Geomorphology 49 (2002) 231 – 254
www.elsevier.com/locate/geomorph
Surface sediment characteristics and tower karst dissolution,
Guilin, southern China
Tao Tang*
Department of Geography and Planning, State University of New York College at Buffalo, 1300 Elmwood Avenue, Buffalo,
NY 14222-1095, USA
Received 9 November 2001; received in revised form 9 May 2002; accepted 17 May 2002
Abstract
Dissolution of extensive outcrops of limestone and dolostone in humid tropical and subtropical southern China produced
numerous caves and residual hills that are referred as tower karst. This study identifies and relates the physical and chemical
characteristics of the surface sediment with the limestone bedrock in Guilin to assess the influence of the limestone dissolution
process on sediment composition.The results of this study indicated that (i) both limestone and dolostone of the region are very
pure (99.5% and 98.5% of CaCO3 and MgCO3, respectively); (ii) the material composition of limestone and dolostone is
different from that of soil and sediment of the region: constituents of surface sediments are highly related with the clastic
sedimentary rocks, such as the mudstone, but show negative correlation with limestone and dolostone; (iii) the limestone
formations are highly resistant to physical weathering and disintegration; their durability versus physical weathering and their
high susceptibility to chemical dissolution account for why residual towers can form and persist; (iv) a dual-zone environmental
structure exists vertically downward from the surface in Guilin: the zone of unconsolidated clastic sediments that is
predominantly acidic, and the zone of karstified limestone that is predominantly basic. The evidence suggests that the
environment and processes differ in these two zones. The chemical dissolution of limestone that formed tower karst of the
region is not mainly responsible for the accumulation of clastic sediment on the surface.
D 2002 Elsevier Science B.V. All rights reserved.
Keywords: Limestone dissolution; Tower karst; Surface sediment; Rock hardness; X-ray fluorescence; X-ray diffraction
1. Introduction
In south China, karst terrains have been heavily
dissected because of neo-tectonic uplift. Previous
research identified two groups of positive karst features: fenglin or ‘‘peak forest’’ and fengcong or ‘‘peak
cluster’’ (Yuan, 1981, 1986; Zhang, 1981; Zhu, 1988;
*
Tel.: +1-716-878-4138; fax: +1-716-878-4009.
E-mail address: [email protected] (T. Tang).
Zhu et al., 1988). The former are individual, isolated
residual hills rising from flood or corrosion plains.
The latter comprise a group of residual hills emerging
from a common bedrock basement and often incorporating closed depressions between the clusters of
peaks. These are generally considered tower karst by
western observers. The tower karst in Guilin, Guangxi
Zhuang Autonomous Region is regarded as one of the
most beautiful and spectacular landform assemblages
in China as well as in the world (Sweeting, 1995).
0169-555X/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 1 6 9 - 5 5 5 X ( 0 2 ) 0 0 1 8 8 - 5
232
T. Tang / Geomorphology 49 (2002) 231–254
The objectives of the current study are (i) to
identify the physical and chemical characteristics of
the surface sediment in Guilin and (ii) to relate the
sediment properties to that of the limestone bedrock of
the region to seek the evidence of the influence of
limestone dissolution process on the sediment concentration. The major focus of this research is to
analyze and compare the material composition, physical and chemical properties of surface sediments with
the characteristics of limestone, dolostone and clastic
sedimentary rock of the region.
2. The study area
Guilin is located in the subtropical monsoon climate region (Huang et al., 1988) of southern China.
Affected by both the SW maritime monsoon from the
Indian Ocean and the SE maritime monsoon from the
western Pacific Ocean, the average annual precipitation in Guilin is 1873.6 mm and the annual average
temperature is 18.8 jC (Yuan, 1992). The exposed
bedrock strata in Guilin mainly represent the Paleozoic, Mesozoic and Cenozoic systems. Of these, the
major outcrops of the region are carbonate rocks from
the Middle Devonian Donggangling Formation (D2d)
to the lower Carboniferous Yanguang Formation
(C1y), which total 4622 m in thickness (Deng et al.,
1988) (Fig. 1). The upper Triassic and upper Cretaceous strata of the Mesozoic are represented by a
continental depositional environment and accumulative conglomerates and silty mudstones. The mudstones are distributed sporadically in the center of the
Guilin syncline. Age of the Cretaceous mudstones
determined by the Ru – Sr method is 145 F 5 Ma
(Weng, 1987), which can be used chronologically as
a reference marker for the other bedrock formations in
Guilin. Quaternary deposits are mainly alluvial and
fluvial sediments, plus in situ weathering residuum
derived from the mudstones.
Outcrops of early Paleozoic age, mainly representing the Cambrian and the Ordovician systems, occur in
the SW and NW corners as well as along the eastern
edge of the region (Fig. 1). The strata of the Cambrian
system are shallow-sea facies sandstones and shales of
the Qinxi Group (Eq) and Bianxi Group (Eb). Bedrock
of Ordovician age consists of shallow-sea facies strata
of interbedded sandstones, shales and limestones. For-
mations of the Ordovician are divided into the White
Cave Group of clayey limestones and dolostones
(O1b), the Huangai Group (O1h) of sandy shales, the
Shengping group of anthrinoid shales or silty shales
(O1s) of lower Ordovician and the mid- to upper
Ordovician clayey sandstones and shales (O2 + 3).
The main geological structure in Guilin is a NNE –
SSW-trending synclinorium. Superimposed on which
are a N – S-trending thrust fault and WNW – ESEtrending transverse as well as shear fault systems
(Deng et al., 1988). The Guilin area is surrounded
by mountains of noncarbonate rocks. The Li River,
which is the major drainage system of the region for
both surface runoff and ground runoff (Huang et al.,
1988), enters the Guilin area from the adjacent sandstone mountainous area of Yuechengling to the north.
The limestone towers become most numerous around
Guilin City. The Li River below the city flows southeastward to Daxu where it turns south and cuts
through a gorge in the Devonian limestone over 60
km in length. The karst ground water discharges into
the Li River and its tributaries ultimately in the form
of either surface springs or submerged springs. The
underground (karst) runoff constitutes the base flow of
the river system and accounts for 40% to 50% of the
total regional runoff.
3. Methods and approaches
Three types of tower karst combinations can be
identified in Guilin: (i) peak-forest (fenglin) dominated areas, which mainly occur in the central part of
the Guilin Syncline; (ii) peak-cluster (fengcong)
dominated areas, which mainly occupy the flanks of
the syncline; (iii) areas of mixed fengcong and fenglin, which occur particularly around Yangshuo, in the
southern part of the Guilin district. In order to incorporate the varying karst types and distributions, three
sample areas were selected for detailed surveys and
data collection. Each site measures 2 – 6 km2. These
are (i) the Experimental Station of the Institute of
Karst Geology and Hydrology (EXS), which is representative of fengcong and is located in the NE part
of the region (Figs. 1 and 2); (ii) the Putao Township
(GPT), which is a representative of fenglin and
located in the central part of the region (Figs. 1 and
3); (iii) Yangshuo (YS), which is an area of mixed
T. Tang / Geomorphology 49 (2002) 231–254
Fig. 1. Bedrock rock distribution and sampling areas in Guilin, China.
233
234
T. Tang / Geomorphology 49 (2002) 231–254
Fig. 2. Fengcong in the EXS area.
fengcong and fenglin and located in the town of
Yangshuo, in the southern part of the region (Figs. 1
and 4). Detailed sampling of water, bedrock and soil
and sediment were conducted within the selected
sample areas unless a particular concerned item was
not available in the area.
Fig. 3. Fenglin in the GPT area.
T. Tang / Geomorphology 49 (2002) 231–254
235
Fig. 4. Mixed fengcong and fenglin in the YS area.
3.1. Analyses of surface sediment and bedrock
properties
The main objective of investigating the sediment
and soil in the sampling areas is to identify the
relationship between slope forms, processes and surface sediments and soils in the region. The field
investigation revealed that there is no soil and sediment cover on the upper and middle slopes of towers,
simply because the slopes are too steep for regolith to
accumulate (Tang and Day, 2000). Soils and sediments typically occur at the bases of towers and on the
flood plain or in the basins of fengcong terrain.
Sediments and soils were only sampled at the bases
of towers where they are preserved with their natural
conditions.
