspond to the same surface. Surface deformation reflects the growth of the structure underground in which uplift folding occurred in the hinge zone and tilted to the east in the fold limb, representing the southern extension of the Pakua anticline. However, the different geometries on both the surface and the subsurface from southern Pakua Anticline indicate a mechanism of fault–bend folding for Touliu Hill. The relationship of WPDI values with soil age provides an approximation of the development rate of Pleistocene soils in Taiwan. According to the comparison with representative soils from Mediterranean and Temperate environments, a slightly faster soil development in the study area is revealed. However, the minor differences suggest that the climatic effect on soil pedogenesis becomes less important on soils developed from Pleistocene times. Acknowledgements The authors are grateful to Professors Zueng-Sang Chen and two anonymous reviewers for their critical reading and suggestion. The gratitude also extends to Mr. Hon-Yi Huang, Yan-Gu Chen and Miss. Ya-Wen Hsiao, Fu-Lang Kuo, Hui-Mei Lin for their help in fieldworks and laboratory analyses. The study was financially supported by the project (NSC 94-2116-M018 -001) from the National Science Council, Taiwan. References Alonso, P., Sierra, C., Ortega, E., Dorronsoro, C., 1994. Soil development indices of soils developed on fluvial terraces (Peòaranda de Bracamonte, Sala manca, Spain). Catena 23, 295–308. Angelier, J., 1986. Geodynamics of Eurasia—Philippine Sea Plate boundary. Preface of Tectonophysics 125, IX–X. Arduino, E., Barberis, E., Carraro, F., Forno, M.G., 1984. Estimating relative ages from iron-oxide/total-iron ratios of soils in the western Po Valley, Italy. Geoderma 33, 39–52. Arduino, E., Barberis, E., Marson Ajmone, F., Zanini, E., Franchini, M., 1986. Iron oxides and clay minerals within profiles as indicators of soil age in northern Italy. Geoderma 37, 45–55.
212
H. Tsai et al. / Geomorphology 88 (2007) 201–213
Biq, C., 1992. Another coastal range on Taiwan. Ti-Chih 12, 1–14. Birkeland, P.W., 1984. Holocene soil chronofunctions, southern Alps, New Zealand. Geoderma 34, 115–134. Birkeland, P.W., 1999. Soils and Geomorphology, 3rd ed. Oxford Univ. Press, New York. Blume, H.P., Schwertmann, U., 1969. Genetic evaluation of profile distribution of aluminum, iron, and manganese oxides. Soil Science Society of America Proceedings 33, 438–444. Bockheim, J.G., 1980. Solution and use of chronofunction in studying soil development. Geoderma 24, 71–85. CGS, 2004. Geologic map of Nantou and Yunlin. Central Geological Survey of Taiwan, scale 1:50,000, 2 sheets. Chen, J.S., 1978. A comparative study of the refraction and reflection seismic data obtained on the Changhua Plain to the Peikang Shelf, Taiwan. Petroleum Geology of Taiwan 15, 199–217. Chen, F.Y., Liew, P.M., 1990. Palynological study of the sungshan formation, Taipei basin. Proceedings Geological Society of China 33, 21–37. Chen, Y.G., Liu, T.K., 1991. Radiocarbon dates of river terraces along the lower Tahanchi, northern Taiwan: their tectonic and geomorphic implications. Journal of the Geological Society of China 34, 337–347. CPC, 1982. Geologic map of Taiwan: Chinese Petroleum Corporation, Taiwan, scale 1:100,000, 4 sheets. Deffontaines, B., Lacombe, O., Angelier, J., Chu, H.T., Mouthereau, F., Lee, C.T., Deramond, J., Lee, J.F., Yu, M.S., Liew, P.M., 1997. Quaternary transfer faulting in the Taiwan Foothills: evidence from a multisource approach. Tectonophysics 274, 61–82. Delcaillau, B., 2001. Geomorphic response to growing fault-related folds: example from the foothills of central Taiwan. Geodinamica Acta 14, 265–287. Delcaillau, B., Deffontaines, B., Floissac, L., Angelier, J., Deramond, J., Souquet, P., Chu, H.T., Lee, J.F., 1998. Morphotectonic evidence from lateral propagation of an active frontal fold; Pakuashan anticline, foothills of Taiwan. Geomorphology 24, 263–290. Engel, S.A., Gardner, T.W., Ciolkosz, E.J., 1996. Quaternary soil chronosequences on terraces of the Susquehnna river, Pennsylvania. Geomorphology 17, 273–294. Gee, G.W., Bauder, J.W., 1986. Particle-size analysis. In: Klut, A. (Ed.), Methods of Soil Analysis, Part 1. Physical and Mineralogical Methods. Agronomy Monograph, vol. 9. ASA and SSSA, Madison, WI, pp. 383–411. Harden, J.W., 1982. A quantitative index of soil development from field descriptions, examples from a chronosequence in Central California. Geoderma 28, 1–28. Harden, J.W., 1990. Soil development on stable landforms and implication for landscape studies. Geomorphology 3, 391–398. Harden, J.W., Taylor, E.M., 1983. A quantitative comparison of soil development in four climatic regimes. Quaternary Research 20, 342–359. Ho, C.S., 1988. An Introduction to the Geology of Taiwan: Explanatory Text of the Geologic Map of Taiwan, 2nd ed. Central Geologic Survey, Taiwan, Republic of China. Hsiao, P.T., 1968. Seismic study of the Taichung area, Taiwan. Petroleum Geology of Taiwan 6, 209–216. ISSS Working Group WRB, 1998. World reference bases for soil resource: introduction. In: Deckers, J.A., Nachtergaele, F.O., Spaargaren, O.C. (Eds.), International Society of Soil Science, International Soil Reference and Information Centre (ISRIC), and Food and Agriculture Organization (FAO). Acco, Leuven. 165 pp. Jenny, H., 1941. Factors in Soil Formation. McGraw-Hill, New York. 281 pp.
Kao, H., Chen, W.P., 2000. The Chi-chi earthquake sequence: active outof-sequence thrust faulting in Taiwan. Science 288, 2346–2349. Ku, C.C., 1963. Photogeologic study of terraces in northwestern Taiwan. Proceedings Geological Society of China 6, 51–60. Leamy, M.L., Milne, J.D.G., Pullar, W.A., Bruce, J.G., 1973. Paleopedology and stratigraphy in the New Zealand Quaternary succession. New Zealand Journal of Geology and Geophysics 16, 723–744. Lee, J.C., Lu, C.Y., Chu, H.T., Delcaillau, B., Angelier, J., Deffontaines, B., 1996. Active deformation and paleostressanalysis in the Pakua anticline area, western Taiwan. Terrestrial, Atmospheric and Oceanic Sciences 7, 431–446. Lee, T.Q., Lin, S.F., Chou, H.C., 1999. Preliminary magnetic study of the Quaternary red-soil bed on Linkou terrace, northern Taiwan. Terrestrial, Atmospheric and Oceanic Sciences 10, 763–776. Liew, P.M., 1988. Quaternary stratigraphy in western Taiwan: palynological correlation. Proceedings Geological Society of China 31, 169–180. Liew, P.M., Hsieh, M.L., 2000. Late Holocene (2 ka) sea level, river discharge and climate interrelationship in the Taiwan region. Journal of Asian Earth Sciences 18, 499–505. Liew, P.M., Huang, S.Y., 1994. Pollen analysis and their paleoclimatic implication in the middle Pleistocene lake deposits of the Ilan district, northeastern Taiwan. Journal of the Geological Society of China 37, 115–124. Lin, C.C., 1957. Topography of Taiwan: Publ. of the Taiwan Provincial Documentary Committee. 