- Email: [email protected]
Fabric analysis of till clasts in the upper Urumqi River, Tian Shan, China
ARTICLE IN PRESS
Quaternary International 154–155 (2006) 19–25
Fabric analysis of till clasts in the upper Urumqi River, Tian Shan, China Dewen Lia,b,, Chaolu Yia,b, Baoqi Mac, Pengling Wanga,b, Chunmei Mab, Gongbi Chengb a
Institute of Tibetan Plateau Research, Chinese Academy of Sciences, 100085 Beijing, PR China b Department of Geography, Nanjing University, 210093 Nanjing, PR China c Institute of Crustal Dynamics, China Earthquake Administration, 100085 Beijing, PR China Available online 13 March 2006
Abstract In-situ measurement of clast fabric within till and other sediment was performed at locations in Hayisa (HY), Upper-Wangfeng (UW), Lower-Wangfeng (LW) and Balatigou (BL), in the upper Urumqi River in the Tian Shan of NW China. Subglacial tills have stronger fabric of both a-axis and a–b-plane than supraglacial till. This can be explained as a result from the different stress and transport histories of till clasts. The preferred orientation of a-axes of till fabrics intersect the former ice-flow direction at a sharp angle (627161); whereas the a–b-plane, with its normal intersecting the former ice-flow at 84741, is oriented approximately parallel to the ice flow direction. Thereby, subglacial relief appears to strongly influence the till fabric. In particular, the a-axis fabric is more sensitive to landforms than that of the a–b plane. In simple straight troughs, such as the Wangfeng Valley, fabric varies little. In the complex section of the trough, such as the Hayisa Lipolith, fabric characteristics are different at different sites of the Hayisa Drumlin. The fabric data can be used to understand glacial sedimentary processes in the study area. r 2006 Elsevier Ltd and INQUA. All rights reserved.
1. Introduction Fabric analysis is the study of the preferred orientation of clasts within sediment. For more than 70 yr, fabric analysis has been widely applied in material sciences and geosciences (Wenk, 1985). Recently, both theory and practice of fabric analysis have rapidly advanced with the advancement of measuring devices and processing techniques (Mills, 1977; Hooyer and Iverson, 2000; Benn and Ringrose, 2001; Henriksen et al., 2001; Sakai et al., 2002). Traditional in-situ fabric measuring is the most effective method in the fabric analysis of coarse-size clastic deposits. The formation of till deposits is essentially controlled by dynamic processes. Up to now, fabric analysis is still one of the main methods to obtain information about former glacier dynamics. Till fabrics provide important evidence on glacier movement orientation, sedimentary processes Corresponding author. Institute of Tibetan Plateau Research, Chinese Academy of Sciences, P.O. Box 2871, No. 18 Shuangqing Rd, 100085 Beijing, PR China. E-mail address: [email protected] (D. Li).
and environmental conditions. The spatial variation of fabric and its origin in local settings are significant for the study of glaciogenic deposits. It is still debated whether fabric can be used to distinguish different types of till. For example, Dowdeswell et al. (1985) and Dowdeswell and Sharp (1986) believed that fabric varies with different till types and that the eigenvalues of a-axis of tills can be effectively used to distinguish till types. In contrast, Bennett et al. (1999) argued that fabric data are unable to independently determine origin of tills. Their results indicate that the application of one uniform model in explaining till fabric is neither reliable nor practical. Therefore, further study is necessary for application of fabric parameters for the reconstruction of Quaternary glaciations and regional sedimentary history. The upper Urumqi River is located on northern slope of the Kalaucheng Mountain in the middle of the Tian Shan of China. It is one of the most-studied Quaternary glaciercovered areas in the world (Fig. 1). Several, mostly qualitative, previous fabric studies on tills in this area have been published since the 1980s (Cui, 1981; Li et al., 1981; Feng and Qin, 1984; Li and Zhou, 1984; Yi and Cui,
1040-6182/$ - see front matter r 2006 Elsevier Ltd and INQUA. All rights reserved. doi:10.1016/j.quaint.2006.02.016
ARTICLE IN PRESS D. Li et al. / Quaternary International 154–155 (2006) 19–25
20
2001; Yi et al., 2002). This paper attempts to quantify diversity and variation of fabric characteristics in a variety of settings and for different till types. 2. Geological profile and study methods The Urumqi River originates from Glacier No. 1 and flows eastward. The study area is located within the Daxigou Valley (Fig. 1). A glaciated lopolith is in its western part, called Hayisa Lopolith, about 1 km wide and with many bedrock exposures. A circular hill, the Hayisa Drumlin, is situated in the center of the lopolith. The middle part of the study area is the Wangfeng Valley. There are widespread lower moraines in the more western part and higher moraine ridges in the eastern part. The eastern part of the study area is a deeply incised valley. There are uncemented deposits in the side tributary gullies of the main valley. Four profiles, viz. Hayisa (HY), Upper Wangfeng (UW), Lower Wangfeng (LW) and Balatigou (BL), respectively, were investigated (Fig. 1 and Table 1). The HY profile, a natural one formed by recent incision of the river, is located in the left bank of the river near the Hayisa Hydrometric Station, in the South side of the Hayisa
BL(Fig.6)
Hayisa Drumlin UW( Fig. 4)
Gla cier No. 1
LW(Fig.5)
HY( Fig. 3) Balatigou RMS
Wangf eng RMS 60°E
Modern gla cier
Sediments
Profile
River
90°E
120°E
30°N
Drumlin (Fig. 1). This profile comprises a till layer, which is consolidated and about 4–8 m thick and underlying lower bedrock. The three measured sites are distributed within the till layer in the stoss (TS07), lee (TS09) and top sides (TS08) (Fig. 3). The UW and LW profiles are about 500 and 100 m away, respectively, to the West from the Wangfeng Road Maintenance Station (RMS) and are located within road cuttings along Daxigou section of the Urumqi–Korla National Road (Fig. 1). The UW, deposited during the Last Glaciation (Yi et al., 2002, 2004), comprises two till layers. The lower is pale yellow till with compact texture and high silt content; while the upper grey unit has loose texture and dipping coarse beds. The six measuring sites were situated in the UW: the TS10, TS11 and TS12 within the upper layer and TS4, TS5 and TS6 within the lower layer (Fig. 4). The LW is composed of till older than the UW (Wang, 1981; Yi et al., 2002, 2004). Sampling of TS3 and TS2 was undertaken in this profile (Fig. 5). The clasts are very coarse in the upper part of the LW. Especially in the more western part (TS3) of the exposure, the clasts reach several meters in diameter. The BL is about 200 m away, to West from the Balatigou RMS, and situated in a gully on the North side of the Urumqi River (Fig. 1). Measurement site TS1 is located within this profile (Fig. 6). The error of in-situ measurement of clasts fabric is closely related to its fieldwork method, especially to profile selection. Some scientists, such as Millar and Nelson (2001a, b) and Klein and Davis (2002), think that large errors will result if only the clasts that are distributed on the exposed surface are measured. To help eliminate this error, the fabrics measured in this study were performed in deep pits. A total of at least 100 clasts (except for TS01 with 70) with long axis X3 cm were counted in each site. The measured items included the length of the three axes and the attitudes of the a–b-plane and a-axis. The total number of the measured clasts was 1170. Information for the samples is summarized in Table 1. The procedures included:
10°N 1km
Fig. 1. Location of the study area and distribution of measuring profiles.