Soils and sediments are in large the products of
interactions between the atmosphere and bedrock
formations. More specifically, the process of sediment
formation is called ‘‘geogenesis’’, and the development of soil horizons in the sediment is called ‘‘pedogenesis’’ (Van Wambeke, 1992). From a geomorphic
perspective, although soils have pedogenic characteristics that distinguish them from their parent materials,
sediments and soils may be seen as having common
origin, in that both are largely products of geomorphic
processes. The analysis of the soil and sediment in a
region can reveal what kind of geomorphic processes
was responsible for their formation, although the
processes that are responsible for the formation of
sedimentary landforms may not be the major processes that produce residual landforms.
A total of 10 soil and sediment profiles were
excavated in the three sampling areas. All of them
were located in the basal areas of the karst towers,
where soil and sediment appeared undisturbed. Seven
profiles were excavated in the EXS area, two in the
GPT area and one in the YS area (Fig. 5). Content of
soil organic matter and carbonate was determined by
the loss on ignition method (Gee and Bauder, 1986),
and soil pH was measured by a pH meter (Accumet
925 pH/ion meter with accumet gel-filled polymer
body electrodes, made by Fisher Scientific) with
standard buffers of pH 4 and pH 7.
Textural analysis was performed for sediment and
soil samples following the method of Gee and Bauder
(1986). Clay, silt and sand fractions were measured by
the hydrometer method. The standard particle size
classes used in this study are sand (2 – 0.0625 mm),
silt (0.0625 –0.002 mm) and clay ( < 0.002 mm). The
sand fraction was collected by wet sieving (sieve mesh
#230) from the hydrometer cylinders. The collection
was then dried and dry-sieved with 1U interval (sieve
mesh #10, 18, 35, 60, 120 and 230). Textural analysis
236
T. Tang / Geomorphology 49 (2002) 231–254
Fig. 5. Particle analysis of surface sediment.
T. Tang / Geomorphology 49 (2002) 231–254
of soil and sediment may elucidate their potential
sources, the depositional environment.
Chemical purity is one of the major properties of
limestone and dolostone that may be directly related
to dissolution processes and landform morphology. In
general, if all other conditions remain constant, the
purer the limestone, the higher the solubility and the
more susceptible to dissolution (Ford and Williams,
1989). Purity and solubility of limestone and dolostone samples were tested in this study by determining
the percentage content by weight of their insoluble
residues.
The objective of this part of the investigation was
to gain a preliminary understanding of the purity of
the limestone on which the tower karst of the region
was developed, and to make comparisons with that of
dolostone in the region. All the tests were carried out
using 3N hydrochloric acid (HCl) in laboratory conditions. A total of four limestone and one dolostone
samples were tested (Table 1). Among them, one was
from the Donggangling Formation (Rock#1) and three
were from Rong County Formation (Rock#2, #3,
#4). The dolostone was from Yangguan Formation
(Rock#5).
Geomorphologists have long attempted implicitly
to associate compressive and shear strengths of bedrock with its susceptibility to weathering and erosion,
although the relationship appears not to be straightforward. One of the major difficulties of this approach is
to measure compressive strength and shear strength of
bedrock formations in the field. Day and Goudie
Table 1
Summary of insoluble contents of carbonate rocks in Guilin
Sample
number
Lithology
Location
Total
weight
(g)
Residue
weight
(g)
Percent of
residue
content
(% by
weight)
Rock#1
limestone
(D2d)
limestone
(D3r)
limestone
(D3r)
limestone
(D3r)
dolostone
(C2)
YS
100.00
0.53
0.53
EXS
100.00
0.64
0.64
EXS
100.00
0.62
0.62
GPT
100.01
0.42
0.42
LinGui
County
104.61
1.95
1.86
Rock#2
Rock#3
Rock#4
Rock#6
237
(1977) introduced a method of in situ testing of rock
hardness or compressive strength using the Schmidt
Concrete Test Hammer. Day (1980, 1981, 1982)
examined the role of rock hardness in carbonate rock
weathering and karst terrain development.
Hardness of the carbonate rocks and other bedrock
of the region was tested by a Schmidt Hammer. The
Schmidt Hammer measures the distance of rebound
(R-value) of a controlled impact on a rock surface.
Because the recovery distance depends on the hardness of surface and hardness is related to compressive
strength, the R-value gives a relative measure of
surface hardness or compressive strength. The R-value
ranges from 10 to 100 and is shown on a scale on the
side of the instrument (Day and Goudie, 1977).
Three different types of limestone surface on the
towers were tested at each of the locations: freshbroken surfaces, weathered surfaces and lichen-covered surfaces. A fresh, broken surface is one that was
exposed by the investigator in order to obtain a
measure of the original compressive strength of the
limestone. A weathered surface is a natural surface on
a slope that had been weathered but which otherwise
appeared intact. No evidence is available of case
hardening or formation of hard weathering crusts on
the limestone in Guilin. Obvious weathering effects
on the slopes of towers are very limited in depth and
generally do not exceed 3 mm deep from the surface.
A lichen-covered surface is one that has at least 30%
of the total area is covered with lichens. This is the
case on roughly 40– 50% of the slope surfaces surveyed on the towers. The major species of lichens
growing on the limestone surface in Guilin are the
Rhizocarpon and Iecanora. Biological weathering of
limestone by lichens is also limited to the upper layer
beneath the surface with a range of < 5 mm.
Six to eight impacts of the Schmidt Hammer were
conducted for each of the three types of rock surface
at each of the locations tested. Readings of each of the
impacts were recorded in the field and the mean Rvalues were calculated after returning from the field.
Tests of limestone hardness were only conducted in
two of the three sampling areas, EXS (fengcong) and
GPT (fenglin). Because of instrument malfunction in
the latter period of the fieldwork, no test results were
obtained from the YS area. A total of 26 tests were
conducted: 18 in the EXS area and 8 in the GPT area
(Table 2).
238
T. Tang / Geomorphology 49 (2002) 231–254
Table 2
Compressive strength test of bedrocks in the field in Guilin
Test
number
Sampling Average
Weathered
area
R-value:
surface
fresh-broken
surface
Compressive strength test of limestone
HD#1
EXS
46.9
HD#2
EXS
39.1
HD#3
EXS
39.3
HD#4
EXS
40.7
HD#5
EXS
50.2
HD#6
EXS
44.1
HD#7
EXS
53.0
HD#8
EXS
52.1
HD#9
EXS
56.5
HD#10
EXS
42.3
HD#11
EXS
39.5
HD#12
EXS
61.3
HD#13
EXS
45.7
HD#14
EXS
45.8
HD#15
EXS
47.4
HD#16
EXS
45.3
HD#17
EXS
43.0
HD#18
EXS
41.6
mean: 46.3
maximum:
61.3
minimum:
39.1
in the field
33.1
30.0
32.0
32.2
36.1
36.0
35.8
36.8
39.4
37.1
35.3
37.7
34.7
43.4
35.2
31.5
36.0
36.3
mean: 35.5
maximum:
43.4
minimum:
30.0
HD#19
HD#20
HD#21
HD#22
HD#23
HD#24
HD#25
HD#26
34.3
35.5
39.6
37.4
34.5
38.3
34.5
39.8
mean:
36.7
maximum:
39.8
minimum:
34.3
GPT
GPT
GPT
GPT
GPT
GPT
GPT
GPT
41.5
48.8
57.3
55.3
47.6
48.0
40.9
42.0
mean:
47.7
maximum:
57.3
minimum:
40.9
Lichen
covered
surface
19.0
26.2
22.8
19.3
26.5
24.0
28.0
32.0
22.5
15.8
32.0
30.7
mean: 24.9
maximum:
32.0
minimum:
15.8
19.3
22.0
16.0
18.0
mean:
18.8
maximum:
22.0
minimum:
16.0
Compressive strength test of dolostone, mudstone (redbed) and
conglomerate in the field
Dolostone:
HD#27
LinGui 39.7
21.9
County
HD#28
25.0
14.5
HD#29
26.0
14.0
HD#30
25.7
14.5
HD#31
31.2
14.8
HD#32
27.5
16.0
Table 2 (continued )
Test
number
Sampling Average
Weathered
area
R-value:
surface
fresh-broken
surface
Lichen
covered
surface
Compressive strength test of dolostone, mudstone (redbed) and
conglomerate in the field
mean: 29.2 mean: 16.0
maximum: maximum:
39.7
21.9
minimum: minimum:
25.0
14.0
Mudstone:
HD#33
Guilin
mean: 21.8
Airport
Conglomerate:
HD#34
cementing
material
(mean): 25.3
major content
(mean): 36.8
3.2. Elemental composition analysis by XRF
X-ray fluorescence (XRF) spectrometry was applied in this study to analyze the elemental composition of rock and sediment samples. XRF is a physical
analysis that is used in quantitative identification of the
elemental content of the earth surface materials. It
differs from chemical analyses of elemental concentrations and does not differentiate between the portions
of elements that are acid soluble or acid insoluble.