424 pp. [in Chinese]. Lin, C.W., Huang, H.C., Lu, S.T., Shih, T.S., Huang, W.J., 2000a. Active fault map of Taiwan, 2nd ed. Central Geological Survey of Taiwan, scale 1:500,000, 1 sheet. Lin, C.W., Lu, S.T., Huang, W.J., Shih, T.S., Chang, H.C., 2000b. The Chi-Chi earthquake fault and structural analyses of the area south of Choshuihsi, central Taiwan. Central Geological Survey of Taiwan, Special Publication, vol. 12, pp. 89–111 (Special issue for the Chi-Chi earthquake, 1999.). Liu G.Y. 2004. A Study of subsurface structures in the Touliu terrace of Tongsuhu fault. MS thesis of Nation Central University, Taiwan, 94 pp. [in Chinese]. Lu, C.Y., Jeng, F.S., Chang, K.J., Jian, W.T., 1998. Impact of basement high on the structure and kinematics of the western Taiwan thrust wedge: insights from sandbox models. Terrestrial, Atmospheric and Oceanic Sciences 9, 533–550. Ma, K.F., Lee, C.T., Tsai, Y.B., Shin, T.C., Mori, J., 1999. The Chi-chi, Taiwan earthquake: large surface displacements on an island thrust fault. EOS, Transactions 80, 605–611. MacFadden, L.D., Hendricks, D.M., 1985. Changes in the content and composition of pedogenic iron oxyhydroxides in a chronosequence of soils in southern California. Quaternary Research 23, 189–204. Maejima, Y., Nagatsuka, S., Higashi, T., 2002. Application of crystallinity ratio of free iron for dating soils developed on the raised coral reef terraces of Kikai and Minami-Daito islands, southwest Japan. The Quaternary Research (Daiyonki-Kenkyu) 41, 485–493. McKeague, J.A., Day, J.H., 1966. Dithionite and oxalate extractable Fe and Al as acids in different various classes of soils. Canadian Journal of Soil Science 46, 13–22. Mehra, O.P., Jackson, M.L., 1960. Iron oxides removed from soils and clays by a dithionite–citrate system buffered with sodium bicarbonate. Clays and Clay Minerals 7, 317–327. Morrison, R.B., 1968. Means of time-stratigraphic division and longdistance correlation of Quaternary successions. In: Morrison, R.B.,
H. Tsai et al. / Geomorphology 88 (2007) 201–213 Wright Jr., H.E. (Eds.), Means of Correlation of Quaternary Successions. Int. Assoc. Quat. Res., VII Congress., Proc., vol. 8, pp. 1–113. Mouthereau, F., Lacombe, O., Deffontaines, B., Angelier, J., Chu, H.T., Lee, C.T., 1999. Quaternary transfer faulting and belt front deformation at Pakuashan (Western Taiwan). Tectonics 18, 215–230. Mulcahy, M.J., Churchward, H.M., 1973. Quaternary environments and soils in Australia. Soil Science 116, 156–169. Nagatsuka, S., 1972. Studies on genesis and classification of soils in warm–temperate region of Southwest Japan, part 3. Some features in distribution and mode of existence of free iron and aluminum oxides in the soil profile. Soil Science and Plant Nutrition 18, 147–154. Nagatsuka, S., Maejima, Y., 2001. Dating of f soils on the raised coral reef terraces of Kikai Islands in Ryukyus, southwest Japan. The Quaternary Research (Daiyonki-Kenkyu) 40, 137–147. Ota, Y., Shyu, B.H., Chen, Y.G., Hsieh, M.L., 2002. Deformation and age of fluvial terraces south of the Choushui River, central Taiwan, and their tectonic implications. Western Pacific Earth Sciences 2, 251–260. Reisenauer, H.M., 1982. Chromium. In: Page, A.L., Miller, R.H., Keeney, D.R. (Eds.), Methods of Soil Analysis, Part 2. Chemical and Microbiological Properties. Agronomy Monograph, vol. 9. ASA and SSSA, Madison, WI, pp. 337–346. Rodbell, D.T., 1990. Soil–age relationships on late Quaternary moraines, Arrowsmith Range, Southern Alps, New Zealand. Arctic and Alpine Research 22, 355–365. Seno, T., Stein, S., Gripp, A.E., 1993. A mode for the motion of the Philippine Sea plate consistent with NUVEL-1 and geological data. Journal of Geophysical Research 98, 17941–17948. Soil Survey Staff, 1993. Examination and description of soils in the field. Soil Survey Manual, Handbook, vol. 18. USDA-SCS, Washington, D.C., pp. 56–196. Soil Survey Staff, 2003. Keys to Soil Taxonomy, 9th eds. USDANRCS, Washington, D.C. 332 pp. Sung, Q.C., Chen, Y.C., 2004. Geomorphic evidence and kinematic model for Quaternary transfer faulting of the Pakuashan anticline, Central Taiwan. Journal of Asian Earth Sciences 24, 389–404. Sung, Q.C., Lu, M.T., Tsai, H., Liu, P.M., 1997. Discussion on the genetics and the correlation of river terraces in Taiwan. J. Geol. Soc. China 40, 31–46.
213
Suppe, J., 1981. Mechanics of mountain building in Taiwan. Memoir of the Geological Society of China 4, 67–89. Suppe, J., 1983. Geometry and hinematics of fault–bend folding. American Journal of Science 283, 648–721. Suppe, J., 1985. Principles of Structural Geology. Prentice Hall, New Jersey. Suppe, J., Namson, J., 1979. Fault–bend origin of frontal folds of western Taiwan fold-and-thrust belt. Proc. Geol. Soc. China, 16, 11Chang, S. L., 1971. Subsurface Geologic Study of the Taichung Basin, Taiwan. Petrol. Geol. Taiwan, vol. 8, pp. 21–45. Torrent, J., Schwertmann, U., Schulze, D.J., 1980. Iron oxide mineralogy of some soils of two river terrace sequences in Spain. Geoderma 23, 191–208. Torrent, J., Schwertmann, U., Fechter, H., Alferez, F., 1983. Quantitative relationship between soil color and hematite content. Soil Science 136, 354–358. Tsai, Y.B., 1986. Seismotectonics of Taiwan. Tectonophysics 125, 17–38. Tsai, H., Huang, W.S., Hseu, Z.Y., Chen, Z.S., 2006. A river terrace soil chronosequence of the Pakua tableland in Taiwan. Soil Science 171, 167–179. Tseng, M.H., Liew, P.M., Chi, W.R., Shih, T.S., 1992. Pollen analysis of the Tananwan formation, northern Taiwan. Journal of the Geological Society of China 35, 247–259. Vidic, N.J., 1998. Soil–age relationships and correlations: comparison of chronosequences in the Ljubljana Basin, Slovenia and USA. Catena 34, 113–129. Vidic, N.J., Lobnik, F., 1997. Rates of soils development of the chronosequence in the Ljubljana Basin, Slovenia. Geoderma 76, 35–64. Wang, C.Y., Kuo, S.Y, Shyu, W.L., Hsiao, J.W., 2003. Investigating near-surface structures under the Changhua fault, west-central Taiwan by the reflection seismic method. Terrestrial, Atmospheric and Oceanic Sciences 14, 343–367. Yang, K.S., 1986. The geomorphological study on active faults in Taiwan—discussion on relationship between active fault and geomorphic surface. Ph.D. thesis of Chinese Culture University, Taiwan, 178 pp. [in Chinese]. Yu, S.B., Chen, H.Y., Kuo, L.C., 1997. Velocity field of GPS stations in the Taiwan area. Tectonophysics 274, 41–59.