(a) In-situ measurement of the attitudes of a-axis (the largest of three axes) and a–b plane (the maximum oblate plane) of clasts.
Table 1 Distribution and number of fabric sites, Urumqi Riverhead, NW China District
Altitude above sealevel (m)
Strike of valley (deg.)
Location
Total number
Sampling number
Hayisa
3500
085
3
TS07, 08, 09
Upper Wangfeng
3300
125
Lower Wangfeng Balatigou
3200 2800
120 085
Left bank of Daxigou in the downstream from weather station Upper layer of road-cutting profile Lower layer of road-cutting profile Road-cutting profile Gully profile
3 3 2 1
TS10, 11, 12 TS04, 05, 06 TS02, 03 TS01
ARTICLE IN PRESS D. Li et al. / Quaternary International 154–155 (2006) 19–25
(b) Computation of the morphologic parameters of clasts: c/a and (ab)/(ac), provided by Sneed and Folk (1958). (c) Computation of the fabric parameters with eigenvalue method (Mark, 1973; Woodcock, 1977), which computes the orientation (V1, V2 and V3) and strength (S1, S2 and S3) of eigenvectors from a preferred matrix. The preferred matrix is defined as 2 2 3 xi xi yi xi zi n n X X 6x y 2 yi z i 7 A¼ xi xTi ¼ 4 i i yi 5, i¼1 i¼1 xi zi yi zi z2i
21
where xi, yi and zi are projections of i-vector to three coordinate axes (X, Y and Z), respectively. Fabric ‘‘shape’’ can be represented as a preferred ellipsoid (PE). The lengths of its three axes are, respectively, three eigenvalues (the longest S1, the middle S2 and the shortest S3) (Mark, 1973; Benn, 1994). The above calculation is carried out using StereoNett (Version 2.46) software, developed by J. Duyster, at the Institute of Geology, Mineralogy and Geophysics, the RuhrUniversity, Bochum, Germany. (d) Computation of the isotropy (I ¼ S3/S1) and elongation (E ¼ 1(S2/S1)) values, which are introduced
Table 2 The a-axis fabric parameters of till clasts, Urumqi Riverhead, Tian Shan, China Location
No.
n
S1
S2
S3
V1
y (deg.)
I
E
Hayisa
TS07 TS08 TS09
100 100 100
0.49 0.62 0.60
0.37 0.31 0.33
0.14 0.07 0.08
137.1, 23.0 159.6, 22.2 172.0, 16.8
55.57 75.77 87.13
0.29 0.11 0.13
0.24 0.50 0.45
Upper Wangfeng
TS12 TS10 TS11 TS04 TS05 TS06
100 100 100 100 100 100
0.60 0.55 0.57 0.63 0.63 0.64
0.31 0.33 0.35 0.28 0.30 0.27
0.09 0.12 0.08 0.09 0.07 0.08
224.6, 186.2, 134.0, 190.3, 190.2, 193.5,
4.8 10.3 11.2 14.9 18.6 19.9
80.43 61.71 14.33 66.18 66.58 69.84
0.15 0.22 0.14 0.14 0.11 0.13
0.48 0.40 0.39 0.56 0.52 0.58
Lower Wangfeng
TS03 TS02
100 100
0.57 0.50
0.33 0.32
0.10 0.18
205.9, 12.6 187.2, 12.0
86 67.73
0.18 0.36
0.42 0.36
Balatigou
TS01
70
0.53
0.34
0.13
72.7, 13.9
18.48
0.25
0.36
n is number of clasts. S1, S2 and S3 are eigenvalues. V1 is the dip and obliquity of the preferred orientation of the S1. y is the angle at which V1 intersects strike of the local valley. I and E are the isotropy (S3/S1) and elongation (1(S2/S1)) values introduced by Benn.
Table 3 The a–b-plane fabric parameters of till clasts, Urumqi Riverhead, Tian Shan, China Location
No.
n
S1
S2
S3
V1 (deg., deg.)
y (deg.)
I
E
Hayisa
TS07 TS08 TS09
100 100 100
0.75 0.76 0.75
0.15 0.15 0.18
0.10 0.09 0.07
292.1, 71.2 346.8, 75.7 349.1, 74.0
73.33 87.98 88.38
0.13 0.12 0.09
0.80 0.80 0.76
Upper Wangfeng
TS12 TS10 TS11 TS04 TS05 TS06
100 100 100 100 100 100
0.73 0.75 0.67 0.76 0.77 0.74
0.16 0.14 0.20 0.16 0.17 0.17
0.10 0.11 0.12 0.09 0.06 0.09
26.5, 85.9 20.8, 76.8 250.9, 76.0 25.9, 75.9 334.8, 75.6 358.0, 73.5
89.39 86.79 81.84 87.79 77.54 80.16
0.14 0.15 0.18 0.12 0.08 0.12
0.78 0.81 0.70 0.79 0.78 0.77
Lower Wangfeng
TS03 TS02
100 100
0.81 0.67
0.10 0.21
0.09 0.12
8.3, 80.3 43.0, 77.3
86.43 87.17
0.11 0.18
0.88 0.69
Balatigou
TS01
70
0.57
0.25
0.18
213.1, 76.9
81.96
0.32
0.56
The meanings of n, S1, S2, S3, V1, V1, y, I and E are same as in Table 2.
ARTICLE IN PRESS D. Li et al. / Quaternary International 154–155 (2006) 19–25
22
a −b plane
by Benn (1994) and used to describe the shape of the PE.
3. Results and analysis 3.1. Clast morphology and fabric characteristics
TS
TS10
TS11
a− axis
All results are listed in Tables 2 and 3, and shown in Figs. 2–6. The total morphologic features of the clasts are given in Fig. 2, with c/a ¼ 0.3670.14 and (ab)/ (ac) ¼ 0.4970.21 (n ¼ 1170). As shown in Fig. 2, the TS11
S125 °E 10m
Bl
TS10
TS Dml
Dml
Dmm
Dmm 0 300m
b:a
TS06
TS04
TS05
a −b plane
c:a
TS06
TS04
TS05
Fig. 2. Sneed and Folk (1957) triangle of the till clasts morphology (n ¼ 1700)
a− axis
(a - b) / (a - c)
a − b pla ne
Fig. 4. Sampling sites and its density contours of till clasts fabric of the Upper Wangfeng profile (UW).