More specifically, atomic spectrometry as an analytical
technique only measures the acid-soluble portion of
the elemental content in a specimen, while XRF
measures the total value of elemental concentrations
in the specimen, identifying the elements present and
their percent in weight in the sample. This avoids
issues of whether the concentration is acid soluble or
acid insoluble and makes it convenient in the comparison of rock and sediment samples.
XRF identifies an element by measurement of its
characteristic X-ray emission wavelength or energy.
The method allows quantification of a given element
by first measuring the emitted characteristic line
intensity and then relating that intensity to elemental
concentration (Jenkins et al., 1981; Tertian and
Claisse, 1982; Williams, 1987). The XRF unit used
in this study is a Phillips PW 1410 X-ray Spectrometer located in the Center for Great Lakes Studies at
T. Tang / Geomorphology 49 (2002) 231–254
the University of Wisconsin-Milwaukee. The analytical method and procedure followed was from Jenkins
(1988).
The purpose of this analysis was to identify the
similarities and differences in terms of elemental
composition between the different rocks in the study
region. Data from this analysis was also used in
comparison with the results of soil and sediment
analysis to trace any evidence of processes that may
be responsible for landform development. Samples of
the Carboniferous dolostone (C1y) were collected
from Lingui County in suburban Guilin, and those
of the Cretaceous mudstone (K2) were collected from
Guilin Airport for comparative purposes.
The total percentage concentration was measured
for eight major elements for both the surface sediment
and bedrock samples. These are silicon (Si), aluminum (Al), calcium (Ca), magnesium (Mg), iron (Fe),
titanium (Ti), potassium (K) and sodium (Na). Following the procedure, all of the samples were pretreated by loss of ignition in order to eliminate the
carbonate content and organic matter. Because CaCO3
and MgCO3 are the major components of limestone
and dolostone, the weight of carbonate content by
loss of ignition was added with Ca and Mg in the
calculation of percentage concentration for all the
samples.
3.3. Mineral composition analysis by XRD
X-ray diffraction spectrometry (XRD) was applied
in this study to analyze the mineral composition of
rock and sediment samples. Each XRD pattern is
characteristic of a certain mineral because each mineral is made up of a unique combination and structure
of atoms (Allen and Hajek, 1989). In practice, the
intensities (or peaks) of diffraction are recorded and
plotted against a horizontal scale in degrees of 2h,
which is the angle of the detector rotation in order to
catch the diffracted X-ray (Moore and Reynolds,
1989, 1997).
XRD is used to identify the mineral content of the
materials both qualitatively and quantitatively. Only
qualitative analyses of mineral contents of the surface
sediment and bedrocks of the region by XRD were
conducted in this research. Two major sample preparation techniques were presented for XRD analysis
(Brindley and Brown, 1980; Moore and Reynolds,
239
1989, 1997): (i) the powder method and (ii) the
glass-slide method. The powder method was applied
to analyze the rock samples and silt fractions of the soil
and sediment in this study in order to ensure a random
orientation of the grains (Bish and Reynolds, 1989).
The glass-slide method was applied to the clay fractions of soil and sediment to achieve a perfect orientation of the clay mineral flakes. (Moore and Reynolds,
1989, 1997). The instrument used for the XRD analysis in this study was a Siemens Analytical X-ray
System in the Center for Great Lakes Studies at the
University of Wisconsin-Milwaukee.
A total of 18 soil samples from five soil profiles
(EXS#1, EXS#4, GPT#1, GPT#2, and YS#1) were
selected for XRD analysis to represent the three sampling areas. The silt and clay fractions of the soil
samples were separated and extracted, and then were
subjected to analysis by XRD spectrometry. The procedure for qualitative identification of mineral content
begins by searching for a mineral whose diffraction
pattern will explain the strongest peak or peaks, then
confirming the choice by finding the positions of
weaker peaks for the same mineral. Once a set of peaks
is confirmed as belonging to a mineral, these peaks are
eliminated from further consideration. This procedure
is repeated until all the peaks are identified. The array
of minerals recognized in this process constitutes the
mineral composition of the sample. By ranking, the
largest mineral content is correspondent to the strongest
peaks, and the smallest mineral content is correspondent to the lowest peaks. Each of the natural minerals has
its own pattern of XRD peaks referenced by the 2h
scale. The standard patterns of different minerals were
established in mineralogical studies (Bish and Reynolds, 1989; Moore and Reynolds, 1989, 1997; Bish
and Post, 1993).
4. Results
4.1. Characteristics of bedrock
The results of the insoluble residue determination
of limestone and dolostone are shown in Table 1. The
result indicates that the limestone of the region is very
pure and contains more than 99% soluble CaCO3 and
MgCO3. Dolostone of the region is also very pure,
containing 98.2% CaCO3 and MgCO3.
240
T. Tang / Geomorphology 49 (2002) 231–254
Mean R-values of limestone hardness at different
locations in the sampling areas are shown in Table 2.
The mean R-value of fresh, broken surfaces in the EXS
area was 46.3, ranging from 39.1 to 61.3. In the GPT
area, the mean R-value was 47.7, ranging from 40.9 to
57.3. By contrast, weathered surfaces had a mean Rvalue of 35.5 with a range from 30.0 to 43.4 in the EXS
area and a mean of 36.7 in the GPT area, ranging from
34.3 to 39.8. The lichen-covered surfaces in both areas
have even lower R-values with means of 18.8 in the
GPT area and 24.9 in the EXS area.
The results indicate that (i) no significant difference
was found between R-values in the fengcong (EXS)
and fenglin (GPT) areas (Table 2); the limestone of the
two sites belongs to the same formation, the Rong
County Formation of the Devonian; (ii) the mean Rvalues decrease from fresh, broken surfaces to weathered surfaces and from weathered surfaces to lichencovered surfaces; (iii) although weathering on the
slopes of towers decreases the surface hardness, the
significance of weathering penetration by the atmosphere and growing lichen is limited and reaches a
maximum of up to 5 mm below the surface; (iv) ranking
the different lithologies of the region by compressive
strength, limestone is the hardest (R = 40 – 50), dolostone is second (R = 25 – 40), conglomerate is third
(R = 25 –37) and the remnants of mudstone are the
softest (R = 21.8).
4.2. Characteristics of surface sediment and soil
The results of particle size analysis show that both
the clay and silt contents of soil and sediment are very
high across all the profiles excavated (Fig. 5). The
mean clay content is 61%, and the mean silt content is
33%. By contrast, the mean sand content is only 6%. A
clear trend is presented in which the clay content
increases and the silt and sand contents decrease as
the depth of soil increases (Fig. 5). In general, clay
content of soils in the study area increases from 40% to
50% in the surface horizon (0– 30 cm) to 80% to 90%
down to 100 – 150 cm in depth. Silt content decreases
from 35% to 50% on the surface to 10% to 30% at the
bottom of the soil profiles, depending on the location.
Sand fractions diminish downward through profiles.
The fenglin of the GPT area has the lowest mean clay
content (54%) and the highest mean sand content
(13%) of the three sampling areas.