Pedogenic correlation of lateritic river terraces in central Taiwan Heng Tsai a,⁎, Wen-Shu Huang a , Zeng-Yei Hseu b b
a Department of Geography, National Changhua University of Education, Changhua 50007, Taiwan Department of Environmental Science and Engineering, National Pingtung University of Science and Technology, Pingtung 91201, Taiwan
Received 12 June 2006; received in revised form 6 November 2006; accepted 6 November 2006 Available online 13 December 2006
Abstract The Pakua, Chushan and Touliu terraces are three groups of lateritic river terraces resulting from different fluvial systems in central Taiwan. Their geomorphic features do not enable regional correlations to be made. In this study, an alternative scheme based on the quantification of a pedogenic development is proposed for this correlation. Four soil pedons from the Chushan and Touliu terraces are classified as Ultisols (CS-1, CS-2 and TL-1) and an Inceptsol (TL-4) following the Soil Taxonomy. An additional six soil pedons, including an Oxisol (PK-1), Ultisols (PK-2, -3, -4 and -5), and an Inceptsol (PK-6), from the Pakua terraces were included from a previous study [Tsai, H., Huang, W.S., Hseu. Z.Y., Chen, Z.S., 2006. A river terrace soil chronosequence of the Pakua tableland in Taiwan. Soil Science 171, 167–179.]. The weighted mean profile development indices (WPDI) for the pedons suggest a chronological order of I (PK-1), II (PK-2), III (PK-3), IV (PK-4, CS-1 and TL-1), V (PK-5 and CS-2), VI (PK-6) and VII (TL-4). The correlation suggests the Chushan and Touliu terraces correspond to the same surface of an anticline, separated by the Chinshui River. Surface deformation reflects the growth of the underlying structure in which uplift folding occurred in the hinge zone and tilted to the east along the fold limb, as a result of the southern extension of the Pakua anticline. However, the different geometries on both the surface and subsurface from southern Pakua Anticline indicate a mechanism of fault-bend folding for Touliu Hill. © 2006 Elsevier B.V. All rights reserved. Keywords: River terrace; Profile development index; Pedogenesis; Soil classification; Anticline; Quaternary
1. Introduction River terraces are one of the most prominent geomorphic features in Taiwan, and constitute an excellent marker for the study of active tectonics. However, the absence of absolute datings and geochronological information means that ages are typically a problem in extracting information from deformed surfaces. Previous studies have correlated terraces mainly on the basis of ⁎ Corresponding author. No.1, Chinte Road, Changhua 50007, Taiwan. Tel.: +886 4 7232105 2819; fax: +886 4 7211186. E-mail address: [email protected] (H. Tsai). 0169-555X/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.geomorph.2006.11.004
geomorphic and pedogenic parameters such as altitude above sea level or present river channel, continuity and height of terrace scarp, surface gradient and color of topsoil (Lin, 1957; Ku, 1963; Yang, 1986). However, their results are usually inconsistent and are not satisfactory. Jenny (1941) indicated that most soil properties are time-dependent chronofunctions, and some of them have been used to correlate Quaternary successions (Morrison, 1968; Leamy et al., 1973; Mulcahy and Churchward, 1973; Birkeland, 1984; Engel et al., 1996). The activity (Feo/Fed) and crystallinity [(Fed − Feo)/Fet] ratios of free iron are good time-dependent indexes for
202
H. Tsai et al. / Geomorphology 88 (2007) 201–213
red soils (Blume and Schwertmann, 1969; Nagatsuka, 1972), and have been successfully used in northern Italy (Arduino et al., 1984, 1986), southern California (MacFadden and Hendricks, 1985) and the Kikai Island of southwestern Japan (Maejima et al., 2002). Harden's (1982) profile development index (PDI) has been proven as another successful method in quantifying field attributes of soil in order to rank soil development (Alonso et al., 1994). Moreover, Harden and Taylor (1983) have demonstrated that the index may be suitable for correlating soil chronosequences on different parent materials and in different climates. Although Vidic (1998) questioned the validity of the chronofunction, it was agreed that the parent material composition does not affect the soil development index significantly. There are many lateritic river terraces in central Taiwan. The Pakua, Chushan and Touliu terraces represent three different groups of river terraces developed in various drainage basins. No satisfactory regional correlation of terraces can be achieved based on the geomorphic parameters mentioned above. Tsai et al. (2006) reported a soil chronosequence of the Pakua terraces, and estimated the ages by the ratio of (Fed − Feo)/Fet. In this study, the weighted mean of the PDIs (hereafter termed WPDI) and the pedogenic iron are calculated and compared as a basis for regional terrace correlation. The rate
of soil development based on the WPDIs is proposed as an effective chronological indicator. 2. Study area 2.1. Geologic background Taiwan is geologically located at the convergent boundary between the Philippine Sea plate and the continental margin of the Eurasian plate (Fig. 1A). The Philippine Sea plate subducts beneath the Eurasian plate offshore eastern Taiwan, and overrides the South China Sea floor of the Eurasian plate south of Taiwan (Suppe, 1981; Tsai, 1986; Angelier, 1986; Ho, 1988). Active tectonics in Taiwan are manifested by continuing shortening and widespread seismic activity. A shortening rate of 8.5 cm/yr is revealed by GPS measurements across the 200-km-wide island (Seno et al., 1993; Yu et al., 1997). There are a series of active subparallel west-verging thrusts in the fold-and-thrust belt along the western Foothills of Taiwan, resulting from a thrust migrating westward by tectonic compression since the Plio–Pleistocene (Lee et al., 1996) (Fig. 1B). The largest inland earthquake (Mw = 7.6, ML = 7.3) of the 20th century in Taiwan was caused by the displacement of the Chelungpu Fault (Ma et al., 1999; Kao and Chen, 2000).
Fig. 1. (A) The tectonic setting of Taiwan. (B) Geological map and cross-section of the studied area in central Taiwan.
H. Tsai et al. / Geomorphology 88 (2007) 201–213
The Changhua Fault, as the active frontal thrust of the fold-and-thrust belt, results in an asymmetric anticline (also known as Pakua Tableland) with a gentler east limb (less than 10°) and a steeper west limb (up to 30°) (Lu et al., 1998; Wang et al., 2003) (Figs. 1B and 2). There is a basement high, the Peikang High, located west of the fold-and-thrust belt (Biq, 1992; Lu et al., 1998). Its obstacle effect results in the deformation of the tableland in a sigmoidal shape trending NNW–SSE in the north, and N–S in the south (Lee et al., 1996; Mouthereau et al., 1999; Sung and Chen, 2004). Leftlateral strike slip faulting associated with transtensional deformation were identified from the microstructures in the northern part of the anticline (Lee et al., 1996; Delcaillau et al., 1998; Mouthereau et al., 1999). No clear outcrops of the Changhua Fault have been found in the field (Lin et al., 2000a). Yang (1986), based on aerial photo interpretation, suggested three segments of the fault along the west of the Pakua Tableland.
203
Seismic and borehole evidence indicates the Changhua Fault as a “blind-thrust” buried underground (Hsiao, 1968; Chen, 1978; Wang et al., 2003). The balanced cross-section in Deffontaines et al. (1997) indicates that the Pakua Anticline has accommodated about 2–3 km of shortening at the southern end and 4–5 km at the northern end since 0.4–0.5 Ma. According to the available geologic maps, the Changhua Fault extends along the west of the Pakua Tableland and the Touliu Hill (CPC, 1982; CGS, 2004). Touliu Hill is an anticline that plunges to the north, where the topography does not comply with the geometry of the underground structure. The relatively symmetrical form of surface is underlain by the strata of west-limb (50–80°) and east-limb (10– 40°) (CGS, 2004; Liu, 2004). Shallow seismic reflection profiles show highly bending layers without any trace of fault offsets along the Touliu Hill (Liu, 2004). Ota et al. (2002) argued that the south segment of the fault extends along the east side of the Chinshui river valley, a
Fig. 2. (A) The geographic distribution of the Pakua, Chushan and Touliu terraces. (B) Enlarged map of the Touliu terraces. (C) The 3D relief of the study area (in 40 × 40 m resolution) and two geomorphic profiles. The geologic structure along the BB′ profile was interpreted based on geological map of the CGS (2004).
204
H. Tsai et al. / Geomorphology 88 (2007) 201–213
Table 1 Geomorphic and pedologic characteristics of the Chushan, Touliu and Pakua terraces Terrace/ pedon Chushan CS-1 CS-2 Touliu TL-1 TL-4 Pakua5 PK-1 PK-2 PK-3 PK-4 PK-5 PK-6
Slope (%)
Altitude (m)1 a.s.l.
a.c.r.
2.1 0.7
250–160 190–150
130–50 70–30
0.6 1.7
330–280 240–230
3.8 3.3 2.8 2.1 1.4 1.4
440–375 430–330 410–250 385–320 330–240 325–170
Parent material2
Classification3
Pedologic correlation
Age4 (ka)
WRB
Taxonomy
PM1 PM1
Acrisol Acrisol
Paleudult Paleudult
IV V
92 31
190–180 133–100
PM2 PM2
Acrisol Cambisol
Paleudult Dystrudept
IV VII
∼ 92a, 78 ± 34b b30a
345–275 320–180 310–140 295–225 230–130 145–45
PM1 PM1 PM1 PM1 PM1 PM1
Ferralsol Acrisol Acrisol Acrisol Acrisol Cambisol
Hapludox Paleudult Paleudult Hapludult Paleudult Dystrudept
I II III IV V VI
500–350 335 ± 77b 267 ± 65b 164 ± 48b, ~92a, ∼ 31a b30a
1: “a.s.l.” represent the altitude above sea level, “a.c.r.” represent the altitude above current river. 2: Sources of parent material from Choshui River (PM1) and Chinshui River (PM2). 3: WRB from ISSS Working Group WRB (1998), Taxonomy from Soil Survey Staff (2003). 4: Ages of CS-1, CS-2 from Ota et al. (2002), and the age of PK-1 in the range from Liew (1988), Mouthereau et al. (1999) and Tsai et al. (2006). The age with superscript “a” based on pedologic correlation, and superscript “b” interpolated from Fig. 8. 5: Soil data from Tsai et al. (2006).
northward flowing tributary of the Choshui River, indicating two different tectonic units between Pakua Tableland and the Touliu Hill (Fig. 2). These fold structures are mainly composed of the Toukoshan Formation, which constitutes synorogenic deposits (Ho, 1988; Lee et al., 1996). The east-dipping strata of Toukoshan Formation resulting in a monocline structure inclined at less than 10° beneath the river terraces located east of the Chinshui River (Fig. 2). 2.2. River terraces and soils The Choshui River, the largest river in Taiwan, entrenched the fold-and-thrust belt in central Taiwan as an antecedent stream, and formed groups of a series of wide unpaired river terraces such as the Pakua and Chushan terraces (Lin, 1957) (Fig. 2). The Touliu terraces were developed by the Chinshui River, a south tributary of the Choshui River. The Pakua terraces are located on the southern part of the Pakua Tableland. The Touliu and Chushan terraces are separated from the Pakua terraces by the Choshui River, and are respectively located on the east and west sides of the Chinshui River (Fig. 2). The very dark sediments (Munsell 5GY 3/1) of the Choshui River are mainly eroded from the slate, phyllite and quartzite of the Tertiary metamorphic terrain in the Central Range. The sediments of the Chinshui River are considerably lighter in color (5Y 4/ 2) and are derived from a non-metamorphic rocks of sandstone and shale bedrock.