TS09
a −axis
TS08
20m
TS08
TS07
TS09
Br
E Dmm
0m 300m
600m
Fig. 3. Sampling sites and its density contours of till clasts fabric of the Hayisa profile (HY), the Upper Urumqi River, Tianshan (equal area projection of lower hemisphere, the values of ordinal contours are 1%, 5%, 9%, etc.).
clasts are mainly anisometric and available for analysis and comparison of both a-axis and a–b-plane fabrics. The statistical results of a-axis- and a–b-plane fabric of till clasts analyses are listed in Tables 2 and 3, respectively. The density contours of the sediment clasts fabrics in the study area are shown in Figs. 3–6. A trend of concentration distribution is obvious in the figures, but the density contours of both a-axis and ab-plane fabrics in all measured sites have several peaks with different intensities. It is coincident with the facts observed in the field that the till clasts are nearly parallel in individual clusters, but the orientations of different clusters vary from each other. In all the measured sites, the S1 value indicates that the fabric of the ab-plane of till clasts is stronger than that of the a-axis. This characteristic is contrary to the clast fabrics of typical debris flow deposits, which have stronger a-axis fabrics than those of the ab-planes (Zhou et al., 1992). This means that the fabric of till clasts has a preferred maximal plane to maximal axis (Fig. 7). The I value
ARTICLE IN PRESS D. Li et al. / Quaternary International 154–155 (2006) 19–25 a− b plane
23
a− axis TS02
S1 TS03
S2
10m
Bl
S3
S120°
Ts03
TS02 Dml
Dml
Fig. 7. Comparison of preferred strength of fabric between the a-axis and the ab-plane (the circles indicate a-axis fabrics, and the squares–abplane ones).
0 100m
Fig. 5. Sampling sites and its density contours of till clasts fabric of the Lower Wangfeng profile (LW). a−b plane
a−axis
fabric of the a–b-plane is strong with small variations in the HY and the lower layer of UW, while it is weak with great variations in the LW and the upper layer of the UW. The variation of till a-axis fabric is more complex than that for the ab-plane. 3.3. Relation of clasts fabric with ice flow direction
S120 °E
Dm
TS01 10m Dm
0 80m
Fig. 6. Sampling sites and its density contours of sediment clasts fabric of the Balatigou profile (BL).
indicates that the PE shapes of both a-axis and ab-plane is not isotropic, and the E value indicates that the shape of the PE of the a-axis is remarkably different from that of the ab-plane. The PE of the a–b-plane is approximately rodshaped (E is between 0.56–0.88, average is 0.76), but the PE of the a-axis is more oblate (E is between 0.24–0.58, average 0.44, tenth rows of Tables 2 and 3). In all sites, the dip direction of the V1 varies for both the a-axis and the ab-plane, but the dip angle of V1 varies over small values, representing a small angle for a-axis and a large, or approximately right angles for the ab-plane. 3.2. Clasts fabric comparison between the different profiles As shown in Tables 2–4 and Fig. 7, all profiles have strong a-axis and ab-plane fabrics, except for BL. The
In the study area, the Late Pleistocene alpine glacier was more than 10 km long, and the former ice flow directions can be inferred from the modern valley system. Based on the fact that till-fabric sampling settings are adjacent to the local trough thalweg, it is presumed that the former ice flow direction at the sampling sites parallels approximately the trough strike. The angles (y) of maximum preferred orientation (V1) to valley strikes (third row of Table 1) are separately computed and listed in the eighth rows of Tables 2 and 3. The a-axis y has a large value (eighth row of Table 3, yE627161), indicating that the long axes of clasts tend to intersect ice flow direction at high angles to some degree. The c-axis y, the normal line of the a b-plane, shows that the c-axis is nearly perpendicular to valley strike (eighth row of Table 3, yE84741). This may mean that the maximum oblate plane (ab-plane) of clasts is nearly parallel with the former ice flow direction. 4. Discussion 4.1. Relation of clasts fabric with sediment types The sedimentary facies of the study profiles were studied in detailed by many scholars (Wang and Zhang, 1981; Cui and Xiong, 1989; Xiong, 1991; Cui et al., 1998). The HY is considered to be formed subglacially and by lodgment process according to its geomorphologic and sedimentary features (Cui, 1981; Cui and Xiong, 1989). The UW, as a
ARTICLE IN PRESS D. Li et al. / Quaternary International 154–155 (2006) 19–25
24
Table 4 Comparison of averages of till clast fabrics, Tian Shan, China Profile
HY UW(upper) UW(lower)
a-Axis
c-Axis or ab-plane
Type
S1
y (deg.)
I
E
S1
y (deg.)
I
E
0.5770.07 0.5770.03 0.6370.01
73716 52734 6872
0.18 0.17 0.13
0.40 0.42 0.55
0.7570.01 0.7270.04 0.7670.02
8379 8674 8275
0.12 0.15 0.11
0.79 0.77 0.78
Subglacial Supraglacial Subglacial
S1, y, I and E are as for Table 2.
part of an end moraine, was formed by melting-out processes. Its lower layer is believed to be formed in a subglacial environment because many shear zones are present in the profile (Cui, 1981; Derbyshire, 1984; Feng and Qin, 1984), dipping up-glacier at angles of 18–281 and with a length of several to more than 10 m and thicknesses of 0.8–2 m (Yi et al., 2004). The LW and upper layer of UW, with loose texture and crude bedding, are believed to be supraglacial till (Xiong, 1991). The BL is debris flow deposits, predominately originating from the upsteam tills (Deng, 1995). However, in the study area the fabrics of tills are stronger than non-till fabrics (BL). Here, the subglacial till has the strongest fabrics. The strong subglacial till fabrics probably formed by clast orientation during shear and squeezing, whereas the weaker fabrics of debris flow deposits probably reflect a wide variety of flow conditions during gravitational reworking. The supraglacial clasts originated mainly from the two sides of the glacial troughs. It moved passively on the glacier’s surface, and had little or no chance to enter into englacal transport. Their fabric is partly controlled by glacier dynamics, surface relief and valley form, and partly inherited from those fabrics formed by other processes, e.g., debris flow on the glacier surface, which is not associated with glacier dynamics. The diversity of controlling factors may explain why the surpaglacial till has a weaker and more changeable fabric in the upper layer of UW than the subglacial till in the lower layer UW.
retreat, except for the HY profile. Compaction or creeping after deposition appears to have had little influence on the clast fabrics because the till profiles are not very high and are distributed within glaciated U-trough with low relief. There is no other evidence showing post-depositional modification. However, the clast size seems to have an influence on the fabrics. In the West end of LW, where there are many till clasts with the biggest average grain size among all sampling sites, the fabric is very strong. It is suggested that clast sizes may influence development of till fabric. 5. Conclusions Results of eigenvector/eigenvalue analysis of clast fabric suggest that the tills here have stronger a-axis and a–bplane fabrics than non-tills, and that the fabrics of the subglacial tills is stronger than that of the supraglacial one. The a–b-plane fabrics are stronger than a-axis fabrics for all till types and are near-parallel to the former ice-flow direction. The a-axis intersects former ice-flow direction at high (/not right) angles, and is more sensitive than is the a–b-plane to the variation of local relief. In straighter valleys in the UW there are consistent fabrics. In contrast, there are different fabrics on the top, lee-side and stoss-side of the drumlin, e.g., in the HY. Acknowledgments
4.2. Influence of local relief and clast sizes It is possible that local relief has influence on the fabric within till deposits. For example, the HY and the lower layer of UW were all formed subglacially. The latter, situated in the straighter section of the Wangfeng Valley, has more consistent fabrics (Table 4) than the former, which is located near the Hayisa Drumlin within the Hayisa Lopolith. In the HY, there are different fabric parameters on the top, the ice-lee side and stoss side of the drumlin (see Tables 2 and 3). The development of till fabrics is controlled by many factors and they contain information on sedimentary processes during, after and before deposition (Glen et al., 1957). The tills in all profiles studied were deposited during the Last Glacial and predominately formed during glacier
This work was supported by Grants from the NSFC (40301004, 40241011, 40271016), the LICCRE Grant BX2001-02, the 100 Talents project of the CAS, 2005, and Tianshan Glacier Observatory. We are grateful to Professor Lewis Owen for improvement of this paper. The comments from Professor Doug Benn and another anonymous referee are acknowledged. References Benn, D.I., 1994. Fabric shape and the interpretation of sedimentary fabric data. Journal of Sedimentary Research A 64, 910–915. Benn, D.I., Ringrose, T.J., 2001. Random variation of fabric eigenvalues: implications for the use of a-axis fabric data to differentiate till facies. Earth Surface Processes and Landforms 26 (3), 295–306.
