Quartz Sand Grain Size and Surface Morphology in some ... · Quartz Sand Grain Size and Surface Morphology in some Tropical Soils S. Thanachit1, A. Suddhiprakarn1,*, I. Kheoruenromne1

www.thaiagj.org Thai Journal of Agricultural Science 2009, 42(1): 41-60

Quartz Sand Grain Size and Surface Morphology in some Tropical Soils

S. Thanachit1, A. Suddhiprakarn1,*, I. Kheoruenromne1 and R.J. Gilkes2

1Department of Soil Science, Faculty of Agriculture, Kasetsart University,

Bangkok 10900, Thailand 2School of Earth and Geographical Sciences, Faculty of Natural and Agricultural Science,

University of Western Australia, Crawley 6009, Australia

Corresponding author. Email: [email protected]

Abstract This study was conducted on catenae at Nam Phong and Khon Buri, Thailand to identify the

spatial variation in quartz sand grain morphology using SEM and particle size distribution data. Profiles were sampled at six positions; on the summit (SU), shoulder (SH), upper midslope (UM), lower midslope (LM), footslope (FS) and toeslope (TS) of the Nam Phong catena which has developed on sandstone in the upper part and shale in the lower part. Four profiles developed from basalt and colluvium on crest (CT), backslope (BS), footslope (FS) and valley floor (VF) were sampled on the Khon Buri catena.

Dissolution and precipitation features on quartz grain surfaces including pits, silica precipitation, crystal growth and adhering forms were present in soils at all positions. Evidence of mechanical damage including conchoidal fracture, angular edges, rounded edges and cracks was also present on quartz grains from all positions on both catenae. The particle size distributions indicate that changes in size sorting due in part to colluvial transportation and addition have occurred in both catenae. For the Nam Phong catena, sand content increases and clay content decreases from summit to footslope. There is a large increase in the clay content in the toeslope soil due to a combination of sedimentation, crystallization of clay minerals from solution and a possible contribution of finer material from underlying shale. In the Nam Phong catena the particle size distribution also indicates a discontinuity of parent material for shoulder, midslope and toeslope soils, possibly due to colluvium over residuum. Summit to footslope soils are composed mostly of transported sandstone material over weathered sandstone and the toeslope soil has developed on residual shale beneath surficial transported sandstone material, and this profile also contains authigenic clay minerals. The abundant quartz in Khon Buri Soils is inconsistent with a purely basaltic origin so colluvial or aeolian materials have been introduced into the profiles. Particle size analysis and quartz grain surface analysis did not provide unambiguous information on the origins of soil materials or their weathering history for either catena. This is possible due to the catenae being in a tropical region, where intense chemical weathering is pervasive and to the diverse origins of the soil materials.

Keywords: quartz grain surface, sand fractionation, SEM, tropical soils

Introduction The surface texture of quartz sand grains in soils

may identify the origins and weathering history of soil materials (Xiao et al., 1995; Helland et al.,

1997). The weathering intensity in soils may be revealed by surface rounding, etching or overgrowth on quartz grains (Asumadu et al., 1987; Marcelino et al., 1999).

42 S. Thanachit et al. Thai Journal of Agricultural Science

Scanning electron microscopy of surface features on quartz grains may help determine the sedimentary history of sand. Surface textures can be original, resulting from crystallization and post magmatic change occurring in rocks or can result from mechanical and chemical alteration of grains after liberation from the parent rock (Doornkamp and Krinsley, 1971). Douglas and Platt (1977) proposed that in general the surface texture of quartz grains becomes more complex with soil age. Some caution must be observed as grain surface features usually associated with a particular weathering or transport process may also be produced by other processes and in addition sample treatment may introduce artifacts (Darmody, 1985).