The chemical test results indicate that soil organic
matter (OM) decreases as the soil depth increases
(Table 3). Carbonate contents of soils are very low
Table 3
Chemical properties of the surface sediment and soil in the sampling
areas of tower karst in Guilin
Sampling area
and profile number
Depth
(cm)
Lab ID
OM
(%)
Carbonate
(%)
pH
EXS#1
EXS#1
EXS#1
EXS#1
0 – 15
15 – 60
60 – 110
110 – 180
1
2
3
4
13.9
8.5
6.6
7.2
4.2
4.3
4.5
4.6
6.6
6.2
5.9
5.9
EXS#2
EXS#2
EXS#2
EXS#2
0 – 10
10 – 25
25 – 45
45 – 100
5
6
7
8
7.1
6.1
5.5
4.9
7.9
8.2
7.1
8.3
5.5
4.7
4.5
4.8
EXS#3
EXS#3
EXS#3
EXS#3
0 – 15
15 – 40
40 – 55
55 – 100
9
10
11
12
11.6
10.7
9.2
8.0
12.5
9.9
10.7
11.5
5.4
5.1
5.3
5.4
EXS#4
EXS#4
EXS#4
0 – 10
10 – 35
35 – 70
13
14
15
15.0
12.3
11.5
4.5
4.6
4.5
5.4
5.3
5.6
EXS#5
EXS#5
EXS#5
0 – 15
15 – 40
40 – 70
16
17
18
14.2
9.8
8.1
12.8
10.5
10.4
6.5
5.9
6.2
EXS#6
EXS#6
EXS#6
EXS#6
0–8
8 – 25
25 – 55
55 – 110
19
20
21
22
11.7
7.2
6.1
6.2
8.9
9.6
9.5
6.6
4.5
4.3
4.4
4.5
EXS#7
EXS#7
EXS#7
EXS#7
0–6
6 – 30
30 – 65
65 – 110
34
35
36
37
9.6
9.5
6.3
5.4
4.4
4.4
9.6
13.2
4.9
4.5
4.5
4.9
GPT#1
GPT#1
GPT#1
GPT#1
0 – 15
15 – 30
30 – 60
60 – 120
26
27
28
29
9.9
8.2
6.7
6.4
5.7
4.5
5.2
6.4
6.2
6.1
6.5
6.4
GPT#2
GPT#2
GPT#2
0 – 20
20 – 50
50 – 100
23
24
25
7.6
6.3
5.5
5.9
4.4
4.7
6.6
6.7
6.3
YS#1
YS#1
YS#1
YS#1
0–8
8 – 20
20 – 70
70 – 110
30
31
32
33
11.4
9.7
8.3
9.7
5.4
5.9
6.3
5.3
6.4
6.1
6.1
6.0
T. Tang / Geomorphology 49 (2002) 231–254
241
two major formations of limestone (Rong County and
Dongangling) was found in the region in terms of Si,
Al, Fe, Ti, K and Na concentrations. However, the
concentration of MgCO3 in the Dongangling Formation (D2d, 12%) is much higher than that of the Rong
County Formation (D3r, 1 –3%).
The major proportions of chemical concentration in
the dolostone are CaCO3 and MgCO3, with the CaCO3
occupying a rather higher percentage. The mean
CaCO3 content is 55.44%, and the mean MgCO3
content is 44.44%. Similar to the limestone formations,
no Ti was found in the dolostone either. Chemical
contents of all the other five elements in the dolostone
are lower than those in the limestone by about one
order of magnitude, ranging from 0.001% to 0.003%.
By contrast, the major elemental content in Cretaceous mudstone (K2) is Si, which accounts for a mean
of 67.4%. The mean concentration of Al is 11.3% and
that of Fe is 10.1%. Ti, K and Na in the mudstone
account for 0.83%, 1.39% and 0.45%, respectively.
The CaCO3 content of the mudstone is very low, with a
mean of only 0.6%. The mean MgCO3 content is
7.9%. This may result from a relatively high Mg
concentration.
and range from 4% to 13%. In general, carbonate
content decreases with increasing soil depth. All soil
pH values are below 7, ranging from 4 to 6.5. Soil pH
decreases as the soil profile depth increases. Overall,
the pH trend in the soil environment is weakly acidic to
acidic from the surface downward.
4.3. Elemental and mineral compositions of bedrock
Total elemental content of the limestone, the dominant bedrock of the region and the two other major
outcrops of the region, dolostone and mudstone, was
analyzed by XRF spectrometry. The results are shown
in Table 4. The dominant chemical component of
limestone in the region is CaCO3, which accounts
for a mean of 95.1% of the samples, ranging from
86.8% to 98.2%. Mean MgCO3 content of the limestone is 3.74% with a range from 0.39% to 12.23%.
Silicon content of limestone samples is very low. The
mean Si content is 0.167% with a range from 0.130%
to 0.241%. No Ti was found in the limestone. The
mean concentration of Al is 0.025% and that of Fe is
0.032%. The mean concentrations of K and Na are
both 0.011%. No significant difference between the
Table 4
Elemental content of bedrocks in Guilin by X-ray fluorescence (XRF) spectrometry
Sample
number
Location Rock Type
Rock#1
YS
Rock#2
EXS
Rock#3
EXS
Rock#4
GPT
Rock#5
EXS
Rock#6
Age
Percentage of elemental content
Si
LinGui
County
Rock#7 LinGui
County
Rock#8 Guilin
Airport
Rock#9 Guilin
Airport
Rock#10 Guilin
Airport
limestone, Dongangling
Formation
limestone, Rong
County Formation
limestone, Rong
County Formation
limestone, Rong
County Formation
limestone, Rong
County Formation
dolomite, Yanguan
Formation
dolomite, Yanguan
Formation
mudstone, Shale
(Red Bed)
mudstone, Shale
(Red Bed)
mudstone, Shale
(Red Bed)
Al
CaCO3
MgCO3
Fe
Ti
K
Na
Devonian (D2d)
0.130
0.009
86.767
12.231
0.015
0.000
0.005
0.001
Devonian (D3r)
0.130
0.009
97.328
1.702
0.015
0.000
0.005
0.000
Devonian (D3r)
0.130
0.007
98.158
0.811
0.018
0.000
0.006
0.000
Devonian (D3r)
0.206
0.015
98.221
0.389
0.024
0.000
0.008
0.024
Devonian (D3r)
0.241
0.084
94.777
3.582
0.088
0.000
0.033
0.029
Carboniferous (C1y)
0.000
0.000
55.902
44.004
0.001
0.000
0.000
0.003
Carboniferous (C1y)
0.002
0.001
54.971
44.872
0.002
0.000
0.001
0.003
Cretaceous (K2)
66.729
11.432
0.679
7.996
10.272
0.818
1.484
0.590
Cretaceous (K2)
67.663
11.215
0.571
7.878
10.093
0.822
1.346
0.411
Cretaceous (K2)
67.799
11.256
0.564
7.780
10.062
0.840
1.344
0.355
242
T. Tang / Geomorphology 49 (2002) 231–254
Table 5
Mineral content of rock samples identified by x-ray diffraction (‘‘yes’’ indicates a positive identification)
Sample
number
Rock type
Rock#1
Rock#2
Rock#3
Rock#4
Rock#5
Rock#6
Rock#7
Rock#8
Rock#9
Rock#10
limestone
limestone
limestone
limestone
limestone
dolostone
dolostone
mudstone
mudstone
mudstone
Minerals
Silicates
Iron oxides
Carbonates
Alpha-quartz
Feldspars
Hematite
Goethite
Calcite
Magnesite
Dolomite
Ankerite
very little
very little
very little
very little
very little
very little
very little
yes (high)
yes (high)
yes (high)
no
no
no
no
no
no
no
yes (low)
yes (low)
yes (low)
no
no
no
no
no
no
no
yes (low)
yes (low)
yes (low)
no
no
no
no
no
no
no
yes (low)
yes (low)
yes (low)
yes
yes
yes
yes
yes
yes
yes
no
no
no
yes
yes
yes
yes
yes
yes
yes
no
no
no
(low)
(low)
(low)
(low)
(low)
(low)
(low)
yes (low)
yes (low)
very little
very little
yes (high)
yes (high)
yes (high)
no
no
no
very
very
very
very
very
yes
yes
no
no
no
(high)
(high)
(high)
(high)
(high)
(low)
(low)
little
little
little
little
little
Sample
number
Clay minerals
Chlorite high-Fe
Chlorite low-Fe
Vermiculite
Smectite
Calcite
Carbonates
Magnesite
Dolomite
Ankerite
Rock#8
Rock#9
Rock#10
yes (high)
yes (high)
yes (high)
yes (high)
yes (high)
yes (high)
yes
yes
yes
yes (low)
yes (low)
yes (low)
no
no
no
no
no
no
no
no
no
no
no
no
The results of the mineral content of limestone,
dolostone and mudstone by XRD analysis are
shown in Table 5 and the X-ray diffractograms
are shown in Figs. 6 and 7, respectively. The
dominant mineral content of the limestone is calcite, with the minor constituent being magnetite.