ARTICLE IN PRESS D. Li et al. / Quaternary International 154–155 (2006) 19–25 Bennett, M.R., Waller, R.I., Glasser, N.F., Hambrey, M.J., Huddart, D., 1999. Glacigenic clast fabrics: genetic fingerprint or wishful thinking? Journal of Quaternary Science 14, 125–135. Cui, Z.J., 1981. Types and features of the moraines and till deposits in the Urumqi Riverhead, Tian Shan. Journal of Glaciology and Geocryology 3 (supp.), 36–48 (in Chinese with English summary). Cui, Z.J., Xiong, H.G., 1989. The sedimentary facies model and characteristics of mountain glacier. Quaternary Sciences 3, 254–267 (in Chinese with English summary). Cui, Z.J., Xiong, H.G., Liu, G.N., Zhu, C., Yi, C.L., 1998. Landform Processes and Sedimentary Features on the Cryosphere of the Central Tianshan. ShiJiazhang. Hebei Science and Technology Publishing House, pp. 1–305 (in Chinese with English summary). Deng, Y.X., 1995. Process of accumulation and characteristics of glacial debris flow deposits transformed by moraine. Acta Sedimentologica Sinica 13 (4), 37–48 (in Chinese with English summary). Derbyshire, E., 1984. Till properties and glacier regime in parts of high Asia: Karakoram and Tian Shan. In: White, R.O. (Ed.), The Evolution of the East Asian Environment. University of Hong Kong Press, Hong Kong, pp. 84–110. Dowdeswell, J.A., Hambrey, M.J., Wu, R.T., 1985. A comparison of clast fabric and shape in late Precambrian and modern glacigenic sediments. Journal of Sedimentary Petrology 55 (5), 691–704. Dowdeswell, J.A., Sharp, M.J., 1986. Characterization of gravel fabric in modern terrestrial glacigenic sediments. Sedimentology 33, 699–710. Feng, Z.D., Qin, D.H., 1984. Sedimentary types and processes of end moraines in the Urumqi Riverhead of the Tian Shan since the last glaciation. Journal of Glaciology and Geocryology 6 (3), 39–49 (in Chinese with English summary). Glen, J.W., Donner, J.J., West, R.G., 1957. On the mechanism by which stones in till become oriented. American Journal of Science 255, 194–205. Henriksen, M., Mangerud, J., Maslenikova, O., Matiouchkov, A., Tveranger, J., 2001. Weichselian stratigraphy and glaciotectonic deformation along the lower Pechora, Arctic Russia. Global and Planetary Change 31, 297–319. Hooyer, T.S., Iverson, N.R., 2000. Clast-fabric development in a shearing granular material: implications for subglacial till and fault gouge. Geological Society of America Bulletin 112 (5), 683–692. Klein, E., Davis D., 2002. Surface Sample Bias and Clast Fabric Interpretation Based on Till, Ditch Plains, Long Island. Long Island Geologists, State University of New York. Li, J.J., Zhou, S.Z., 1984. Morphology and surface features of till clasts. Journal of Glaciology and Geocryology 6 (3), 27–29 (in Chinese with English summary). Li, S.D., Cui, Z.J., Wang, J.T., Zhang, Z.Q., 1981. Petrology and morphology of pebbles of glaciogenic, glacioalluvial and alluvial deposits, the Urumqi Riverhead, Tian Shan. Journal of Glaciology and Geocryology 3 (Suppl.), S78–S91 (in Chinese with English summary).
25
Mark, D.M., 1973. Analysis of axial orientation data, including till fabrics. Geological Society of America Bulletin 84, 1369–1374. Millar, S.W.S., Nelson, F.E., 2001a. Sampling-surface orientation and clast macrofabric in Periglacial colluvium. Earth Surface Processes and Landforms 26, 523–529. Millar, S.W.S., Nelson, F.E., 2001b. Clast fabric in relict Periglacial colluvium, Salamanca re-entrant, southwestern New York, USA. Geografiska Annaler 83A, 145–156. Mills, H.H., 1977. Basal till fabrics of modern alpine glaciers. Geological Society of America Bulletin 88, 824–828. Sakai, T., Yokokawa, M., Kubo, Y., Endo, N., Masuda, F., 2002. Grain fabric of experimental gravity flow deposits. Sedimentary Geology 154, 1–10. Sneed, E.D., Folk, R.L., 1958. Gravels in the lower Colorado River, Texas, a study in particle morphogenesis. Journal of Geology 66, 114–150. Wang, J.T., 1981. Ancient glaciers at the head of Urumqi River, Tian Shan. Journal of Glaciology and Geocryology 3 (supp.), 56–63 (in Chinese with English summary). Wang, J.T., Zhang, Z.Q., 1981. Glacier sediment at the head of Urumqi river, Tian Shan. Journal of Glaciology and Geocryology 3 (Suppl), S49–S56 (in Chinese with English summary). Wenk, H.R. (Ed.), 1985. Preferred Orientation in Deformed Metals and Rocks: An Introduction to Modern Texture Analysis. Academic press, Inc., New York. Woodcock, N.H., 1977. Specification of fabric shapes using an eigenvalue method. Geological Society of America Bulletin 88, 1231–1236. Xiong, H.G., 1991. Comparative research of till with debris flow deposits and glacial-geomorphic process of the Tianshan Mountains. Ph.D. Dissertation, Peking University, pp.118–128 (in Chinese with English summary). Yi, C.L., Cui, Z.J., 2001. Subglacial deformation: evidence from microfabric studies of particles and voids in till from the upper Urumqi River valleys, Tian Shan, China. Journal of Glaciology 47 (159), 607–612. Yi, C.L., Jiao, K.X., Liu, K.X., He, Y.Q., Ye, Y.G., 2002. ESR dating of the sediments of the Last Glaciation at the source area of the Urumqi River, Tian Shan Mountains, China. Quaternary International 97–98, 141–146. Yi, C.L., Liu, K.X., Cui, Z.J., Jiao, K.X., Yao, T.D., He, Y.Q., 2004. AMS radiocarbon dating of Late Quaternary glacial landforms, source of the Urumqi River, Tien Shan—a pilot study of 14C dating on inorganic carbon. Quaternary International 121 (1), 99–107. Zhou, B.G., Li, G., Wang, J.G., Pan, L.L., 1992. Genesis analysis of Quaternary clay–gravel deposits in the Huangshan, East of China. Journal of Glaciology and Geocryology 14 (4), 360–365 (in Chinese with English summary).