The Pakua terraces can be divided into six levels, from PK-1 (highest) to PK-6 (lowest), in the order of descending altitude, indicating a long-term northward river migration (Yang, 1986) (Table 1; Fig. 2). However, the lowering of the river by erosion is insufficient to keep pace with the rapid uplift of the anticline, preventing the development of river terraces located lower than PK-6 (Yang, 1986; Tsai et al., 2006). The defeated river diverted around the nose of the anticline at the southern end of the tableland. Delcaillau et al. (1998) indicate that the uplift of the Pakua Anticline began about 400–500 ka, which gives a maximum age estimation for the oldest river terraces developed on the surface. Liew (1988) proposed, based on palynological analyses, that the terrace sediments were probably deposited during some high sea-level intervals between 70–350 ka. The Chushan terraces consist of four levels of river terraces that have become progressively eastward-tilted. The higher the terraces, the steeper of slope on the surface. Radiocarbon dating gives an age of about 31 ka for CS-2, and indicates an age of around 92 ka for CS-1 by assuming a constant rate of tilting (Ota et al., 2002). The later age is reasonably constrained because the tilting gives a maximum uplift rate of 2.7 mm/yr (250 m/ 92,000 yr) within the range estimated by Delcaillau et al. (1998) for the Pakua terraces. The Touliu terraces also consist of four levels of river terraces distributed along the hinge of the anticlinal hill. However, they are fragmented and small in size due to erosion. Most of the river terraces mentioned above are covered by red soils and gravels, and hence are designated as
H. Tsai et al. / Geomorphology 88 (2007) 201–213
205
“lateritic river terraces” by Lin (1957). Unfortunately, material for 14C or biostratigraphic dating is poorly preserved under highly oxidized and weathering conditions. It is generally believed that the red-colored soils in Taiwan could be developed prior to 30 ka (Liew, 1988; Chen and Liu, 1991; Sung et al., 1997; Ota et al., 2002). The oldest age for red soil was reported as ≤ 900 ka based on paleomagnetic correlation (Lee et al., 1999). Tsai et al. (2006) compared the (Fed − Feo)/Fet ratios of the Kikai Island soils, and suggest ages of 40–400 ka for the soils of the Pakua terraces. Liew (1988) indicated a lateritic weathering phase occurred in latest Pleistocene time about 140 ka ago, after the major transgression associated with the Last Interglacial event. 2.3. Climate and vegetation Taiwan has a tropical climate in the southern and western plains and subtropical climate in the northern and mountainous regions. The study area is located in the tropical zone, and the climate is characterized by high temperature and humidity, substantial rainfall and summer tropical cyclones. According to the Central Weather Bureau of Taiwan (http://www.cwb.gov.tw), the mean annual temperature and precipitation for the area are 23 °C and 1642 mm/yr respectively. The seasonal oscillations in precipitation, with a dry winter and a wet summer, comprise a wet–dry tropical climate (Cwa) based on Köppen's climate classification. Subtropical broad-leaved evergreen forests are the predominant vegetation in this area. The paleoclimate is characterized by a warm–cold– warm cycle during glacial–interglacial periods of the Pleistocene. The cold and dry phase that existed during the mid-Pleistocene is considered to be younger than the Jaramillo event (about 0.7–0.9 Ma) (Tseng et al., 1992; Liew and Huang, 1994). The warm–temperate to subtropical climatic conditions continued and were separated by three very humid intervals during the early and mid-Holocene (Chen and Liew, 1990; Liew and Hsieh, 2000). Palynological evidence indicates that a warmer annual temperature of 1–2 °C with a relatively higher amount of precipitation prevailed during the warm interval than in the cool–dry interval in mid-Pleistocene times (Liew and Hsieh, 2000). 3. Materials and methods A total of 4 pedons were sampled from the Chushan and Touliu terraces, and were labelled as their terraces for easy illustration (Fig. 2). The classification and sampling of each horizon of the soils was conducted following Soil
Fig. 3. Flow chart for deriving the PDI or WPDI (redrawn and modified from Harden and Taylor, 1983).
Survey Staff (1993). Soil samples ≤2 mm in diameter are prepared for physical and chemical analyses. The soil properties of another 6 pedons of the Pakua terraces are reported previously in Tsai et al. (2006). Based on the guidelines of Harden (1982) and Harden and Taylor (1983) with minor modification (Fig. 3), the PDIs of the pedons were calculated based on the following six properties: (i) rubification (moist color hue and chroma from the Munsell color chart); (ii) total texture (texture class plus type of stickiness and plasticity of the wet consistence); (iii) structure (type and degree of development of soil structure); (iv) moist consistence (class); (v) dry consistence (class); and (vi) clay coatings (abundance, thickness, and location). The pH and color values (lightening or darkening) of soil used in Harden's (1982) original method were not included in our calculation. The pH was omitted because of its susceptibility to the disturbance by human activities, i.e. ploughing and fertilization in the study area. The color value of the soils has been considered an important soil property mainly in the early stages of pedogenesis because organic matter content seems to increase relatively rapid and reach a steady state faster than any other soil properties (Birkeland, 1999). The organic matter in soils will then decrease to a very low content after long-term development, leading to an insignificant variation in color value. As a result, the color value is also ignored in the calculation. The index values are calculated by comparing the soil properties of a horizon with those of the assumed parent
206
H. Tsai et al. / Geomorphology 88 (2007) 201–213
Touliu terraces (Table 1). The PDI values were further divided by the depth of the described soil profile to produce WPDI normalized values ranging between 0 (no development) and 1 (maximum development) (Birkeland, 1999). In addition to soil morphology quantified in the field, particle size distribution was determined in the laboratory using the pipette method (Gee and Bauder, 1986). Total Fe (Fet) was determined by the Reisenauer (1982) method. Free Fe (Fed) was extracted by the dithionite– citrate–bicarbonate (DCB) method (Mehra and Jackson, 1960). Amorphous Fe (Feo) was extracted by 0.2 M ammonium oxalate (pH 3.0) (McKeague and Day, 1966). All the Fe in solution was determined by atomic absorption spectrometry (Hitachi Z-8100 type, Japan). Fig. 4. Particle size distribution of the pedons and the soil texture classes (soil data of the Pakua terraces from Tsai et al., 2006).
material. The sediments of the present active floodplains were used to represent fresh and unoxidized parent material of the soils. The soils containing fragments of detrital slate or phyllites observed in the Chushan terraces indicate that the parent material was deposited by the Choshui River. In contrast, the alluvium from the Chinshui River represents the parent material of the
4. Results 4.1. Soil characteristics and classification The soils graded from yellow to reddish in hue, showing a trend that roughly agrees with the ascending altitude of terraces (the higher the terraces, the richer the red color of the soils). Consequently, the highest redness index (RI) is found in pedon TL-1, while the least
Fig. 5. The variations of soil profile distribution in clay contents (shaded region) and Fed (solid circle).