Some limestone formations contain a small amount
of dolomite. The major mineral content of the
dolostone is dolomite, with the calcite as the
second largest content ranging from 20% to 30%.
The patterns of the X-ray diffractograms indicate
that all of the dolostone samples contain a small
amount of magnetite and ankerite. Only trace
amounts of alpha-quartz were found in the limestone and dolostone samples.
By contrast, the dominant mineral constituent in
the mudstone is alpha-quartz. Minor minerals include
feldspars and iron-rich oxide species of hematite and
goethite. No carbonate mineral was found in the
mudstone samples. The Cretaceous mudstones also
contain a large proportion of clay minerals (about
30%). XRD shows that the dominant clay mineral in
the mudstone is chlorite (Table 5, Fig. 7). Vermiculite
is present in the mudstone as a minor coexistent clay
mineral. A trace amount of smectite was also found in
the mudstone.
4.4. Elemental and mineral compositions of surface
sediment
The eight major elemental species in soil and
sediment that are the same as those tested in the
bedrock were analyzed by XRF spectrometry. The
results of percentage contents in the soil and sediment
are shown in Table 6. The major component of the soil
and sediment is Si, which accounts for a mean of 43%
ranging from 32% to 55%. The second highest elemental concentration in the soil and sediment is Al
with a mean of 21.7%. Iron has a mean content of
12.9%. Owing to its high solubility, the lowest elemental concentration is that of Na with a mean of
0.35%. Titanium is present in all the soil samples, at a
much higher concentration (1.6%) than in the Cretaceous (K2) mudstone samples measured. Concentrations of CaCO3, MgCO3 and K are minor as a result
of leaching; mean contents are 3.5%, 5.9% and 2.1%,
respectively. The pattern of elemental composition of
the soil and sediment is generally similar to that of
mudstone, but different from those of limestone and
dolostone in the region.
The results do not show a clear trend of changes in
Si content through soil and sediment profiles, but the
Si content of each of the profiles is roughly the same.
T. Tang / Geomorphology 49 (2002) 231–254
Fig. 6. X-ray diffractogram of limestone and dolostone.
243
244
T. Tang / Geomorphology 49 (2002) 231–254
Fig. 7. X-ray diffractogram of the mudstone samples.
T. Tang / Geomorphology 49 (2002) 231–254
245
Table 6
Elemental content of soil and sediment in Guilin by X-ray fluorescence (XRF) spectrometry
Sample
number
Sampling area
and profile number
Depth (cm)
1
2
3
4
EXS#1
EXS#1
EXS#1
EXS#1
5
6
7
8
Percentage of elemental content
Si
Al
CaCO3
MgCO3
0 – 15
15 – 60
60 – 110
110 – 180
41.978
45.113
47.788
43.724
20.234
21.489
20.776
22.878
4.116
2.915
2.799
2.983
EXS#2
EXS#2
EXS#2
EXS#2
0 – 10
10 – 25
25 – 45
45 – 70
52.823
53.386
51.163
49.963
16.847
17.346
18.928
19.722
9
10
11
12
EXS#3
EXS#3
EXS#3
EXS#3
0 – 15
15 – 40
40 – 55
55 – 100
32.116
31.843
32.111
32.404
13
14
15
EXS#4
EXS#4
EXS#4
0 – 10
10 – 35
35 – 70
16
17
18
EXS#5
EXS#5
EXS#5
19
20
21
22
Fe
Ti
K
Na
2.810
3.436
3.599
3.409
13.100
13.939
13.438
14.666
1.464
1.665
1.830
1.878
2.103
2.556
2.813
2.885
0.296
0.387
0.356
0.376
3.758
2.970
2.785
2.520
6.264
6.878
7.832
8.281
10.036
10.012
10.856
11.205
1.703
1.780
1.801
1.720
1.505
1.597
1.968
2.121
0.364
0.333
0.267
0.367
25.315
26.887
27.757
27.928
5.716
4.956
5.430
5.708
9.831
10.516
10.294
10.159
13.402
14.065
13.878
14.172
1.117
1.155
1.049
1.041
1.231
1.345
1.471
1.532
0.272
0.234
0.510
0.357
38.413
39.908
40.944
23.099
23.660
23.299
3.324
3.064
3.410
3.459
3.616
3.314
12.804
13.370
13.293
1.466
1.564
1.525
2.101
2.142
2.319
0.333
0.375
0.397
0 – 15
15 – 40
40 – 70
35.679
39.760
40.960
19.537
21.152
21.638
11.349
8.601
7.933
5.101
5.211
5.560
10.978
11.833
11.985
1.273
1.430
1.399
1.627
1.686
1.928
0.255
0.527
0.497
EXS#6
EXS#6
EXS#6
EXS#6
0–8
8 – 25
25 – 55
55 – 110
43.598
44.616
45.020
49.863
19.612
21.266
21.584
19.100
2.436
2.547
2.482
2.580
7.461
8.119
7.977
7.052
11.632
12.622
12.596
11.796
1.765
1.830
1.853
2.080
1.381
1.528
1.690
1.814
0.414
0.271
0.699
0.616
23
24
25
GPT#2
GPT#2
GPT#2
0 – 20
20 – 50
50 – 100
53.003
54.382
54.032
15.914
16.640
17.427
3.682
2.914
2.071
3.193
3.252
3.901
9.898
10.630
11.104
1.697
1.915
1.769
3.092
3.303
3.433
0.319
0.264
0.263
26
27
28
29
GPT#1
GPT#1
GPT#1
GPT#1
0 – 15
15 – 30
30 – 60
60 – 120
46.268
46.178
47.384
42.372
19.455
20.850
20.476
23.023
2.600
2.134
1.955
2.005
4.063
4.313
4.469
5.282
12.018
13.036
12.696
14.475
1.754
1.817
1.886
1.619
2.663
2.924
2.930
3.356
0.179
0.149
0.404
0.467
30
31
32
33
YS#1
YS#1
YS#1
YS#1
0–8
8 – 20
20 – 70
70 – 110
38.959
37.096
37.171
33.483
22.967
25.748
26.051
28.567
3.074
1.810
1.774
1.802
5.697
6.145
5.947
6.624
14.014
15.488
15.924
16.041
1.330
1.270
1.384
1.103
1.687
1.814
1.889
1.733
0.373
0.330
0.261
0.347
34
35
36
37
EXS#7
EXS#7
EXS#7
EXS#7
0–6
6 – 30
30 – 65
65 – 110
43.630
47.019
43.781
41.297
22.381
21.136
21.938
21.184
1.698
1.279
2.200
4.431
3.736
4.179
8.485
9.748
15.400
13.066
13.450
14.576
1.810
1.985
1.867
1.713
1.551
1.505
1.500
1.468
0.194
0.331
0.479
0.184
43.222
54.382
31.843
21.725
28.567
15.914
3.508
11.349
1.279
5.925
10.516
2.810
12.905
16.041
9.898
1.603
2.080
1.041
2.059
3.433
1.231
0.352
0.699
0.149
Mean:
Maximum:
Minimum:
246
T. Tang / Geomorphology 49 (2002) 231–254
Table 7
Mineral content of sediments and soils identified by X-ray diffraction (‘‘yes’’ indicates a positive identification)
Sample
number
Mineral
Silt#1
Silt#2
Silt#3
Silt#4
Sampling area Depth (cm) Silicates
and profile
Alpha-quartz
number
content of silt fraction of soil
EXS#1
0 – 15
EXS#1
15 – 60
EXS#1
60 – 110
EXS#1
110 – 180
and sediment
yes (high)
yes (high)
yes (high)
yes (high)
Iron oxides
Carbonates
Feldspars
Hematite
Goethite
Calcite Magnesite Dolomite Ankerite
yes
yes
yes
yes
very
very
very
very
yes
yes
yes
yes
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
(low)
(low)
(low)
(low)
little
little
little
little
(low)
(low)
(low)
(low)
Silt#13
Silt#14
Silt#15
EXS#4
EXS#4
EXS#4
0 – 10
10 – 35
35 – 70
yes (high)
yes (high)
yes (high)
yes (low)
yes (low)
yes (low)
very little
very little
very little
yes (low) no
yes (low) no
yes (low) no
no
no
no
no
no
no
no
no
no
Silt#23
Silt#24
Silt#25
GPT#2
GPT#2
GPT#2
0 – 20
20 – 50
50 – 100
yes (high)
yes (high)
yes (high)
yes (low)
yes (low)
yes (low)
very little
very little
very little
yes (low) no
yes (low) no
yes (low) no
no
no
no
no
no
no
no
no
no
Silt#26
Silt#27
Silt#28
Silt#29
GPT#1
GPT#1
GPT#1
GPT#1
0 – 15
15 – 30
30 – 60
60 – 120
yes
yes
yes
yes
(high)
(high)
(high)
(high)
yes
yes
yes
yes
(low)
(low)
(low)
(low)
very
very
very
very
little
little
little
little
yes
yes
yes
yes
(low)
(low)
(low)
(low)
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
Silt#30
Silt#31
Silt#32
Silt#33
YS#1
YS#1
YS#1
YS#1
0–8
8 – 20
20 – 70
70 – 110
yes
yes
yes
yes
(high)
(high)
(high)
(high)
yes
yes
yes
yes
(low)
(low)
(low)
(low)
very
very
very
very
little
little
little
little
yes
yes
yes
yes
(low)
(low)
(low)
(low)
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