Quaternary International 154–155 (2006) 19–25
Fabric analysis of till clasts in the upper Urumqi River, Tian Shan, China Dewen Lia,b,, Chaolu Yia,b, Baoqi Mac, Pengling Wanga,b, Chunmei Mab, Gongbi Chengb a
Institute of Tibetan Plateau Research, Chinese Academy of Sciences, 100085 Beijing, PR China b Department of Geography, Nanjing University, 210093 Nanjing, PR China c Institute of Crustal Dynamics, China Earthquake Administration, 100085 Beijing, PR China Available online 13 March 2006
Abstract In-situ measurement of clast fabric within till and other sediment was performed at locations in Hayisa (HY), Upper-Wangfeng (UW), Lower-Wangfeng (LW) and Balatigou (BL), in the upper Urumqi River in the Tian Shan of NW China. Subglacial tills have stronger fabric of both a-axis and a–b-plane than supraglacial till. This can be explained as a result from the different stress and transport histories of till clasts. The preferred orientation of a-axes of till fabrics intersect the former ice-flow direction at a sharp angle (627161); whereas the a–b-plane, with its normal intersecting the former ice-flow at 84741, is oriented approximately parallel to the ice flow direction. Thereby, subglacial relief appears to strongly influence the till fabric. In particular, the a-axis fabric is more sensitive to landforms than that of the a–b plane. In simple straight troughs, such as the Wangfeng Valley, fabric varies little. In the complex section of the trough, such as the Hayisa Lipolith, fabric characteristics are different at different sites of the Hayisa Drumlin. The fabric data can be used to understand glacial sedimentary processes in the study area. r 2006 Elsevier Ltd and INQUA. All rights reserved.
1. Introduction Fabric analysis is the study of the preferred orientation of clasts within sediment. For more than 70 yr, fabric analysis has been widely applied in material sciences and geosciences (Wenk, 1985). Recently, both theory and practice of fabric analysis have rapidly advanced with the advancement of measuring devices and processing techniques (Mills, 1977; Hooyer and Iverson, 2000; Benn and Ringrose, 2001; Henriksen et al., 2001; Sakai et al., 2002). Traditional in-situ fabric measuring is the most effective method in the fabric analysis of coarse-size clastic deposits. The formation of till deposits is essentially controlled by dynamic processes. Up to now, fabric analysis is still one of the main methods to obtain information about former glacier dynamics. Till fabrics provide important evidence on glacier movement orientation, sedimentary processes Corresponding author. Institute of Tibetan Plateau Research, Chinese Academy of Sciences, P.O. Box 2871, No. 18 Shuangqing Rd, 100085 Beijing, PR China. E-mail address: [email protected] (D. Li).
and environmental conditions. The spatial variation of fabric and its origin in local settings are significant for the study of glaciogenic deposits. It is still debated whether fabric can be used to distinguish different types of till. For example, Dowdeswell et al. (1985) and Dowdeswell and Sharp (1986) believed that fabric varies with different till types and that the eigenvalues of a-axis of tills can be effectively used to distinguish till types. In contrast, Bennett et al. (1999) argued that fabric data are unable to independently determine origin of tills. Their results indicate that the application of one uniform model in explaining till fabric is neither reliable nor practical. Therefore, further study is necessary for application of fabric parameters for the reconstruction of Quaternary glaciations and regional sedimentary history. The upper Urumqi River is located on northern slope of the Kalaucheng Mountain in the middle of the Tian Shan of China. It is one of the most-studied Quaternary glaciercovered areas in the world (Fig. 1). Several, mostly qualitative, previous fabric studies on tills in this area have been published since the 1980s (Cui, 1981; Li et al., 1981; Feng and Qin, 1984; Li and Zhou, 1984; Yi and Cui,
1040-6182/$ - see front matter r 2006 Elsevier Ltd and INQUA. All rights reserved. doi:10.1016/j.quaint.2006.02.016
ARTICLE IN PRESS D. Li et al. / Quaternary International 154–155 (2006) 19–25
20
2001; Yi et al., 2002). This paper attempts to quantify diversity and variation of fabric characteristics in a variety of settings and for different till types. 2. Geological profile and study methods The Urumqi River originates from Glacier No. 1 and flows eastward. The study area is located within the Daxigou Valley (Fig. 1). A glaciated lopolith is in its western part, called Hayisa Lopolith, about 1 km wide and with many bedrock exposures. A circular hill, the Hayisa Drumlin, is situated in the center of the lopolith. The middle part of the study area is the Wangfeng Valley. There are widespread lower moraines in the more western part and higher moraine ridges in the eastern part. The eastern part of the study area is a deeply incised valley. There are uncemented deposits in the side tributary gullies of the main valley. Four profiles, viz. Hayisa (HY), Upper Wangfeng (UW), Lower Wangfeng (LW) and Balatigou (BL), respectively, were investigated (Fig. 1 and Table 1). The HY profile, a natural one formed by recent incision of the river, is located in the left bank of the river near the Hayisa Hydrometric Station, in the South side of the Hayisa
BL(Fig.6)
Hayisa Drumlin UW( Fig. 4)
Gla cier No. 1
LW(Fig.5)
HY( Fig. 3) Balatigou RMS
Wangf eng RMS 60°E
Modern gla cier
Sediments
Profile
River
90°E
120°E
30°N
Drumlin (Fig. 1). This profile comprises a till layer, which is consolidated and about 4–8 m thick and underlying lower bedrock. The three measured sites are distributed within the till layer in the stoss (TS07), lee (TS09) and top sides (TS08) (Fig. 3). The UW and LW profiles are about 500 and 100 m away, respectively, to the West from the Wangfeng Road Maintenance Station (RMS) and are located within road cuttings along Daxigou section of the Urumqi–Korla National Road (Fig. 1). The UW, deposited during the Last Glaciation (Yi et al., 2002, 2004), comprises two till layers. The lower is pale yellow till with compact texture and high silt content; while the upper grey unit has loose texture and dipping coarse beds. The six measuring sites were situated in the UW: the TS10, TS11 and TS12 within the upper layer and TS4, TS5 and TS6 within the lower layer (Fig. 4). The LW is composed of till older than the UW (Wang, 1981; Yi et al., 2002, 2004). Sampling of TS3 and TS2 was undertaken in this profile (Fig. 5). The clasts are very coarse in the upper part of the LW. Especially in the more western part (TS3) of the exposure, the clasts reach several meters in diameter. The BL is about 200 m away, to West from the Balatigou RMS, and situated in a gully on the North side of the Urumqi River (Fig. 1). Measurement site TS1 is located within this profile (Fig. 6). The error of in-situ measurement of clasts fabric is closely related to its fieldwork method, especially to profile selection. Some scientists, such as Millar and Nelson (2001a, b) and Klein and Davis (2002), think that large errors will result if only the clasts that are distributed on the exposed surface are measured. To help eliminate this error, the fabrics measured in this study were performed in deep pits. A total of at least 100 clasts (except for TS01 with 70) with long axis X3 cm were counted in each site. The measured items included the length of the three axes and the attitudes of the a–b-plane and a-axis. The total number of the measured clasts was 1170. Information for the samples is summarized in Table 1. The procedures included:
10°N 1km
Fig. 1. Location of the study area and distribution of measuring profiles.