Typical chemically induced surface features on quartz grains in tropical soils are v-shape or triangular etch pits (Eswaran and Stoops, 1979). Pye and Mazzullo (1994) noted that silica dissolution and grain etching is particularly prevalent in the podzol A-horizon under conditions of high rainfall, low soil pH (3.0-4.5) and with a high concentration of organic acids derived from acidophyllous dune vegetation cover. Thus quartz in soil at a summit position may be less weathered than at other positions in a catena due to differences in drainage condition (Marcelino et al. 1999). Well drained soils on slopes are more intensely leached and experience stronger chemical weathering, which is reflected in the presence of more triangular etch pits and chatter marks on the surface of quartz grains. In the case of an imperfectly drained soil on a valley bottom the dissolution of quartz is inhibited by oversaturation in silica in the drainage water and deposition of secondary silica may occur altering the surface morphology of sand grains (Douglas and Platt, 1977). Quartz grains liberated from crystalline rocks are highly nonspherical, have angular outlines including crystal faces due to crystallization and deformation at elevated temperature and pressure. Quartz grains from cemented sedimentary rocks often have a moderately angular outline and well defined quartz overgrowths (Mazzullo and Magenhiemer, 1986). Quartz sand grains from mature, leached soils under a tropical climate may have severely altered surfaces and interiors due to them having reached advanced stages of the various dissolution and precipitation processes (Little et al., 1978).

V-shaped impact pits, conchoidal breakage and ridges can be produced by mechanical abrasion during transport (Hurst, 1981). Eswaran and Stoops (1979) attempted to relate conchoidal features to sedimentary environments including conditions under which the grain was dislodged from the rock prior to transport and sedimentation. Evidently classification of quartz grain shape and surface texture may help identify the uniformity of soil parent materials and pedogenic modification due to weathering.

In contrast to the extensive research on quartz grain surface features carried out on soils and sediments elsewhere, there has been relatively little research on the morphology of sand grains in tropical soils, which is the focus of this paper.

Materials and Methods Field Setting of Study Area

The Khorat Plateau in the north-east of Thailand covers approximately 170,000 km2 which is one third of the total area of the country (Parry, 1996). The Plateau is bounded to the north and east by the Mekong river, to the south by the Dong Rek mountain range and to the west by Petchabun and Dong Phrayayen mountain ranges. Regional topography is low (< 250 m MSL) and the surface is gently undulating and dotted with low hills and small shallow lakes. The regional plant cover consists of a range of deciduous to mixed deciduous communities. A large area is flooded during the wet season and in most parts soils are shallow and deficient in nutrients. The Khorat Plateau has received significant scientific attention by local quaternary geologists who consider it to have received an aeolian (loess-like) mantle of late Pleistocene to Holocene age (Hoang Ngoc Ky., 1989; Udomchake, 1989; Boonsener, 1991). Aeolian processes would be expected to have operated at much higher rates than at present during the colder periods of the last and earlier glacial cycles. During glacial periods the paleo-environmental of the Khorat Plateau was clearly cooler and probably drier than at present but there is no evidence to indicate periods of extreme aridity in this region associated with glacial maxima (Penny, 2001). Subsequent to aeolian deposition of the sands, bioturbation by termites

Vol. 42, No.1, 2009 Surface morphology of sand in some Tropical soils 43

and other organisms has modified this cover layer (mantle) (Löffler and Kubiniok, 1991, 1996; Sanderson et al., 2001).

The representative Nam Phong and Khon Buri catenae are in Khon Kaen and Nakhon Ratchasima Provinces respectively on the Khorat Plateau. The locations of soil profiles in the catena are shown in Figure 1. All soils were classified according to Soil Taxonomy (Soil Survey Staff, 1999). The Nam Phong catena can be divided into 6 geomorphic positions: summit (SU: Typic Kandiustult), shoulder (SH: Psammentic Paleustalf), upper midslope (UM: Psammentic Haplustalf), lower midslope (LM: Psammentic Haplustalf), footslope (FS: Oxyaquic Haplustalf) and toeslope (TS: Aeric Endoaqualf). Altitudes are 209, 206, 202, 199, 191 and 180 m MSL respectively. The length of the catena is about 1 km. The Nam Phong catena has developed on sedimentary rocks of the upper Triassic to Cretaceous Khorat group and lower Tertiary to lower Pleistocene Krabi group. Soils from the summit to footslope positions are on the Khok Kruat formation of the Khorat group. The soil on the toeslope position is on quaternary deposits of the Krabi group. The Khok Kruat formation consists of red to purplish gray bedded sandstone, siltstone and shale with interbedded calcareous conglomeratic sandstone, sandy limestone and calcareous siltstone which is mottled with purplish gray and greenish gray colors. The quaternary Khok Kruat floodplain deposit comprises unconsolidated silt, clay, sand and gravel (Suwanasing, 1972). Soils on summit to footslope positions are occupied by cultivated cassava fields but the main land use at the toeslope position of this catena is paddy rice field.