Sample
number
Sampling area Depth (cm) Clay minerals
and profile
Chlorite high-Fe Chlorite low-Fe Vermiculite Smectite
number
Mineral
Clay#1
Clay#2
Clay#3
Clay#4
content of clay fraction of soil and sediment
EXS#1
0 – 15
yes (high)
EXS#1
15 – 60
yes (high)
EXS#1
60 – 110
yes (high)
EXS#1
110 – 180
yes (high)
yes
yes
yes
yes
(low)
(low)
(low)
(low)
Carbonates
Calcite Magnesite Dolomite Ankerite
(high)
(high)
(high)
(high)
yes
yes
yes
yes
yes
yes
yes
yes
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
Clay#13 EXS#4
Clay#14 EXS#4
Clay#15 EXS#4
0 – 10
10 – 35
35 – 70
yes (high)
yes (high)
yes (high)
yes (high)
yes (high)
yes (high)
yes
yes
yes
yes (low) no
yes (low) no
yes (low) no
no
no
no
no
no
no
no
no
no
Clay#23 GPT#2
Clay#24 GPT#2
Clay#25 GPT#2
0 – 20
20 – 50
50 – 100
yes (high)
yes (high)
yes (high)
yes (high)
yes (high)
yes (high)
yes
yes
yes
yes (low) no
yes (low) no
yes (low) no
no
no
no
no
no
no
no
no
no
Clay#26
Clay#27
Clay#28
Clay#29
GPT#1
GPT#1
GPT#1
GPT#1
0 – 15
15 – 30
30 – 60
60 – 120
yes
yes
yes
yes
(high)
(high)
(high)
(high)
yes
yes
yes
yes
(high)
(high)
(high)
(high)
yes
yes
yes
yes
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
Clay#30
Clay#31
Clay#32
Clay#33
YS#1
YS#1
YS#1
YS#1
0–8
8 – 20
20 – 70
70 – 110
yes
yes
yes
yes
(high)
(high)
(high)
(high)
yes
yes
yes
yes
(high)
(high)
(high)
(high)
yes
yes
yes
yes
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
T. Tang / Geomorphology 49 (2002) 231–254
Fig. 8. X-ray diffractogram of surface sediment in EXS (EXS#1).
247
248
T. Tang / Geomorphology 49 (2002) 231–254
Fig. 9. X-ray diffractogram of surface sediment in GPT (GPT#1).
T. Tang / Geomorphology 49 (2002) 231–254
Fig. 10. X-ray diffractogram of surface sediment in YS (YS#1).
249
250
T. Tang / Geomorphology 49 (2002) 231–254
The difference in Si contents of different soil profiles
might be significant, but major differences do not
occur between soil horizons in any one profile. Both
Al and Fe contents increase as soil depth increases.
This may be because of the translocation and podzolization of Al- and Fe-rich minerals. However, these
processes are not very intensive through the soil
profiles. Ti, K and Na also increase as the soil depth
increases, but the increase is small. Calcium carbonate
content either remains constant or decreases as soil
depth increases; Mg content either remains constant or
increases slightly with soil depth increase.
Mineral composition of the silt fraction of the soil
and sediment samples as determined by XRD spectrometry is shown in Table 7. The comparative X-ray
diffractograms from the surface horizon to the base of
the profile of three examples in the three sampling
areas (EXS#1, GPT#1 and YS#1) are shown in Figs.
8, 9 and 10, respectively. The dominant mineral in the
silt fraction of soil and sediment in all the samples is
alpha-quartz (d spacing 3.33 or 3.34, degrees of 2h:
26.7). Minor mineral contents are feldspars and goethite, at concentrations much lower than that of the
alpha-quartz. Hematite is present in trace amount. No
carbonate minerals were found in the soil samples
tested.
Interpreting the intensity of XRD patterns, the
contents of alpha-quartz remain relatively constant
through each of the soil profiles but differ between
different profiles. Contents of alpha-quartz increase
from the fengcong area of EXS, through the mixed
fengcong and fenglin area of YS, to the fenglin area of
GPT. Relative contents of feldspars decrease slightly
from the surface downward through soil profiles, and
also decrease from the fengcong area of EXS to the
fenglin area of GPT.
Mineral contents of the clay fraction of soils
identified by XRD spectrometry are also shown in
Table 7. The comparative X-ray diffractograms from
the surface horizon to the base of the profile of three
examples in the three sampling areas (EXS#1,
GPT#1 and YS#1) (Figs. 8, 9 and 10, respectively)
indicated that the dominant clay mineral in the soils
and sediments is chlorite. Vermiculite is present in
the soil and sediment either as a coexisting major
mineral with chlorite or as a minor mineral. Some
soil profiles, particularly those in the fengcong area
(EXS), contain a small amount of smectite. No
carbonate minerals were found, again, in the clay
fraction of the soil and sediment.
Through the interpretation of the relative intensity
of peaks on the X-ray diffractograms, chlorite content
decreases downward in all the soil profiles from the
three sampling areas. The trend of chlorite decline
with increasing soil depth increases from the fengcong
area (EXS) to the fenglin area (GPT). Chlorite content, in general, also increases from the fengcong area
(EXS) to the fenglin area (GPT). Vermiculite is the
major coexisting mineral with chlorite in the soil
profiles of the fengcong area (EXS), but in the fenglin
area (GPT) and the mixed fengcong and fenglin area
(YS), it is a minor mineral. Smectite mainly occurs in
trace amounts in the fengcong area of EXS.
5. Discussion
5.1. Purity of carbonate rocks and rock hardness
The results of limestone purity test are coincident
with those of previous studies in the Guilin as well as
elsewhere in SW China (Yuan et al., 1991). The purity
of the limestone is comparable to that of the limestone
in Jamaica where cockpit karst is developed (Day,
1982), but is much purer than limestone formations in
other locations in the Caribbean and Central America
(Day, 1978). No significant difference in purity was
noted between the two major limestone formations in
the region, namely, the Rong County Formation of the
upper Devonian (D3r) and the Dongangling Formation of the middle Devonian (D2d). However, the
dolostone of the region is less pure than that of the
limestone formations in terms of CaCO3 content
(Table 1).
In situ rock hardness testing by the Schmidt Hammer shows that the compressive strength of the limestone in the region is very high, with the mean R-value
ranging from 46 to 48. The hardness of the limestone
in Guilin is significantly higher than those of limestone
formations in the Caribbean and Central America, but
similar to that of limestone that support impressive
residual hills in other parts of SE Asia (Day, 1980,
1981, 1982). No significant differences in hardness
were found between the different limestone formations
in which the tower karst developed, as well as between
the limestone in which the different types of towers
T. Tang / Geomorphology 49 (2002) 231–254
formed. However, the hardness decreases from fresh,
broken surfaces to weathered surfaces and to lichencovered surfaces (Table 2). Hardness of the dolostone
in the region is much lower than that of the limestone.
No case hardening was detected in Guilin. The high
degree of chemical purity and the considerable compressive strength suggest that the limestone in Guilin is
highly durable in the face of mechanical breakdown
and physical weathering, but has a high susceptibility
to chemical dissolution. These material properties of
the limestone contribute to the development of very
steep slopes on the tower karst in Guilin.