(a) In-situ measurement of the attitudes of a-axis (the largest of three axes) and a–b plane (the maximum oblate plane) of clasts.
Table 1 Distribution and number of fabric sites, Urumqi Riverhead, NW China District
Altitude above sealevel (m)
Strike of valley (deg.)
Location
Total number
Sampling number
Hayisa
3500
085
3
TS07, 08, 09
Upper Wangfeng
3300
125
Lower Wangfeng Balatigou
3200 2800
120 085
Left bank of Daxigou in the downstream from weather station Upper layer of road-cutting profile Lower layer of road-cutting profile Road-cutting profile Gully profile
3 3 2 1
TS10, 11, 12 TS04, 05, 06 TS02, 03 TS01
ARTICLE IN PRESS D. Li et al. / Quaternary International 154–155 (2006) 19–25
(b) Computation of the morphologic parameters of clasts: c/a and (ab)/(ac), provided by Sneed and Folk (1958). (c) Computation of the fabric parameters with eigenvalue method (Mark, 1973; Woodcock, 1977), which computes the orientation (V1, V2 and V3) and strength (S1, S2 and S3) of eigenvectors from a preferred matrix. The preferred matrix is defined as 2 2 3 xi xi yi xi zi n n X X 6x y 2 yi z i 7 A¼ xi xTi ¼ 4 i i yi 5, i¼1 i¼1 xi zi yi zi z2i
21
where xi, yi and zi are projections of i-vector to three coordinate axes (X, Y and Z), respectively. Fabric ‘‘shape’’ can be represented as a preferred ellipsoid (PE). The lengths of its three axes are, respectively, three eigenvalues (the longest S1, the middle S2 and the shortest S3) (Mark, 1973; Benn, 1994). The above calculation is carried out using StereoNett (Version 2.46) software, developed by J. Duyster, at the Institute of Geology, Mineralogy and Geophysics, the RuhrUniversity, Bochum, Germany. (d) Computation of the isotropy (I ¼ S3/S1) and elongation (E ¼ 1(S2/S1)) values, which are introduced
Table 2 The a-axis fabric parameters of till clasts, Urumqi Riverhead, Tian Shan, China Location
No.
n
S1
S2
S3
V1
y (deg.)
I
E
Hayisa
TS07 TS08 TS09
100 100 100
0.49 0.62 0.60
0.37 0.31 0.33
0.14 0.07 0.08
137.1, 23.0 159.6, 22.2 172.0, 16.8
55.57 75.77 87.13
0.29 0.11 0.13
0.24 0.50 0.45
Upper Wangfeng
TS12 TS10 TS11 TS04 TS05 TS06
100 100 100 100 100 100
0.60 0.55 0.57 0.63 0.63 0.64
0.31 0.33 0.35 0.28 0.30 0.27
0.09 0.12 0.08 0.09 0.07 0.08
224.6, 186.2, 134.0, 190.3, 190.2, 193.5,
4.8 10.3 11.2 14.9 18.6 19.9
80.43 61.71 14.33 66.18 66.58 69.84
0.15 0.22 0.14 0.14 0.11 0.13
0.48 0.40 0.39 0.56 0.52 0.58
Lower Wangfeng
TS03 TS02
100 100
0.57 0.50
0.33 0.32
0.10 0.18
205.9, 12.6 187.2, 12.0
86 67.73
0.18 0.36
0.42 0.36
Balatigou
TS01
70
0.53
0.34
0.13
72.7, 13.9
18.48
0.25
0.36
n is number of clasts. S1, S2 and S3 are eigenvalues. V1 is the dip and obliquity of the preferred orientation of the S1. y is the angle at which V1 intersects strike of the local valley. I and E are the isotropy (S3/S1) and elongation (1(S2/S1)) values introduced by Benn.
Table 3 The a–b-plane fabric parameters of till clasts, Urumqi Riverhead, Tian Shan, China Location
No.
n
S1
S2
S3
V1 (deg., deg.)
y (deg.)
I
E
Hayisa
TS07 TS08 TS09
100 100 100
0.75 0.76 0.75
0.15 0.15 0.18
0.10 0.09 0.07
292.1, 71.2 346.8, 75.7 349.1, 74.0
73.33 87.98 88.38
0.13 0.12 0.09
0.80 0.80 0.76
Upper Wangfeng
TS12 TS10 TS11 TS04 TS05 TS06
100 100 100 100 100 100
0.73 0.75 0.67 0.76 0.77 0.74
0.16 0.14 0.20 0.16 0.17 0.17
0.10 0.11 0.12 0.09 0.06 0.09
26.5, 85.9 20.8, 76.8 250.9, 76.0 25.9, 75.9 334.8, 75.6 358.0, 73.5
89.39 86.79 81.84 87.79 77.54 80.16
0.14 0.15 0.18 0.12 0.08 0.12
0.78 0.81 0.70 0.79 0.78 0.77
Lower Wangfeng
TS03 TS02
100 100
0.81 0.67
0.10 0.21
0.09 0.12
8.3, 80.3 43.0, 77.3
86.43 87.17
0.11 0.18
0.88 0.69
Balatigou
TS01
70
0.57
0.25
0.18
213.1, 76.9
81.96
0.32
0.56
The meanings of n, S1, S2, S3, V1, V1, y, I and E are same as in Table 2.
ARTICLE IN PRESS D. Li et al. / Quaternary International 154–155 (2006) 19–25
22
a −b plane
by Benn (1994) and used to describe the shape of the PE.
3. Results and analysis 3.1. Clast morphology and fabric characteristics
TS
TS10
TS11
a− axis
All results are listed in Tables 2 and 3, and shown in Figs. 2–6. The total morphologic features of the clasts are given in Fig. 2, with c/a ¼ 0.3670.14 and (ab)/ (ac) ¼ 0.4970.21 (n ¼ 1170). As shown in Fig. 2, the TS11
S125 °E 10m
Bl
TS10
TS Dml
Dml
Dmm
Dmm 0 300m
b:a
TS06
TS04
TS05
a −b plane
c:a
TS06
TS04
TS05
Fig. 2. Sneed and Folk (1957) triangle of the till clasts morphology (n ¼ 1700)
a− axis
(a - b) / (a - c)
a − b pla ne
Fig. 4. Sampling sites and its density contours of till clasts fabric of the Upper Wangfeng profile (UW).