For the Khon Buri catena, four soils were sampled on an undulating basaltic lava corrosion plain. This catena transect has a northeast-facing slope and soils are considered to have formed on residuum or/and colluvium derived principally from basalt. The elevations above sea level are 275, 271, 233 and 225 m for the four soils on this catena. The soils sampled were on crest (CT), backslope (BS), footslope (FS) and valley floor (VF) positions. The soils were classified according to Soil Taxonomy (Soil Survey Staff, 1999) as Rhodic Kandiustox, Typic Kandiustult, Typic

Plinthustult and Typic Endoaquert for the four positions on the catena. The land uses are mainly cassava cultivation with some jackfruit, mango, sugarcane and rice.

The Nam Phong and Khon Buri catenae are under a tropical savannah climate with an average mean annual rainfall of 1,300 mm year-1. The highest average maximum temperature is 32.7 oC and average lowest temperature is 22.3oC. The average temperature is 26oC.

Laboratory Analysis

Representative profiles at each position on the Nam Phong and Khon Buri catenae were described and sampled by genetic horizon according to standard field study methods (Kheoruenromne, 1999). Bulk soil samples were air-dried and crushed (but not ground) and the fine earth fraction passed through a 2 mm sieve. The particle size distribution of the -2 mm fraction was determined by the pipette method. Carbonate and organic matter were first removed using NaOAc at pH 5.0 and H2O2 respectively. 0.1 N NaOH was used as a dispersing agent. The sand fraction was separated from the silt and clay fractions by wet sieving, oven dried and then fractionated by dry sieving with a Force mechanical shaker into the 5 classes: very coarse sand, coarse sand, medium sand, fine sand and very fine sand (2-1, 1-0.5, 0.5-0.25, 0.25-0.1 and 0.1-0.05 mm) (Day, 1965).

The fine sand fraction (0.25-0.1 mm) from two horizons per soil profile was selected for examination by scanning electron microscopy (SEM). Fine sand size quartz grains are most useful for surface feature analysis, they are large enough to show the surficial features of interest, yet not so large as to be composed of more than one material (Darmody, 1985). X-ray diffraction and optical microscopy showed that most of the soils on the catena have a fine sand fraction composed almost exclusively of quartz, containing no weatherable minerals except for traces of amphibole. Pedogenic calcite and magnesite are present in the toeslope soils but were removed by the pretreatment. The iron oxide cement between grains was removed with dithionite-citrate-bicarbonate solution (Mehra and Jackson, 1960), grains were then washed with deionised water and


NP

KB

THAILAND

180

185

190

195

200

205

2100 200 400 600 800 1000

MSL (m)

Distance (m)

Subangular; Sandy loam - sandy clay loam

Subangular; Sand – loamy sand

Subangular-semi massive; Silty clay

Iron and manganese nodules and gravels

Very gravelly sandy clay loam

(Psammentic Paleustalf)

(Psammentic Haplustalf)

(Aeric Endoaqualf)

SUSH

LM

UM

FS

TS

2040

2335

140

20

45

198

27

80

1003045

30

143

ApE

Bt

ApE

Bt

2C

Ap

Bt

2C

Ap

2BC

2C

Bt

Ap

Bw

Btg

Apg

Btg

BCg

(Oxyaquic Haplustalf)

(Typic Kandiustult)

(Psammentic Haplustalf)

SU: Summit; SH: Shoulder; UM: Upper midslope; LM: Lower midslope; FS: Footslope; TS: Toeslope