H. Tsai et al. / Geomorphology 88 (2007) 201–213
207
Table 2 The morphological characteristics and the quantification of the studied pedons Pedon Horizon Depth (cm) CS-1
CS-2
TL-1
TL-4
PM1 PM2
A Bt1 Bt2 Bt3 Bt4 A Btv1 Btv2
0–20 20–45 45–85 85–120 120–150 0–10 10–25 25–70
Texture a Structure b Consistence c
Munsell color Moist
10YR 4/4 10YR 4/6 7.5YR 6/8 7.5YR 6/8 7.5YR 7/8 10YR 6/6 10YR 5/6 10YR 5/5 (95%) 2.5YR 6/4 (5%) Btv3 70–95 10YR 5/6 (90%) 2.5YR 6/4 (10%) Btv4 95–125 10YR 5/6 (85%) 2.5YR 6/4 (15%) A 0–35 7.5YR 5/6 Bt1 35–65 7.5YR 4/6 Bt2 65–100 7.5YR 4/4 Bt3 100–125 5YR 4/6 Bt4 125–155 5YR 4/4 Bt5 155–180 5YR 4/6 A 0–10 10YR 3/4 AB 10–20 10YR 3/4 Bw1 20–35 7.5YR 3/4 Bw2 35–55 7.5YR 3/4 (Choshui river) 5GY 3/1 (Chinshui river) 5Y 4/2
Dry
Dry Moist Wet
Clay RI e coatings d
HI f
0.39 0.56 (PM1) 0.47 0.61 0.61 0.61 0.44 0.49 (PM1) 0.44
10YR 5/4 10YR 6/6 7.5YR 7/6 7.5YR 7/6 7.5YR 7/6 10YR 6/6 10YR 7/6 10YR 6/6
CL SiCL SiC C C CL CL CL
2f&msbk 2f&mabk 2m&cabk 2m&cabk 2m&abk 2f&msbk 2f&mabk 2f&mabk
sh h h h h sh sh h
fr fir fir fir fir fir fir fir
ssps ssps sp vsp vsp ssps ssps ssps
–h – 1dpf 1dpf 1dpf – – 1fpf
0 0 3.33 3.33 2.86 0 0 0
10YR 6/6
SiC
2m&cabk
sh
fr
sp
1fpf
0
7.5YR 6/6 SiC
2m&cabk
sh
fr
sp
1fpf
7.5YR 5/4 7.5YR 6/4 7.5YR 6/6 5YR 5/6 5YR 5/6 5YR 5/6 10YR 4/4 10YR 4/4 7.5YR 4/6 7.5YR 4/4 5GY 4/1 5Y 5/2
1f&msbk 2f&mabk 3f&mabk 3f&mabk 3m&cabk 3f&mabk 1fgr 1f&msbk 1f&msbk 2f&msbk
sh sh vh h vh vh lo sh sh sh
fr fr fir fir vfir vfir lo fr fr fr
sp sp sp sp vsvp vsvp sopo sopo sp sp
– 1fpf 1fpf 1dpf 2dpf 2dpf – – – –
WPDI g
0.48 0.53
CL SiCL CL C C C SL SCL SCL SC
3 3.75 2.5 7.5 5 7.5 0 0 3.33 3.33
0.53 0.39 0.53 (PM2) 0.46 0.52 0.57 0.64 0.65 0.18 0.32 (PM2) 0.27 0.36 0.39
a
Using the US system; SL = sandy loam, C = clay, CL = clay loam, SiC = silty clay, SiCL = silty clay loam. sg = single grain, 3 = strong, 2 = moderate, 1 = weak, f = fine, vf = very fine, m = medium, c = coarse, gr = granular, abk = angular blocky, sbk = subangular blocky, vc = very coarse. c h = hard, lo = loose, fir = firm, fr = friable, v = very, w = weak, o = none, s = sticky, p = plastic. d 1 = few; 2 = common; f = faint; d = distinct; pf = ped face. e RI; Redness index = ((10 − hue) × chroma) / value (Torrent et al., 1980, 1983). f HI; Horizon index, PDI = The sum of HI multiplied by horizon thickness (Appendix 2 in Birkeland, 1999). g WPDI; Weighted mean of profile development index calculated based on parent material from the Choshui River (PM1) and the Chinshui River (PM2), WPDI = PDI divided by the pedon thickness (Appendix 2 in Birkeland, 1999). h No feature. b
reddish hue or lower RI values were found in pedons TL-4 and CS-2. Except for TL-4, all the soils exhibit fine soil texture with a sand fraction lower than 40% (Table 3). The wide range from sandy clay loam to clay in texture classes exhibits a decreasing trend of sand content with the altitude of the terraces (Fig. 4). Fine and very fine sands are the dominant subdivisions of the sand fraction divided into five classes, reflecting the fine and highly weatherable parent materials of the soils (Table 3). As a result, a high rate of soil development is expected. Pedons CS-1 and TL-1 show a similar particle size distribution, low contents of each sand fraction. This suggests an equally long period of weathering for the parent materials. The fine texture of the soils reflects the morphologies of structure and consistence. The considerable develop-
ment in structure and degree of stickiness and plasticity are in accordance with the abundant clay coatings on the pedsurface observed in the field. Strongly coarse angular and subangular blocky structures are generally found in CS-1, CS-2 and TL-1 pedons, while moderately finer ones are observed in TL-4 pedons. Clear accumulation of illuvial clay is identified from the faint or distinct clay coatings on the pedsurface of subsurface horizons in each pedon, except for pedon TL-4 (Fig. 5; Table 2). Additionally, high clay contents were found in all the soils related to strong chemical weathering (Table 3). The vertical variations of clay and silt along the profiles also indicate a strong degree of leaching and weathering within the soils (Fig. 5). However, the clay contents of the Chushan terraces are generally higher than those of the Touliu terraces. Based on the observed
208
H. Tsai et al. / Geomorphology 88 (2007) 201–213
Table 3 Particle size distribution of the studied pedons Pedon Horizon Size distribution (%)
Ultisols with the Subgroup of Typic Paleudults following the Soil Taxonomy (Soil Survey Staff, 2003). The soil developed on TL-4 is as Typic Dystrudepts because of the absence of clay coatings in its B horizons.
Sand-size class a
Sand Silt Clay VC% C%
M% F%
VF%
4.2. WPDI and pedogenic Fe
Chushan terraces CS-1 A 25 Bt1 18 Bt2 14 Bt3 16 Bt4 17 CS-2 A 38 Btv1 35 Btv2 28 Btv3 18 Btv4 14 B4 14
47 46 42 36 39 30 29 32 42 46 50
28 36 44 48 44 32 36 40 40 40 36
0.3 0.1 0 0.2 0.6 0 0 0 0 0 0
1.8 1.8 1.5 1.9 2.8 0 0.1 0.1 0 0 0.1
2.1 1.9 1.7 2.1 2.3 0.5 0.4 0.3 0.1 0.1 0.1
7.7 4.6 4.9 6.4 5.2 25.9 20.0 20.0 8.6 7.3 1.6
13.3 9.7 6.3 5.0 6.5 11.2 14.5 7.9 9.4 6.6 12.5
Touliu terraces TL-1 A Bt1 Bt2 Bt3 Bt4 Bt5 TL-4 A AB Bw1 Bw2
50 47 42 35 33 32 17 20 21 12
28 36 32 40 44 48 20 24 32 36
0.2 0.1 0 0.1 0.6 0 2.1 2.4 1.9 6.8
0.7 0.4 1.1 1.0 1.1 0.9 11.5 8.4 7.3 11.4
0.8 0.6 1.6 1.6 1.8 1.7 7.6 6.5 5.7 5.5
7.6 3.6 9.2 11.4 7.4 7.4 19.4 24.2 22.3 14.7
12.8 12.0 14.0 11.3 11.8 10.1 21.9 14.0 10.0 13.6
22 17 26 25 23 20 63 56 47 52
a VC = very coarse (2.00–1.00 mm); C = coarse (1.00–0.50 mm); M = medium (0.50–0.25 mm); F = fine (0.25–0.10 mm); VF = very fine (0.10–0.05 mm).
trend of clay accumulation in the subsurface horizons as well as on the evidence of clay coating and clay contents, the soils of CS-1, CS-2 and TL-1 are classified as
The field morphology of the soils was quantified following the procedure mentioned above (Fig. 3). The degree of differentiation between each soil horizon is revealed by the vertical variation of Horizon Indices (HI). In general, the HI value increases with soil depth with transition to a sharp increase in subsurface B horizons along the profile (Fig. 6). Such differentiation becomes progressively greater for pedons located at higher altitudes, such as CS-1 and TL-1. The WPDI values of the soils range from 0.32 to 0.56 (Tables 1 and 2). The higher the WPDI value, the older the age is indicated for the soil. All the pedons show a different degree of rubification. This is attributed to different degrees of chemical weathering associated with Fe released from silicate minerals. In addition, the soils bearing Ferromagnesianrich minerals from detrital slates or phyllites possess high Fet contents. Furthermore, the free Fe (Fed) content of soils reached up to 30 g/kg (Table 4). However, the amorphous Fe (Feo) is relatively low, such that the ratio of oxalate to DCB extractable Fe (Feo/Fed value) is always b 0.5, indicating a high degree of Fe crystallization (Blume and Schwertmann, 1969). The accumulation of Fed in subsurface horizons indicates the migration of free Fe from the eluvial to the illuvial horizons, particularly in the soils on the terraces at high
Fig. 6. The profile variation of horizontal index (HI) of the soils from (A) Chushan and Touliu terraces, and (B) Pakua terraces (from Tsai et al., 2006).