TS09
a −axis
TS08
20m
TS08
TS07
TS09
Br
E Dmm
0m 300m
600m
Fig. 3. Sampling sites and its density contours of till clasts fabric of the Hayisa profile (HY), the Upper Urumqi River, Tianshan (equal area projection of lower hemisphere, the values of ordinal contours are 1%, 5%, 9%, etc.).
clasts are mainly anisometric and available for analysis and comparison of both a-axis and a–b-plane fabrics. The statistical results of a-axis- and a–b-plane fabric of till clasts analyses are listed in Tables 2 and 3, respectively. The density contours of the sediment clasts fabrics in the study area are shown in Figs. 3–6. A trend of concentration distribution is obvious in the figures, but the density contours of both a-axis and ab-plane fabrics in all measured sites have several peaks with different intensities. It is coincident with the facts observed in the field that the till clasts are nearly parallel in individual clusters, but the orientations of different clusters vary from each other. In all the measured sites, the S1 value indicates that the fabric of the ab-plane of till clasts is stronger than that of the a-axis. This characteristic is contrary to the clast fabrics of typical debris flow deposits, which have stronger a-axis fabrics than those of the ab-planes (Zhou et al., 1992). This means that the fabric of till clasts has a preferred maximal plane to maximal axis (Fig. 7). The I value
ARTICLE IN PRESS D. Li et al. / Quaternary International 154–155 (2006) 19–25 a− b plane
23
a− axis TS02
S1 TS03
S2
10m
Bl
S3
S120°
Ts03
TS02 Dml
Dml
Fig. 7. Comparison of preferred strength of fabric between the a-axis and the ab-plane (the circles indicate a-axis fabrics, and the squares–abplane ones).
0 100m
Fig. 5. Sampling sites and its density contours of till clasts fabric of the Lower Wangfeng profile (LW). a−b plane
a−axis
fabric of the a–b-plane is strong with small variations in the HY and the lower layer of UW, while it is weak with great variations in the LW and the upper layer of the UW. The variation of till a-axis fabric is more complex than that for the ab-plane. 3.3. Relation of clasts fabric with ice flow direction
S120 °E
Dm
TS01 10m Dm
0 80m
Fig. 6. Sampling sites and its density contours of sediment clasts fabric of the Balatigou profile (BL).
indicates that the PE shapes of both a-axis and ab-plane is not isotropic, and the E value indicates that the shape of the PE of the a-axis is remarkably different from that of the ab-plane. The PE of the a–b-plane is approximately rodshaped (E is between 0.56–0.88, average is 0.76), but the PE of the a-axis is more oblate (E is between 0.24–0.58, average 0.44, tenth rows of Tables 2 and 3). In all sites, the dip direction of the V1 varies for both the a-axis and the ab-plane, but the dip angle of V1 varies over small values, representing a small angle for a-axis and a large, or approximately right angles for the ab-plane. 3.2. Clasts fabric comparison between the different profiles As shown in Tables 2–4 and Fig. 7, all profiles have strong a-axis and ab-plane fabrics, except for BL. The
In the study area, the Late Pleistocene alpine glacier was more than 10 km long, and the former ice flow directions can be inferred from the modern valley system. Based on the fact that till-fabric sampling settings are adjacent to the local trough thalweg, it is presumed that the former ice flow direction at the sampling sites parallels approximately the trough strike. The angles (y) of maximum preferred orientation (V1) to valley strikes (third row of Table 1) are separately computed and listed in the eighth rows of Tables 2 and 3. The a-axis y has a large value (eighth row of Table 3, yE627161), indicating that the long axes of clasts tend to intersect ice flow direction at high angles to some degree. The c-axis y, the normal line of the a b-plane, shows that the c-axis is nearly perpendicular to valley strike (eighth row of Table 3, yE84741). This may mean that the maximum oblate plane (ab-plane) of clasts is nearly parallel with the former ice flow direction. 4. Discussion 4.1. Relation of clasts fabric with sediment types The sedimentary facies of the study profiles were studied in detailed by many scholars (Wang and Zhang, 1981; Cui and Xiong, 1989; Xiong, 1991; Cui et al., 1998). The HY is considered to be formed subglacially and by lodgment process according to its geomorphologic and sedimentary features (Cui, 1981; Cui and Xiong, 1989). The UW, as a
ARTICLE IN PRESS D. Li et al. / Quaternary International 154–155 (2006) 19–25
24
Table 4 Comparison of averages of till clast fabrics, Tian Shan, China Profile
HY UW(upper) UW(lower)
a-Axis
c-Axis or ab-plane
Type
S1
y (deg.)
I
E
S1
y (deg.)
I
E
0.5770.07 0.5770.03 0.6370.01
73716 52734 6872
0.18 0.17 0.13
0.40 0.42 0.55
0.7570.01 0.7270.04 0.7670.02
8379 8674 8275
0.12 0.15 0.11
0.79 0.77 0.78
Subglacial Supraglacial Subglacial
S1, y, I and E are as for Table 2.
part of an end moraine, was formed by melting-out processes. Its lower layer is believed to be formed in a subglacial environment because many shear zones are present in the profile (Cui, 1981; Derbyshire, 1984; Feng and Qin, 1984), dipping up-glacier at angles of 18–281 and with a length of several to more than 10 m and thicknesses of 0.8–2 m (Yi et al., 2004). The LW and upper layer of UW, with loose texture and crude bedding, are believed to be supraglacial till (Xiong, 1991). The BL is debris flow deposits, predominately originating from the upsteam tills (Deng, 1995). However, in the study area the fabrics of tills are stronger than non-till fabrics (BL). Here, the subglacial till has the strongest fabrics. The strong subglacial till fabrics probably formed by clast orientation during shear and squeezing, whereas the weaker fabrics of debris flow deposits probably reflect a wide variety of flow conditions during gravitational reworking. The supraglacial clasts originated mainly from the two sides of the glacial troughs. It moved passively on the glacier’s surface, and had little or no chance to enter into englacal transport. Their fabric is partly controlled by glacier dynamics, surface relief and valley form, and partly inherited from those fabrics formed by other processes, e.g., debris flow on the glacier surface, which is not associated with glacier dynamics. The diversity of controlling factors may explain why the surpaglacial till has a weaker and more changeable fabric in the upper layer of UW than the subglacial till in the lower layer UW.
retreat, except for the HY profile. Compaction or creeping after deposition appears to have had little influence on the clast fabrics because the till profiles are not very high and are distributed within glaciated U-trough with low relief. There is no other evidence showing post-depositional modification. However, the clast size seems to have an influence on the fabrics. In the West end of LW, where there are many till clasts with the biggest average grain size among all sampling sites, the fabric is very strong. It is suggested that clast sizes may influence development of till fabric. 5. Conclusions Results of eigenvector/eigenvalue analysis of clast fabric suggest that the tills here have stronger a-axis and a–bplane fabrics than non-tills, and that the fabrics of the subglacial tills is stronger than that of the supraglacial one. The a–b-plane fabrics are stronger than a-axis fabrics for all till types and are near-parallel to the former ice-flow direction. The a-axis intersects former ice-flow direction at high (/not right) angles, and is more sensitive than is the a–b-plane to the variation of local relief. In straighter valleys in the UW there are consistent fabrics. In contrast, there are different fabrics on the top, lee-side and stoss-side of the drumlin, e.g., in the HY. Acknowledgments
4.2. Influence of local relief and clast sizes It is possible that local relief has influence on the fabric within till deposits. For example, the HY and the lower layer of UW were all formed subglacially. The latter, situated in the straighter section of the Wangfeng Valley, has more consistent fabrics (Table 4) than the former, which is located near the Hayisa Drumlin within the Hayisa Lopolith. In the HY, there are different fabric parameters on the top, the ice-lee side and stoss side of the drumlin (see Tables 2 and 3). The development of till fabrics is controlled by many factors and they contain information on sedimentary processes during, after and before deposition (Glen et al., 1957). The tills in all profiles studied were deposited during the Last Glacial and predominately formed during glacier
This work was supported by Grants from the NSFC (40301004, 40241011, 40271016), the LICCRE Grant BX2001-02, the 100 Talents project of the CAS, 2005, and Tianshan Glacier Observatory. We are grateful to Professor Lewis Owen for improvement of this paper. The comments from Professor Doug Benn and another anonymous referee are acknowledged. References Benn, D.I., 1994. Fabric shape and the interpretation of sedimentary fabric data. Journal of Sedimentary Research A 64, 910–915. Benn, D.I., Ringrose, T.J., 2001. Random variation of fabric eigenvalues: implications for the use of a-axis fabric data to differentiate till facies. Earth Surface Processes and Landforms 26 (3), 295–306.