Remarks

225

235

245

255

265

2750 500 1000 1500 2000 2500 3000 3500 4000

(Rhodic Kandiustox)(Typic Kandiustult)

(Typic Plinthustult)

(Typic Endoaquert)

Ap

Bto

Bto

Bo

Ap

Bt

BCr

30

130

185

30

110

180

30

150

160

30

110

MSL (m)

Distance (m)

Subangular blocky structure parting to granular

Subangular blocky/semi-angular/angular

Subangular blocky; Fe/Mn nodules

Subangular blocky/semi-massive; slickensides

Basalt texture

Remarks

CT

VF

FSAp

Btc

BCr

Cr

Crg

BCg

Bg

Apg

BS

CT: Crest; BS: Backslope; FS: Footslope; VF: Valley floor

NP: Nam Phong catena

KB: Khon Buri catena

(Stone line)

(Stone line)

(Stone line)

Figure 1 The study sites on the Khorat Plateau of Northeast Thailand at Num Phong and Khon Buri and the catenary distribution of the soil profiles sampled in this research.

oven dried at 80 oC. Up to 100 fine sand grains were mounted on an aluminum stub with double sided carbon tape and coated with about 20 nm platinum before investigation under a 6400JEOL SEM. The machine was operated at 15 kV and 0.5 mA. Twenty randomly chosen fine sand grains were characterized in detail for surface features by

taking a photograph of the entire grain and then further photographs of edges and faces at higher magnification. The occurrence of surface features on grains in each sample is reported semiquantitatively using the terms abundant (>75%), common (35-75%), present (5-35%), few (1-5%) and absent (0%).


The Statistical Computations Particle size distribution curves for soils of the

Nam Phong and Khon Buri catenae were expressed on a Ф scale. Folk and Ward’s (1957) graphical method was adopted to calculate mean size (Mz), sorting (σI), skewness (SKI) and kurtosis (K'

G). This method involves the measurement of several percentiles from cumulative curves (Ф5, Ф16, Ф25, Ф50, Ф75, Ф84 and Ф95). The formulae are as follows:

Ф = -log2 G where G = the grain size (mm)

(i.e. sieve mesh opening)

Mean size Mz = Ф16 + Ф50 + Ф84

3

Sorting σI = Ф84 - Ф16 + Ф95 - Ф5

4 6.6

Skewness SKI = Ф16 + Ф84 - 2 Ф50 + Ф5 + Ф95 - 2 Ф50

2(Ф84 - Ф16) 2(Ф95 – Ф5)

Kurtosis K'

G = Ф95 – Ф5

2.44(Ф75 – Ф25)

These parameters aid in the characterization of the samples by providing a concise summary of particle size distribution and in some circumstances provide a basis for an interpretation of the environments of transport and deposition. The mean size may indicate the average kinetic energy of the deposition agent but as for all coefficients is also dependent upon the size distribution of the available source materials. The sorting coefficient is an indicator of fluctuations in kinetic energy of the transporting medium during deposition. Skewness of the particle size distribution is a measure of asymmetry and relates to the value of mean size with respect to median size. Negative skewness values indicate that the mean size of the sediment is towards the coarser side of the median whereas the positive skewness values shows that the mean is towards the finer side of the median. Kurtosis is a measure of the ratio of sorting within

the central 90% of the distribution curve to sorting of the central 50% and is sometimes described as a measured of peakedness (Sahu, 1964). It should be stressed that these sorting parameters may be inherited from parent rocks and not be a consequence of subsequent transport or pedogenesis. Statistical analysis was carried out using the Statistica Programme (Version 6.1).

Results and Discussion Soil Morphology

The field morphology of soils along the Nam Phong catena is summarized in Table 1. The summit profile experiences continuous well drained conditions, hence it is red to dark red in color with no mottles. The red color gradually decreases in intensity downslope along the catena. The toeslope profile is at the lowest position on the soil sequence, it is poorly drained and the soil has a yellow color and many mottles which is consistent with its wet environment. The pH of summit to footslope soils range from 4.5-6.5 and is more than 8 for the toeslope profile, which contains both CaCO3 and MgCO3.