H. Tsai et al. / Geomorphology 88 (2007) 201–213
209
altitude (Fig. 5). Except for TL-4, all pedons are within Nagatsuka's (1972) red soil region [Feo/Fed ≤ 0.4, (Fed − Feo)/Fet ≥ 0.5] (Fig. 7). 4.3. Regional terrace correlation The geomorphologic correlation of river terraces is commonly based on parameters like altitude above sea level or the present river channel, gradient, longitudinal profiles etc. However, these geomorphic features are often inadequate to correlating river terraces deformed by active tectonics. The HI and WPDI indexes obtained from various morphological characteristics of the studied soils, in addition to soil color, are applied for this purpose. The river terraces are correlated in a chronological order of I, II, III, IV to n. Based on the degree of soil development, the chronosequence of the Pakua terraces are designated with roman numerals from I (oldest) to VI (youngest) accordingly (Tsai et al., 2006) (Table 1). The soils with similar WPDI values can be correlated. In fact, the correlation is best provided from the vertical variation of HI index along the soil profiles (Fig. 6). In spite of the difference in WPDI values between CS-1 and TL-1, the similar pattern in HI along the profiles strongly suggests an equivalent degree of pedogenesis. Table 4 Free Fe oxides of the studied pedons Pedon
Horizon
Free Fe oxides (g/kg)
Feo/ Fed
(Fed − Feo)/ Fet
Feo
Fed
Fet
Chushan terraces CS-1 A Bt1 Bt2 Bt3 Bt4 CS-2 A Btv1 Btv2 Btv3 Btv4 B4
3.8 4.6 4.7 4.2 3.3 1.5 1.1 1.3 0.8 1.1 0.9
26.0 28.5 33.7 36.5 33.3 22.2 26.3 27.4 31.0 25.0 30.8
41.1 42.0 41.6 42.7 47.0 33.2 37.1 38.8 37.1 40.7 36.3
0.15 0.16 0.14 0.11 0.10 0.07 0.04 0.05 0.03 0.04 0.03
0.54 0.57 0.70 0.76 0.64 0.62 0.68 0.67 0.81 0.59 0.83
Touliu terraces TL-1 A Bt1 Bt2 Bt3 Bt4 Bt5 TL-4 A AB Bw1 Bw2
3.92 5.37 4.80 4.09 3.24 3.76 2.99 3.27 3.42 4.43
15.9 14.5 14.3 16.6 15.8 18.8 8.24 10.2 10.5 9.86
16.0 16.1 17.8 15.4 24.3 26.8 19.2 11.6 25.9 30.6
0.25 0.37 0.34 0.25 0.21 0.20 0.36 0.32 0.33 0.45
0.75 0.57 0.53 0.81 0.52 0.56 0.27 0.60 0.27 0.18
Fig. 7. Plot of Fe (Feo/Fed) versus (Fed − Feo)/Fet for soils of the Chushan, Touliu and Pakua terraces. The solid line represents the regressive power curve of all data. The soils with known ages of the Kikai Island (Maejima et al., 2002) are plotted for comparison.
As a result, a sequence of CS-1 = TL-1 N CS-2 N TL-4 from high to low in the degree of soil development, is proposed as basis for regional correlation of the river terraces. The WPDI and HI values indicate that the pedogenesis of CS-1 and CS-2 are respectively equivalent to PK-4 and PK-5, which are correlated as stages IV and V. TL-4, the youngest soil, is correlated as stage VII in the sequence. Regional correlation is established in Table 1. The resultant correlation suggests that the terraces CS-1 and TL-1 correspond to the same geomorphic surface that was divided by incision of the Chinshui River when PK-4 of the Pakua river terrace was developed. The soils of the terraces based on the correlation from the youngest to oldest agree with the development sequence of Inceptsol → Paleudult → Oxisol indicated in the Soil Taxonomy (Table 1). 5. Discussion 5.1. Pedologic chronometry Soil age can be estimated by its positive relationship to Fe activity (Feo/Fed) and/or negative relationship to Fe crystallinity (Fed − Feo)/Fet (Arduino et al., 1984; MacFadden and Hendricks, 1985; Arduino et al., 1986). In the example from Japan, Maejima et al. (2002) showed a highly positive correlation between the absolute ages (Nagatsuka and Maejima, 2001) and crystallinity ratios of free iron oxides for the lateritic red soils on raised coral reef terraces of Kikai Island, and successfully estimated ages of the red-colored soils developed on Minamidaito Island. Fig. 7 shows the relationship of (Feo/Fed) versus (Fed − Feo)/Fet for the study soils mentioned above. The data of Chushan and
210
H. Tsai et al. / Geomorphology 88 (2007) 201–213
Fig. 8. Relationship between WPDIs and soil ages indicating the rate of soil development.
Touliu terraces roughly agree with those of Kikai Island and Pakua terraces. The wide scattering of data from each pedon overlaps with other pedons, which makes it difficult to estimate a correct age. Moreover, the higher (Feo/Fed) and lower (Fed − Feo)/Fet ratios of CS-1 compared with CS-2 in the observed relationship suggest CS-1 to be younger than CS-2, which does not conform to the likely sequence in terrace formation. There are studies showing the increase in the values of soil development index with time, which is suitable for chronometric application (Harden and Taylor, 1983; Rodbell, 1990; Vidic and Lobnik, 1997). In this study, the pedologic chronometry is established from three soils with known ages by linear regression. The soils of Chushan terraces give WPDI values of 0.51 (CS-2) and 0.56 (CS-1) for ages of 30 ka and 92 ka, respectively (Tables 1 and 2). The highest WPDI value of 0.73 represents the oldest soils (PK-1) of Pakua terraces. The maximum age of the Pakua terraces lies between 350– 500 ka (Liew, 1988; Delcaillau et al., 1998; Tsai et al., 2006). The mean value and the range of ages from PK-1 introduce uncertainties in age estimation of other undated soils or river terraces from regression (Fig. 8; Table 1). Although the extrapolation of the linear trend does not yield proper ages for TL-4 and PK-6, their relative ages can be obtained by pedogenic correlation based on the WPDI and HI values.
Touliu terraces (Lin, 1957; Ota et al., 2002). On the contrary, the pedogenic correlation of the river terraces indicates the Chushan terraces and the Touliu terraces are associated with same growing fold structure that has become uplifted and tilted. The smooth and flat surfaces of the Touliu terraces are uplifted above the hinge of the anticline, and the tilted surfaces of the Chushan terraces represent the east flank of the anticline (Fig. 2). They appear to be the southern extension of the Pakua Anticline folded by the slip of the fault dipping eastward, which favors the argument that the Changhua Fault extends along the west of the Touliu Hill (CPC, 1982; CGS, 2004). Although Suppe and Namson (1979) suggested the geological structure of Touliu Hill fits the fault–bend folding, Lin et al. (2000b) applied the model of faultpropagation-fold to depict the geologic evolution of this area. The geomorphic profiles show that the extension along the CS-1 surface toward the Touliu Hill reaches an altitude much higher above the TL-1 surface, indicating an anticline with a long, flat top, bordered by shorter limbs (Fig. 9A). The flat-topped geometry of anticline presented by the Touliu terraces and the Chushan terraces, in contrast to the narrow and sharp crest of the Pakua anticline, indicates a mechanism by fault–bend folding (Fig. 9B). According to Delcaillau (2001), the
5.2. Geomorphologic implications Previous studies suggested the Chushan Tableland and the Touliu Hill belong to different tectonic units based on their different patterns of terrace deformation, which forms the extension of the Changhua fault along the Chinshui River between the Chushan terraces and
Fig. 9. (A) Geomorphic profiles of the river terraces along the Choshui River. Shaded block indicates the hanging-wall of the Changhua Fault. (Geomorphic profiles of Pakua terraces from Tsai et al., 2006.) (B) Models of fold growth associated with thrust fault. Dotted lines are axial surfaces of the growing fold (Modified after Suppe, 1983, 1985).