Combination of compressive strength data and the
dissolution testing results of the limestone, dolostone
and mudstone suggests that (i) a direct relationship
exists between R-value and resistance of rock to
mechanical weathering and erosion; the limestone in
the region with the highest R-value is least susceptible
to mechanical weathering and breakdown; by contrast,
mudstone in the region with the lowest R-value is most
susceptible to mechanical weathering; (ii) R-value is
inversely related to the susceptibility of rock formations to chemical dissolution; the limestone with the
highest R-value is that most susceptible to chemical
dissolution.
5.2. Chemical environment of surface sediment
In contrast to the basic environment of the subsoil
karst system, soil and sediment analyses indicate that
the soil environment is predominantly acidic and that
the acidity increases with increasing soil depth. This
suggests that the soil and sediment environment is
more affected by the subtropical climate than by the
adjacent outcrops of limestone. The acidic environment is favorable to disintegration of coarse minerals,
oxidation and translocation of the clay minerals in the
soil and sediment. Particle size is coincident with the
pH value; the lower the pH, the finer the particle size
(Table 3 and Fig. 5).
The concentration and translocation of clay minerals through the soil profiles indicate intensive
processes of physical disintegration and chemical
decomposition in the study area. Previous studies
recognized (Rebertus et al., 1986; Van Wambeke,
1992) that some clay minerals were formed by
comminution of sand- and silt-sized particles; as
weathering proceeds, the clay percentages increase
251
mainly because of the formation of secondary minerals in the fraction. The high content of silt,
particularly on the surface, may reflect the extensive
influence of fluvial and alluvial processes. The
higher sand content of the sediment samples from
fenglin of the GPT area might suggest that the
fenglin areas are more influenced by fluvial processes than the fengcong and mixed fengcong and
fenglin areas.
Distributional patterns of soil pH and content of
carbonate are coincident visually with each other and
indicate that (i) the soil environment is acidic in spite
of the massive outcrops of limestone bedrock close by;
(ii) calcite, dolomite and carbonate contents in soils are
minor, suggesting that the influence of limestone
dissolution on the surface-unconsolidated materials is
very limited; (iii) acidity increases with increasing soil
depth, indicating further that the processes of geogenesis and pedogenesis in the soil and sediment of the
region are mainly controlled by subtropical climatic
conditions rather than by the dissolution of limestone
bedrock; (iv) a duality exists in the environmental
landscape system in Guilin: the unconsolidated materials of the soil and sediment on the one hand, the
carbonate rock outcrops on the other. The contact
between these two zones, particularly in terms of water
percolation, is one of the major foci of limestone
dissolution and karstification.
5.3. Relations between the surface sediment and the
bedrocks
Analysis of total elemental content indicates that
the distributional pattern of eight major elements in
the soil and sediment is similar to that of the Cretaceous mudstone, but very different from those of
limestone and dolostone (see Tables 4 and 6). Silicon
is the predominant elemental content in surface
sediment (mean 43%) and the mudstone (mean
67%). However, it only occurs in limestone and
dolostone in trace amounts (0.001% and 0.17%,
respectively). In contrast to the relatively high concentration of Fe and Al in the surface sediment
(mean 13% and 22%, respectively) and the mudstone
(mean 10% and 11%, respectively), the iron and
aluminum content of limestone and dolostone are
also very low (mean 0.03% and 0.025%, respectively). The Ti content increases from the mudstone
252
T. Tang / Geomorphology 49 (2002) 231–254
(0.8%) to soils and sediments (1.6%), but no Ti
presents in the limestone and dolostone.
The lower concentration of Si in the soil and
sediment compared to that of the mudstone suggests
that mudstone is not the only source of Si, and the Si
in soil and sediment might also be derived from
weathering of other clastic bedrocks, such as sandstones of the region. The several-magnitudes-higher
Si concentrations in the soil and sediment compared
to those in limestone and dolostone suggest that
limestone and dolostone are not the major source
of surface-unconsolidated materials and might not
even be a minor source. This observation is further
supported by comparing the Ti content of bedrocks
with the soil and sediment. Although no Ti presents
in the limestone or dolostone, a clear Ti content
exists in the mudstone and the soil and sediment.
Iron and aluminum contents of the soil and sediment
are also higher than those of the mudstone because of
oxidation in the subtropical environment. Conversely,
the Fe and Al contents in limestone and dolostone are
very low.
The low concentrations of CaCO3 and MgCO3
in soils and sediments are possibly explained by
two means, noting that more than 99% of chemical
constituents of limestone and dolostone are CaCO3
and MgCO3. First, chemical leaching and dissolution are enhanced and intensified in the subtropical
climate and acidic soil environment. Dissolved
CaCO3 and MgCO3 in soils are transported downward by percolation of soil water and are evacuated
in solution via underground drainage. Second, input
of CaCO3 and MgCO3 to the soil and sediment
depends on the vertical distribution of dissolution
and the relationship of the surface sediment to the
dissolution of limestone and dolostone. Surface
limestone exposures on steep-sloped tower karst
undergo relatively little dissolution because of the
Table 8
Pearson correlation matrix of elemental content of the bedrocks and
surface sediment in Guilin
R
Limestone
Dolostone
Mudstone
Sediment
Limestone
Dolostone
1.000
0.770
0.215
0.222
1.000
0.234
0.284
Mudstone
Sediment
1.000
0.943
1.000
rapid runoff and relative low acidic content in the
runoff (Tang et al., in preparation). Consequently,
major dissolution is focused in the subcutaneous
zone underneath soils and sediments where acidic
runoff is readily available (Williams, 1983, 1985).
Pearson correlation statistics were derived on the
contents of the eight major elements in the soil and
sediment, limestone, dolostone, and mudstone. The
Pearson correlation matrix is shown in Table 8. The
result indicates that Pearson’s correlation coefficient
(R) between the surface sediment and mudstone is
0.943. By contrast, the correlations are negative
between the surface sediment and limestone as well
as the sediment and dolostone.
In comparison of mineral contents of bedrocks
with that of soil and sediment samples, the dominant
mineral in the sediment and in the mudstone is
alpha-quartz. Minor mineral constituents of both
are feldspars, goethite and hematite (see Tables 5
and 7). By contrast, the major mineral in the limestone is calcite and that in the dolostone is dolomite.
Both carbonates contain only trace amounts of
alpha-quartz. Comparing the clay minerals in the
soil and sediment with the clay fraction of mudstone, the dominant clay mineral in both is chlorite
(see Tables 5 and 7). The minor clay minerals
include vermiculite and smectite (Barnhisel and
Bertsch, 1989; Douglas, 1989). No calcite and dolomite were detected in the soil and sediment.
In summary, examination of elemental and mineral
contents of the sediment and bedrock shows that
material composition of the limestone and dolostone
is very different from that of the unconsolidated
materials mantling the surface. This result elucidates
further that processes involved in the accumulation of
surface-unconsolidated materials are not those producing the limestone karst landforms, especially the tower
karst. The dissolution process largely is not responsible for the formation of the soil and sediment in the
region.
The surface sediment in the region was probably
derived from multiple sources, including in situ
weathering of the mudstone, redistribution and dispersion of the residuum by fluvial processes and
fluvial transportation and deposition of weathered
materials from sandstone and other adjacent clastic
rocks. The mechanisms and processes of erosion
and deposition of clastic and chemical sedimentary
T. Tang / Geomorphology 49 (2002) 231–254
rocks are clearly different. Although they may share
the same media, particularly water, the processes of
chemical dissolution and deposition of limestone
and dolostone are different than those of the weathering, transportation and deposition of clastic sediment.
6. Conclusions
The results of this study can be summarized into
the following observations. First, both limestone
and dolostone of the region are very pure in terms
of CaCO3 and MgCO3 concentrations, 99.5% and
98.5%, respectively. Second, compressive strength
of the limestone of the region is very high. No
significant difference in compressive strength between
different formations of the limestone was found, but
significant decreases occur from limestone to dolostone and from fresh-broken to weathered and to
lichen-covered limestone surfaces. Third, both elemental and mineral compositions of the soil and
sediment are different from those of limestone and
dolostone of the region. The elemental constituents of
the soil and sediment are highly correlated with the
clastic sedimentary rocks (0.943), such as the mudstone, but shows negative correlation with limestone
and dolostone. Finally, the limestone formations are
highly resistant to physical weathering and disintegration. The durability of the limestone versus physical
weathering and their high susceptibility to chemical
dissolution account for why residual towers can form
and persist.