The soils are deep at all positions on the catena. The thickness of the Ap-horizon increases from the summit to footslope position and is smaller in the toeslope profiles. Iron and manganese oxide nodules occur at 5-30 cm depth in the toeslope soil due to seasonal redox variations associated with the high water table at this position. In addition, there is evidence of an abrupt change in materials from the Bt-horizon to the 2BC or 2C horizon at the shoulder and midslope positions because of a presence of rock fragment gravel in the 2BC or 2C horizon (i.e. a stone line). Hence, parent material unconformities were recognized in the field at shoulder and midslope positions.

The field morphology of the soils on the Khon Buri catena is given in Table 1. All soils on this catena are deep. The accumulation of the clay in subsoil horizons for soils on the crest, backslope and footslope is apparent in the field. Clay balls and clay coated onto pore walls and ped faces can be seen in soils on the crest, backslope and footslope positions. For soil on the valley floor position, cracks are well expressed in both surface and subsurface horizons; slickensides or pressure


Table 1 Field morphology, particle size distribution, pH and soil features for the Nam Phong and Khon Buri catenae.

Color Particle size distribution(%)

Horizon Depth (cm)

Matrix Mottle

Field pH

Sand Silt Clay

Other

Nam Phong catena Summit: Typic Kandiustult

Ap 0-20 7.5YR5/4 - 6.3 851 104 46 - E 20-40 5YR6/8 - 6.4 830 114 56 Few faint clay bridges among sand grains and few

faint clay coats on ped faces and pore walls. Bt 40-205+ 2.5YR4/8 - 5.0 707 101 192 Common few faint clay bridges among sand grains

and few faint clay coats on ped faces and pore walls.Shoulder: Psammentic Paleustalf

Ap 0-23 10YR5/4 - 6.6 832 118 51 - E 23-35 7.5YR6/6 - 6.9 804 141 56 - Bt 35-140 7.5YR6/6 - 5.1 725 130 146 Common distinct faint clay bridges among sand

grains and common prominent clay coats on ped faces and pore walls.

2C 140-190+ 2.5YR4/6 - 5.0 540 159 301 Few distinct faint clay bridges among sand grains and few faint clay coats on ped faces and pore walls.

Upper midslope: Psammentic Haplustalf Ap 0-30 10YR4/3 - 6.0 909 61 30 - E 30-45 10YR5/4 - 6.3 888 87 25 - Bt 45-198 7.5YR6/6 - 5.9 855 96 49 Few distinct faint clay bridges among sand grains

and clay coats on pore walls. 2C 198-210+ 7.5YR7/6 - 5.3 678 101 221 Few distinct faint clay bridges among sand grains

and clay coats on pore walls. Lower midslope: Psammentic Haplustalf

Ap 0-35 10YR5/3 - 4.8 914 61 25 - Bt 35-80 10YR6/6 - 5.2 872 89 38 Very few faint clay bridges among sand grains. 2BC 80-100 7.5YR6/6 - 5.0 870 75 55 - 2C 100-180+ 2.5YR4/8 - 4.9 680 78 243 Common faint clay coats on pore walls.

Footslope: Oxyaquic Haplustalf Ap 0-30 10YR4/2 - 5.3 895 93 13 - Bw 30-45 10YR7/3 - 5.4 894 101 5 - Bt 45-130+ 7.5YR7/4 - 5.5 884 104 12 Few faint clay bridges among sand grains and clay

coats on pore walls. Toeslope: Aeric Endoaqualf

Apg 0-5 10YR5/3 6.0 723 191 86 - Btcg 5-30 10YR6/4 7.0 401 245 354 Few faint clay bridges among sand grains and few

faint clay coats on ped faces and pore walls. Btg 30-143 2.5Y6/6 8.9 228 372 401 Common distinct clay coats on pore walls and ped

faces. BCg 143-195+ 5Y7/1

2.5Y7/2 2.5YR4/6 2.5Y8/1 2.5Y7/2 10YR5/8

9.1 68 472 460 Few faint clay coats on pore walls and ped faces. Khon Buri catenae

Crest: Rhodic Kandiustox Ap 0-30 10R3/3 - 4.0 106 240 654 Many roots, common clay balls, few rock fragments.Bto 30-130 10R3/3 - 4.5 81 84 835 Many roots, few clay balls, few rock fragments. Bo 130-185+ 10R3/3 - 4.5 71 73 856 Few roots, common clay balls, very few rock

fragments. Backslope: Typic Kandiustult

Ap 0–30 5YR3/3 - 6.5 243 221 536 Common tuberrus roots, few clay balls, common rock fragments.