H. Tsai et al. / Geomorphology 88 (2007) 201–213
Pakua terraces grew as a lateral propagation (westward and southward) of an active anticline, which had greater deformation in the south where a fault-propagation-fold may have formed. However, seismic and borehole evidence proved another fault–bend–fold system under the northern Pakua Tableland (Hsiao, 1968; Chen, 1978; Wang et al., 2003). The transitions between folding mechanisms along the Changhua Fault indicate the fault growth from segment linkage, which agrees with the geomorphologic observations (Yang, 1986; Delcaillau, 2001). The present water gap between Pakua Tableland and Touliu Hill probably marks the boundary between two fault segments beneath the surface (Fig 2). 5.3. Rate of soil development The red soils of the Chushan, Touliu and Pakua terraces form a sequence of Inceptsol, Ultisol and Oxisol following the Soil Taxonomy (Soil Survey Staff, 2003). Their pedogenic development increases with their WPDI values as well as their ages (Table 1). These relationships suggest that the soil WPDIs may assist pedologists not only in soil classification but also in assessing the age of the soils. By relating the WPDI values to soil ages, an approximation of the development rate of Pleistocene soils in Taiwan is revealed (Fig. 8). However, the validity of the linear regression is limited by the number and quality of soil ages. The age (or time) curve of soil development is usually approached by power, exponential or logarithmic chronofunctions (Bockheim, 1980; Birkeland, 1984, 1999). They usually flatten markedly by 10 ka (Rodbell, 1990; Alonso et al., 1994). The linear trend in Fig. 8 represents the rates of soil development of Pleistocene terraces in Taiwan. This rate in a tropical climate is about 2 and 3 times faster than those estimated in Ljubljana Basin, Slovenia (temperate climate) and Salamanca, Spain (Mediterranean type) respectively. Note that the soils of Slovenia (Vidic and Lobnik, 1997) and Spain (Alonso et al., 1994) are compared by recalculating WPDI values with identical soil properties to those used in this study. The increasing trend of the rates toward warmer climate and higher precipitation agrees with the common opinion of climatic effect on soil development (Vidic, 1998). However, their difference is much lower than that reported by Harden (1990), who indicates that the rates in Pleistocene times are more than 10 times slower in semi-arid than in the moister regions. The comparison in this study minimized the bias introduced by soil properties, profile thickness and other factors, thus providing an objective view on soil development.
211
6. Conclusions The soil pedons studied from the Chushan and Touliu terraces in central Taiwan are identified as Inceptsol (TL4) and Ultisol (CS-1, CS-2, TL-1). The WPDI values range from 0.31 to 0.58 for these soils. Based on the degree of soil development, the river terraces (Papua, Chushan and Touliu terraces) of different fluvial systems are chronologically arranged and correlated as I (PK-1), II (PK-2), III (PK-3), IV (PK-4, CS-1 & TL-1), V (PK-5 & CS-2), VI (PK-6) and VII (TL-4). Based on this correlation, the terraces CS-1 and TL-1, separated by the Chinshui River, are considered to correspond to the same surface. Surface deformation reflects the growth of the structure underground in which uplift folding occurred in the hinge zone and tilted to the east in the fold limb, representing the southern extension of the Pakua anticline. However, the different geometries on both the surface and the subsurface from southern Pakua Anticline indicate a mechanism of fault–bend folding for Touliu Hill. The relationship of WPDI values with soil age provides an approximation of the development rate of Pleistocene soils in Taiwan. According to the comparison with representative soils from Mediterranean and Temperate environments, a slightly faster soil development in the study area is revealed. However, the minor differences suggest that the climatic effect on soil pedogenesis becomes less important on soils developed from Pleistocene times. Acknowledgements The authors are grateful to Professors Zueng-Sang Chen and two anonymous reviewers for their critical reading and suggestion. The gratitude also extends to Mr. Hon-Yi Huang, Yan-Gu Chen and Miss. Ya-Wen Hsiao, Fu-Lang Kuo, Hui-Mei Lin for their help in fieldworks and laboratory analyses. The study was financially supported by the project (NSC 94-2116-M018 -001) from the National Science Council, Taiwan. References Alonso, P., Sierra, C., Ortega, E., Dorronsoro, C., 1994. Soil development indices of soils developed on fluvial terraces (Peòaranda de Bracamonte, Sala manca, Spain). Catena 23, 295–308. Angelier, J., 1986. Geodynamics of Eurasia—Philippine Sea Plate boundary. Preface of Tectonophysics 125, IX–X. Arduino, E., Barberis, E., Carraro, F., Forno, M.G., 1984. Estimating relative ages from iron-oxide/total-iron ratios of soils in the western Po Valley, Italy. Geoderma 33, 39–52. Arduino, E., Barberis, E., Marson Ajmone, F., Zanini, E., Franchini, M., 1986. Iron oxides and clay minerals within profiles as indicators of soil age in northern Italy. Geoderma 37, 45–55.
212
H. Tsai et al. / Geomorphology 88 (2007) 201–213
Biq, C., 1992. Another coastal range on Taiwan. Ti-Chih 12, 1–14. Birkeland, P.W., 1984. Holocene soil chronofunctions, southern Alps, New Zealand. Geoderma 34, 115–134. Birkeland, P.W., 1999. Soils and Geomorphology, 3rd ed. Oxford Univ. Press, New York. Blume, H.P., Schwertmann, U., 1969. Genetic evaluation of profile distribution of aluminum, iron, and manganese oxides. Soil Science Society of America Proceedings 33, 438–444. Bockheim, J.G., 1980. Solution and use of chronofunction in studying soil development. Geoderma 24, 71–85. CGS, 2004. Geologic map of Nantou and Yunlin. Central Geological Survey of Taiwan, scale 1:50,000, 2 sheets. Chen, J.S., 1978. A comparative study of the refraction and reflection seismic data obtained on the Changhua Plain to the Peikang Shelf, Taiwan. Petroleum Geology of Taiwan 15, 199–217. Chen, F.Y., Liew, P.M., 1990. Palynological study of the sungshan formation, Taipei basin. Proceedings Geological Society of China 33, 21–37. Chen, Y.G., Liu, T.K., 1991. Radiocarbon dates of river terraces along the lower Tahanchi, northern Taiwan: their tectonic and geomorphic implications. Journal of the Geological Society of China 34, 337–347. CPC, 1982. Geologic map of Taiwan: Chinese Petroleum Corporation, Taiwan, scale 1:100,000, 4 sheets. Deffontaines, B., Lacombe, O., Angelier, J., Chu, H.T., Mouthereau, F., Lee, C.T., Deramond, J., Lee, J.F., Yu, M.S., Liew, P.M., 1997. Quaternary transfer faulting in the Taiwan Foothills: evidence from a multisource approach. Tectonophysics 274, 61–82. Delcaillau, B., 2001. Geomorphic response to growing fault-related folds: example from the foothills of central Taiwan. Geodinamica Acta 14, 265–287. Delcaillau, B., Deffontaines, B., Floissac, L., Angelier, J., Deramond, J., Souquet, P., Chu, H.T., Lee, J.F., 1998. Morphotectonic evidence from lateral propagation of an active frontal fold; Pakuashan anticline, foothills of Taiwan. Geomorphology 24, 263–290. Engel, S.A., Gardner, T.W., Ciolkosz, E.J., 1996. Quaternary soil chronosequences on terraces of the Susquehnna river, Pennsylvania. Geomorphology 17, 273–294. Gee, G.W., Bauder, J.W., 1986. Particle-size analysis. In: Klut, A. (Ed.), Methods of Soil Analysis, Part 1. Physical and Mineralogical Methods. Agronomy Monograph, vol. 9. ASA and SSSA, Madison, WI, pp. 383–411. Harden, J.W., 1982. A quantitative index of soil development from field descriptions, examples from a chronosequence in Central California. Geoderma 28, 1–28. Harden, J.W., 1990. Soil development on stable landforms and implication for landscape studies. Geomorphology 3, 391–398. Harden, J.W., Taylor, E.M., 1983. A quantitative comparison of soil development in four climatic regimes. Quaternary Research 20, 342–359. Ho, C.S., 1988. An Introduction to the Geology of Taiwan: Explanatory Text of the Geologic Map of Taiwan, 2nd ed. Central Geologic Survey, Taiwan, Republic of China. Hsiao, P.T., 1968. Seismic study of the Taichung area, Taiwan. Petroleum Geology of Taiwan 6, 209–216. ISSS Working Group WRB, 1998. World reference bases for soil resource: introduction. In: Deckers, J.A., Nachtergaele, F.O., Spaargaren, O.C. (Eds.), International Society of Soil Science, International Soil Reference and Information Centre (ISRIC), and Food and Agriculture Organization (FAO). Acco, Leuven. 165 pp. Jenny, H., 1941. Factors in Soil Formation. McGraw-Hill, New York. 281 pp.