ARTICLE IN PRESS D. Li et al. / Quaternary International 154–155 (2006) 19–25 Bennett, M.R., Waller, R.I., Glasser, N.F., Hambrey, M.J., Huddart, D., 1999. Glacigenic clast fabrics: genetic fingerprint or wishful thinking? Journal of Quaternary Science 14, 125–135. Cui, Z.J., 1981. Types and features of the moraines and till deposits in the Urumqi Riverhead, Tian Shan. Journal of Glaciology and Geocryology 3 (supp.), 36–48 (in Chinese with English summary). Cui, Z.J., Xiong, H.G., 1989. The sedimentary facies model and characteristics of mountain glacier. Quaternary Sciences 3, 254–267 (in Chinese with English summary). Cui, Z.J., Xiong, H.G., Liu, G.N., Zhu, C., Yi, C.L., 1998. Landform Processes and Sedimentary Features on the Cryosphere of the Central Tianshan. ShiJiazhang. Hebei Science and Technology Publishing House, pp. 1–305 (in Chinese with English summary). Deng, Y.X., 1995. Process of accumulation and characteristics of glacial debris flow deposits transformed by moraine. Acta Sedimentologica Sinica 13 (4), 37–48 (in Chinese with English summary). Derbyshire, E., 1984. Till properties and glacier regime in parts of high Asia: Karakoram and Tian Shan. In: White, R.O. (Ed.), The Evolution of the East Asian Environment. University of Hong Kong Press, Hong Kong, pp. 84–110. Dowdeswell, J.A., Hambrey, M.J., Wu, R.T., 1985. A comparison of clast fabric and shape in late Precambrian and modern glacigenic sediments. Journal of Sedimentary Petrology 55 (5), 691–704. Dowdeswell, J.A., Sharp, M.J., 1986. Characterization of gravel fabric in modern terrestrial glacigenic sediments. Sedimentology 33, 699–710. Feng, Z.D., Qin, D.H., 1984. Sedimentary types and processes of end moraines in the Urumqi Riverhead of the Tian Shan since the last glaciation. Journal of Glaciology and Geocryology 6 (3), 39–49 (in Chinese with English summary). Glen, J.W., Donner, J.J., West, R.G., 1957. On the mechanism by which stones in till become oriented. American Journal of Science 255, 194–205. Henriksen, M., Mangerud, J., Maslenikova, O., Matiouchkov, A., Tveranger, J., 2001. Weichselian stratigraphy and glaciotectonic deformation along the lower Pechora, Arctic Russia. Global and Planetary Change 31, 297–319. Hooyer, T.S., Iverson, N.R., 2000. Clast-fabric development in a shearing granular material: implications for subglacial till and fault gouge. Geological Society of America Bulletin 112 (5), 683–692. Klein, E., Davis D., 2002. Surface Sample Bias and Clast Fabric Interpretation Based on Till, Ditch Plains, Long Island. Long Island Geologists, State University of New York. Li, J.J., Zhou, S.Z., 1984. Morphology and surface features of till clasts. Journal of Glaciology and Geocryology 6 (3), 27–29 (in Chinese with English summary). Li, S.D., Cui, Z.J., Wang, J.T., Zhang, Z.Q., 1981. Petrology and morphology of pebbles of glaciogenic, glacioalluvial and alluvial deposits, the Urumqi Riverhead, Tian Shan. Journal of Glaciology and Geocryology 3 (Suppl.), S78–S91 (in Chinese with English summary).
25
Mark, D.M., 1973. Analysis of axial orientation data, including till fabrics. Geological Society of America Bulletin 84, 1369–1374. Millar, S.W.S., Nelson, F.E., 2001a. Sampling-surface orientation and clast macrofabric in Periglacial colluvium. Earth Surface Processes and Landforms 26, 523–529. Millar, S.W.S., Nelson, F.E., 2001b. Clast fabric in relict Periglacial colluvium, Salamanca re-entrant, southwestern New York, USA. Geografiska Annaler 83A, 145–156. Mills, H.H., 1977. Basal till fabrics of modern alpine glaciers. Geological Society of America Bulletin 88, 824–828. Sakai, T., Yokokawa, M., Kubo, Y., Endo, N., Masuda, F., 2002. Grain fabric of experimental gravity flow deposits. Sedimentary Geology 154, 1–10. Sneed, E.D., Folk, R.L., 1958. Gravels in the lower Colorado River, Texas, a study in particle morphogenesis. Journal of Geology 66, 114–150. Wang, J.T., 1981. Ancient glaciers at the head of Urumqi River, Tian Shan. Journal of Glaciology and Geocryology 3 (supp.), 56–63 (in Chinese with English summary). Wang, J.T., Zhang, Z.Q., 1981. Glacier sediment at the head of Urumqi river, Tian Shan. Journal of Glaciology and Geocryology 3 (Suppl), S49–S56 (in Chinese with English summary). Wenk, H.R. (Ed.), 1985. Preferred Orientation in Deformed Metals and Rocks: An Introduction to Modern Texture Analysis. Academic press, Inc., New York. Woodcock, N.H., 1977. Specification of fabric shapes using an eigenvalue method. Geological Society of America Bulletin 88, 1231–1236. Xiong, H.G., 1991. Comparative research of till with debris flow deposits and glacial-geomorphic process of the Tianshan Mountains. Ph.D. Dissertation, Peking University, pp.118–128 (in Chinese with English summary). Yi, C.L., Cui, Z.J., 2001. Subglacial deformation: evidence from microfabric studies of particles and voids in till from the upper Urumqi River valleys, Tian Shan, China. Journal of Glaciology 47 (159), 607–612. Yi, C.L., Jiao, K.X., Liu, K.X., He, Y.Q., Ye, Y.G., 2002. ESR dating of the sediments of the Last Glaciation at the source area of the Urumqi River, Tian Shan Mountains, China. Quaternary International 97–98, 141–146. Yi, C.L., Liu, K.X., Cui, Z.J., Jiao, K.X., Yao, T.D., He, Y.Q., 2004. AMS radiocarbon dating of Late Quaternary glacial landforms, source of the Urumqi River, Tien Shan—a pilot study of 14C dating on inorganic carbon. Quaternary International 121 (1), 99–107. Zhou, B.G., Li, G., Wang, J.G., Pan, L.L., 1992. Genesis analysis of Quaternary clay–gravel deposits in the Huangshan, East of China. Journal of Glaciology and Geocryology 14 (4), 360–365 (in Chinese with English summary).