Bt 30-119 2.5YR4/6 - 5.5 138 115 747 Few roots, few clay balls, few crack, few rock fragments.

BCr 119-180+ 2.5YR4/6 - 5.5 188 152 660 Very few roots, common clay balls, many rock fragments and gravel, traces of ant’s nest.

Footslope: Typic Plinthustult Ap 0–30 5YR3/3 - 6.5 491 176 334 Many roots, few clay balls, many gravels. Btc 30-150 2.5YR3/6 - 4.5 207 112 681 Common roots, few clay balls, many gravels. BCrt 150-180 5YR3/4 - 4.0 199 177 624 Few roots, common clay balls, many gravels. Cr 180-197+ 2.5Y6/1 - 6.5 - - - Basalt texture.


Table 1 Cont.

Color Particle size distribution(%)

Horizon Depth (cm)

Matrix Mottle

Field pH

Sand Silt Clay

Other

Valley floor: Typic Endoaquert Apg 0-25 10YR3/2 7.5 143 265 592 Many roots, few rock fragments, few pressure faces.Bssg 25-50 2.5Y3/1 8.0 56 224 720 Few roots, few clay balls, common pressure faces,

common slickenside. Bg 50-110 2.5Y4/1 8.0 55 169 776 Very few roots, common clay balls, common

slickenside, few pressure faces, few rock fragments. BCg 110-160 Mixed

10Y3/1, 5Y3/2

8.0 131 171 698 Very few roots, few clay balls, few slickenside, few pressure faces, few rock fragments.

Crg 160-180+ Mixed 2.5Y3/1, 10GY3/2

10YR5/6, 5YR4/6, 2.5Y4/4, 2.5Y3/3, 5Y7/8

8.0 188 296 516 No roots, few pressure faces, few variegated sand

and rock fragments.

fractures are common in this soil. Many iron and manganese nodules occur throughout the soil profile on the footslope position whereas only a few iron and manganese nodules are present in the deep horizons of the soil on the backslope position. The presence of iron and manganese nodules in soils indicates the occurrence of wetting and drying periods inducing pedogenic redox processes (Acquaye et al., 1992). Small fragments of basalt are common in some horizons at all catena positions and weathered basalt bedrock is present in the deepest horizon of the soils on backslope and footslope positions. This evidence led us to conclude that all soils on this catena have formed predominantly from basalt but there is no field evidence to exclude the addition of exogenous materials.

Changes of soil color within the Khon Buri catena are obvious in the field. In soils on the crest, backslope and footslope positions, a dark red (10R, 2.5YR) color dominates with an abrupt change to a gray (10Y) color for the valley floor soil. Soils on the crest, backslope and footslope positions are well-drained and lack redoximorphic features. The soil on the valley floor is poorly drained, olive brown, dark olive gray and with yellow mottles which are features characteristic of aquic conditions occurring throughout the soil profile. The diverse soil properties existing within these two catenae indicate that chemical weathering conditions differ considerably throughout each catena and it is hypothesised that this difference may have resulted

in the development of different surface features on quartz. In particular the soils at higher elevation experience strong leaching whereas valley floor soils are in a zone of leachate accumulation.