Kao, H., Chen, W.P., 2000. The Chi-chi earthquake sequence: active outof-sequence thrust faulting in Taiwan. Science 288, 2346–2349. Ku, C.C., 1963. Photogeologic study of terraces in northwestern Taiwan. Proceedings Geological Society of China 6, 51–60. Leamy, M.L., Milne, J.D.G., Pullar, W.A., Bruce, J.G., 1973. Paleopedology and stratigraphy in the New Zealand Quaternary succession. New Zealand Journal of Geology and Geophysics 16, 723–744. Lee, J.C., Lu, C.Y., Chu, H.T., Delcaillau, B., Angelier, J., Deffontaines, B., 1996. Active deformation and paleostressanalysis in the Pakua anticline area, western Taiwan. Terrestrial, Atmospheric and Oceanic Sciences 7, 431–446. Lee, T.Q., Lin, S.F., Chou, H.C., 1999. Preliminary magnetic study of the Quaternary red-soil bed on Linkou terrace, northern Taiwan. Terrestrial, Atmospheric and Oceanic Sciences 10, 763–776. Liew, P.M., 1988. Quaternary stratigraphy in western Taiwan: palynological correlation. Proceedings Geological Society of China 31, 169–180. Liew, P.M., Hsieh, M.L., 2000. Late Holocene (2 ka) sea level, river discharge and climate interrelationship in the Taiwan region. Journal of Asian Earth Sciences 18, 499–505. Liew, P.M., Huang, S.Y., 1994. Pollen analysis and their paleoclimatic implication in the middle Pleistocene lake deposits of the Ilan district, northeastern Taiwan. Journal of the Geological Society of China 37, 115–124. Lin, C.C., 1957. Topography of Taiwan: Publ. of the Taiwan Provincial Documentary Committee. 424 pp. [in Chinese]. Lin, C.W., Huang, H.C., Lu, S.T., Shih, T.S., Huang, W.J., 2000a. Active fault map of Taiwan, 2nd ed. Central Geological Survey of Taiwan, scale 1:500,000, 1 sheet. Lin, C.W., Lu, S.T., Huang, W.J., Shih, T.S., Chang, H.C., 2000b. The Chi-Chi earthquake fault and structural analyses of the area south of Choshuihsi, central Taiwan. Central Geological Survey of Taiwan, Special Publication, vol. 12, pp. 89–111 (Special issue for the Chi-Chi earthquake, 1999.). Liu G.Y. 2004. A Study of subsurface structures in the Touliu terrace of Tongsuhu fault. MS thesis of Nation Central University, Taiwan, 94 pp. [in Chinese]. Lu, C.Y., Jeng, F.S., Chang, K.J., Jian, W.T., 1998. Impact of basement high on the structure and kinematics of the western Taiwan thrust wedge: insights from sandbox models. Terrestrial, Atmospheric and Oceanic Sciences 9, 533–550. Ma, K.F., Lee, C.T., Tsai, Y.B., Shin, T.C., Mori, J., 1999. The Chi-chi, Taiwan earthquake: large surface displacements on an island thrust fault. EOS, Transactions 80, 605–611. MacFadden, L.D., Hendricks, D.M., 1985. Changes in the content and composition of pedogenic iron oxyhydroxides in a chronosequence of soils in southern California. Quaternary Research 23, 189–204. Maejima, Y., Nagatsuka, S., Higashi, T., 2002. Application of crystallinity ratio of free iron for dating soils developed on the raised coral reef terraces of Kikai and Minami-Daito islands, southwest Japan. The Quaternary Research (Daiyonki-Kenkyu) 41, 485–493. McKeague, J.A., Day, J.H., 1966. Dithionite and oxalate extractable Fe and Al as acids in different various classes of soils. Canadian Journal of Soil Science 46, 13–22. Mehra, O.P., Jackson, M.L., 1960. Iron oxides removed from soils and clays by a dithionite–citrate system buffered with sodium bicarbonate. Clays and Clay Minerals 7, 317–327. Morrison, R.B., 1968. Means of time-stratigraphic division and longdistance correlation of Quaternary successions. In: Morrison, R.B.,
H. Tsai et al. / Geomorphology 88 (2007) 201–213 Wright Jr., H.E. (Eds.), Means of Correlation of Quaternary Successions. Int. Assoc. Quat. Res., VII Congress., Proc., vol. 8, pp. 1–113. Mouthereau, F., Lacombe, O., Deffontaines, B., Angelier, J., Chu, H.T., Lee, C.T., 1999. Quaternary transfer faulting and belt front deformation at Pakuashan (Western Taiwan). Tectonics 18, 215–230. Mulcahy, M.J., Churchward, H.M., 1973. Quaternary environments and soils in Australia. Soil Science 116, 156–169. Nagatsuka, S., 1972. Studies on genesis and classification of soils in warm–temperate region of Southwest Japan, part 3. Some features in distribution and mode of existence of free iron and aluminum oxides in the soil profile. Soil Science and Plant Nutrition 18, 147–154. Nagatsuka, S., Maejima, Y., 2001. Dating of f soils on the raised coral reef terraces of Kikai Islands in Ryukyus, southwest Japan. The Quaternary Research (Daiyonki-Kenkyu) 40, 137–147. Ota, Y., Shyu, B.H., Chen, Y.G., Hsieh, M.L., 2002. Deformation and age of fluvial terraces south of the Choushui River, central Taiwan, and their tectonic implications. Western Pacific Earth Sciences 2, 251–260. Reisenauer, H.M., 1982. Chromium. In: Page, A.L., Miller, R.H., Keeney, D.R. (Eds.), Methods of Soil Analysis, Part 2. Chemical and Microbiological Properties. Agronomy Monograph, vol. 9. ASA and SSSA, Madison, WI, pp. 337–346. Rodbell, D.T., 1990. Soil–age relationships on late Quaternary moraines, Arrowsmith Range, Southern Alps, New Zealand. Arctic and Alpine Research 22, 355–365. Seno, T., Stein, S., Gripp, A.E., 1993. A mode for the motion of the Philippine Sea plate consistent with NUVEL-1 and geological data. Journal of Geophysical Research 98, 17941–17948. Soil Survey Staff, 1993. Examination and description of soils in the field. Soil Survey Manual, Handbook, vol. 18. USDA-SCS, Washington, D.C., pp. 56–196. Soil Survey Staff, 2003. Keys to Soil Taxonomy, 9th eds. USDANRCS, Washington, D.C. 332 pp. Sung, Q.C., Chen, Y.C., 2004. Geomorphic evidence and kinematic model for Quaternary transfer faulting of the Pakuashan anticline, Central Taiwan. Journal of Asian Earth Sciences 24, 389–404. Sung, Q.C., Lu, M.T., Tsai, H., Liu, P.M., 1997. Discussion on the genetics and the correlation of river terraces in Taiwan. J. Geol. Soc. China 40, 31–46.
213
Suppe, J., 1981. Mechanics of mountain building in Taiwan. Memoir of the Geological Society of China 4, 67–89. Suppe, J., 1983. Geometry and hinematics of fault–bend folding. American Journal of Science 283, 648–721. Suppe, J., 1985. Principles of Structural Geology. Prentice Hall, New Jersey. Suppe, J., Namson, J., 1979. Fault–bend origin of frontal folds of western Taiwan fold-and-thrust belt. Proc. Geol. Soc. China, 16, 11Chang, S. L., 1971. Subsurface Geologic Study of the Taichung Basin, Taiwan. Petrol. Geol. Taiwan, vol. 8, pp. 21–45. Torrent, J., Schwertmann, U., Schulze, D.J., 1980. Iron oxide mineralogy of some soils of two river terrace sequences in Spain. Geoderma 23, 191–208. Torrent, J., Schwertmann, U., Fechter, H., Alferez, F., 1983. Quantitative relationship between soil color and hematite content. Soil Science 136, 354–358. Tsai, Y.B., 1986. Seismotectonics of Taiwan. Tectonophysics 125, 17–38. Tsai, H., Huang, W.S., Hseu, Z.Y., Chen, Z.S., 2006. A river terrace soil chronosequence of the Pakua tableland in Taiwan. Soil Science 171, 167–179. Tseng, M.H., Liew, P.M., Chi, W.R., Shih, T.S., 1992. Pollen analysis of the Tananwan formation, northern Taiwan. Journal of the Geological Society of China 35, 247–259. Vidic, N.J., 1998. Soil–age relationships and correlations: comparison of chronosequences in the Ljubljana Basin, Slovenia and USA. Catena 34, 113–129. Vidic, N.J., Lobnik, F., 1997. Rates of soils development of the chronosequence in the Ljubljana Basin, Slovenia. Geoderma 76, 35–64. Wang, C.Y., Kuo, S.Y, Shyu, W.L., Hsiao, J.W., 2003. Investigating near-surface structures under the Changhua fault, west-central Taiwan by the reflection seismic method. Terrestrial, Atmospheric and Oceanic Sciences 14, 343–367. Yang, K.S., 1986. The geomorphological study on active faults in Taiwan—discussion on relationship between active fault and geomorphic surface. Ph.D. thesis of Chinese Culture University, Taiwan, 178 pp. [in Chinese]. Yu, S.B., Chen, H.Y., Kuo, L.C., 1997. Velocity field of GPS stations in the Taiwan area. Tectonophysics 274, 41–59.