The physical and chemical environment of the soil
and sediment differs from those of karstified carbonate rocks of the region. The chemical environment in
the soil and sediment is predominantly acidic despite
the adjacent outcrops of limestone and dolostone. The
pH increases as the soil depth increases. The underground karstified environment is predominantly basic.
Vertically downward from the surface, a dual-zone
environmental structure exists in the landscape system
in Guilin: the zone of unconsolidated clastic materials
on the surface and the zone of karstified limestone.
The environment and the processes differ in these two
zones. The chemical dissolution of limestone is not
mainly responsible for the accumulation of clastic
sediment on the surface.
253
Acknowledgements
The author would like to express his sincere
thanks to Dr. Michael J. Day at Department of
Geography, University of Wisconsin-Milwaukee, for
the thoughtful guidance and encouragement. I would
also like to thank the co-editor, Dr. R.A. Marston,
and the anonymous reviewers of this journal for
their valuable critiques and suggestions in the
process of revision. Many thanks also to Dr. Bruce
Brown, Mr. Robert Paddock, Ms. Kathy Graff and
Mr. Patrick Anderson at the Center for Great Lakes
Studies and Dr. Glen Fredlund and Ms. Mary Jo
Schabel at the Soils and Physical Geography
Laboratory, University of Wisconsin-Milwaukee,
for their help with XRF and XRD analyses. I am
also grateful to Professor Yuan, Daoxian at the
Institute of Karst Geology and Hydrology, Ministry
of Natural Resources, China, for his guidance of
fieldwork. Finally, I would like to dedicate this
paper to Dr. George N. Huppert, a life-long cave
conservation scholar.
References
Allen, B.L., Hajek, B.F., 1989. Mineral occurrence in soil environment. In: Dixon, J.B., Weed, S.B. (Eds.), Minerals in Soil Environments, 2nd ed. Soil Science Society of America, Madison,
WI, pp. 199 – 278.
Barnhisel, R.I., Bertsch, P.M., 1989. Chlorites and hydroxy-interlayered vermiculite and smectite. In: Dixon, J.B., Weed, S.B.
(Eds.), Minerals in Soil Environments, 2nd ed. Soil Science
Society of America, Madison, WI, pp. 729 – 788.
Bish, D.L., Post, J.E., 1993. Quantitative mineralogical analysis
using the Rietveld full-pattern fitting method. Am. Mineral.
78, 358 – 379.
Bish, D.L., Reynolds Jr., R.C., 1989. Sample preparation for X-ray
diffraction. In: Bish, D.L., Post, J.E. (Eds.), Modern Powder
Diffraction. Mineralogical Society of America, Washington,
DC, pp. 73 – 79.
Brindley, G.W., Brown, G., 1980. Crystal structures of clay minerals
and their X-ray identification. Mineralogical Society Monograph, vol. 5. Mineralogical Society, London, 495 pp.
Day, M.J., 1978. Morphology and distribution of residual limestone
hills (mogotes) in the karst of north Puerto Rico. Geol. Soc.
Amer. Bull. 89, 426 – 432.
Day, M.J., 1980. Rock hardness: field assessment and geomorphic
importance. Prof. Geogr. 32 (1), 72 – 81.
Day, M.J., 1981. Rock hardness and landform development in the
Gunong Mulu National Park, Sarawak, E. Malaysia. Earth Surf.
Process. Landf. 6, 165 – 172.
254
T. Tang / Geomorphology 49 (2002) 231–254
Day, M.J., 1982. The influence of some material properties on the
development of tropical karst terrain. Trans. Br. Cave Res. Assoc. 9 (1), 27 – 37.
Day, M.J., Goudie, A.S., 1977. The Schmidt Hammer and field
assessment of rock hardness. Br. Geomorphol. Res. Group,
Tech. Bull. 18, 19 – 29.
Deng, Z., Lin, Y., Zhang, M., Liu, G., Wei, Z., 1988. Karst and
Geological Structure in Guilin. Chongqing Publishing House,
China, 129 pp. (in Chinese).
Douglas, L.A., 1989. Vermiculites. In: Dixon, J.B., Weed, S.B.
(Eds.), Minerals in Soil Environments, 2nd ed. Soil Science
Society of America, Madison, WI, pp. 635 – 728.
Ford, D.C., Williams, P.W., 1989. Karst Geomorphology and Hydrology. Chapman & Hall, London, 601 pp.
Gee, G.W., Bauder, J.W., 1986. Particle-size analysis. In: Klute, A.
(Ed.), Method of Soil Analysis: Part I. Physical and Mineralogical Methods. American Agronomy Society and Soil Science
Society of America, Madison, WI, pp. 384 – 409.
Huang, J., Yan, Q., Wang, M., Zhou, W., Gua, C., Shi, J., Pei, J.,
Huang, J., Li, Q., 1988. Study on Karst Water Resource Evaluations in Guilin and Its Methodology. Chongqing Publishing
House, China, 134 pp. (in Chinese).
Jenkins, R., 1988. X-ray Fluorescence Spectrometry. Wiley, NY,
175 pp.
Jenkins, R., Gould, R.W., Gedcke, D., 1981. Quantitative X-ray
Spectrometry. Marcel Dekker, New York, 586 pp.
Moore, D.M., Reynolds Jr., R.C., 1989. X-ray Diffraction and the
Identification and Analysis of Clay Minerals. Oxford Univ.
Press, Oxford, 332 pp.
Moore, D.M., Reynolds Jr., R.C., 1997. X-ray Diffraction and the
Identification and Analysis of Clay Minerals, 2nd ed. Oxford
Univ. Press, Oxford, 378 pp.
Rebertus, R., Weed, S.B., Buol, S.W., 1986. Transformations of
biotite to kaolinite during saprolite – soil weathering. Soil Sci.
Soc. Am. J. 50, 810 – 819.
Sweeting, M.M., 1995. Karst in China, Its Geomorphology and
Environment. Springer-Verlag, Berlin, 265 pp.
Tang, T., Day, M.J., 2000. Field survey and analysis of hillslopes on
tower karst in Guilin, southern China. Earth Surf. Process.
Landf. 25, 1221 – 1235.
Tang, T., Yuan, D., Day, M.J., 2002. Effect of water runoff to limestone dissolution in a karstified watershed. J. Am. Water Resour.
Assoc. (in preparation).
Tertian, R., Claisse, F., 1982. Principles of Quantitative X-ray Fluorescence Analysis. Heyden, London, 385 pp.
Van Wambeke, A., 1992. Soils of the Tropics, Properties and Appraisal. McGraw-Hill, London, 343 pp.
Weng, J., 1987. Karst and Carbonate Rocks in Guilin. Chongqing
Publishing House, China, 180 pp. (in Chinese).
Williams, P.W., 1983. The role of the subcutaneous zone in karst
hydrology. J. Hydrol. 61, 45 – 67.
Williams, P.W., 1985. Subcutaneous hydrology and the development of doline and cockpit karst. Z. Geomorphol. N.F. 29,
463 – 482.
Williams, K.L., 1987. An Introduction to X-ray Spectrometry: Xray Fluorescence and Electron Microprobe Analysis. Allen and
Unwin, London, p. 370.
Yuan, D., 1981. A Brief Introduction to China’s Research in Karst.
The Institute of Karst Geology, Ministry of Geology, Guilin,
China, 33 pp.
Yuan, D., 1986. New observations on tower karst. In: Gardiner, V.
(Ed.), International Geomorphology 1986—Proceedings of the
First International Conference on Geomorphology. Wiley, Chichester, New York, pp. 1109 – 1123.
Yuan, D., 1992. Karst in southwest China and its comparison with
karst in north China. Quat. Sci. China 4, 352 – 361 (in Chinese).
Yuan, D., Zhu, D., Weng, J., Zhu, X., Han, X., Wang, X., Cai, G.,
Zhu, Y., Cui, G., Deng, Z., 1991. Karst of China. Chinese Geological Publishing House, Beijing, 244 pp.
Zhang, Z., 1981. Karst types in China. Geol. J. 4 (6), 541 – 570.
Zhu, X., 1988. Guilin Karst. Shanghai Scientific and Technical
Publishers, Shanghai, 188 pp.
Zhu, X., Wang, X., Zhu, D., Gong, Z., Tan, H., 1988. Karst Geomorphology and Cave Studies in Guilin. Chinese Geological
Publishing House, Beijing, 249 pp. (in Chinese).