Particle Size Distribution for Nam Phong Catena

The particle size distribution data (Figure 2) for

soils on the catena differ considerably which is consistent with variations in parent material and the action of colluvial transport processes on the catena. Soils from the summit to footslope have sand contents greater than 50% with textures ranging from sand to sandy clay loam. The high sand content is also present in the Ap-horizon of the toeslope soil, with sand content decreasing greatly with depth (Figure 3). Soils on all positions on the catena show a slight to strong increase of silt and clay contents and associated decrease of sand content with depth as is shown in Figure 3. There is evidence of the eluviation of fine material in the soil profiles on the summit, shoulder and toeslope positions but not for soils on midslope and footslope positions. These intermediate positions have highest slopes so that erosion occurs more quickly than elsewhere on the catena and the illuvial accumulation of clay is relatively less important. The soil on the toeslope contains much silt and clay in subsoil horizons. The subsoil may have developed on residuum derived from fine grain sedimentary rock with the surface horizon


Nam Phong catena Khon Buri catena

%Silt

0

10

20

30

40

50

60

70

80

90

100

%Clay

0

10

20

30

40

50

60

70

80

90

100

%Sand0102030405060708090100

%Silt

0

10

20

30

40

50

60

70

80

90

100

%Clay

0

10

20

30

40

50

60

70

80

90

100

%Sand

0102030405060708090100

Fig. 2. Ternary graphs of sand, silt and clay concentrations for soils of the Nam Phong and Khon Buri catenae. (Ap1) being sandy colluvium. The profile particle size distribution plots also indicate a clear lithological break within the solum at the shoulder, midslope and toeslope positions (Figure 3). There is an abrupt increase in clay concentration from the Bt-horizon to the 2C-horizon. It is proposed that the shoulder, midslope and toeslope profiles consist of sandy colluvium over more clayey weathered shale but pedoturbation may have caused considerable mixing of materials.

The depth functions for graphical sorting parameters for the Nam Phong catena are shown in Figure 4. Mean size (Ф units) of profile samples decreases systematically downslope from summit to footslope position as larger sand sizes become more dominant. Mean size (Ф units) is much larger in the toeslope samples due to the very clayey nature of much of this profile. Sorting varies with depth for all profiles. The soils from the summit to the footslope are poorly to very poorly sorted with the exception of the three deepest samples in the toeslope soil which are quite well sorted. The generally poor sorting might be attributed to colluvial transportation and mixing into profiles of materials from upslope positions. All soil materials on this catena exhibit highly positive skewness with no consistent trends with depth or position in the catena.

The coarse sand/fine sand ratio for Nam Phong soils (Figure 5) is somewhat variable but shows no

systematic trend with depth apart from for the lower midslope profile where the ratio increases substantially in the lower part of the profile.

Particle Size Distribution for the Khon Buri Catena

Soils of the Khon Buri catena contain up to 30% quartz sand (Figure 2) which is inconsistent with the materials being derived solely from the weathering of (quartz free) basalt. Prone (2003) considers that aeolian transport of sand occurred in Northeast, Thailand during glacial periods and this process may have contributed quartz to the Khon Buri catena. The major feature of the particle size depth functions for the Khon Buri catena is the pronounced clay bulge in subsoils of the backslope, footslope and valley floor profiles. The clay is mostly a result of weathering of basalt and the reduced clay percentages at the top and base of these three profiles are due respectively to colluvial addition of sand and eluviation of the clay from the topsoil, and the presence of sand and silt size partly weathered basalt grains in the deepest horizons. The crest profile shows a rather uniform distribution of materials with depth apart from an accumulation of silt or strongly aggregated clay in the Ap2 horizon. Figure 3 shows the distributions of sand, silt and clay for the soils of the Khon Buri catena. The increasing concentration of clay with depth probably indicates that translocation of clay


Nam Phong catena

0 20 40 60 80 100

Ap

E

Bt1

Bt2

Bt3

Bt4

Bt5

Bt6

Summit (%)

0 20 40 60 80 100

Ap

E

Bt1

Bt2

Bt3

Bt4

2C

Shoulder (%)

0 20 40 60 80 100

Ap

E

Bt1

Bt2

Bt3

Bt4

Bt5

Bt6

Upper midslope (%)

0 20 40 60 80 100

Ap

Bt1

Bt2

Bt3

2BC

2C1

2C2

Lower midslope (%)80 85 90 95 100

Ap1

Ap2

Bw

Bt

Btg1

Btg2

Footslope (%)0 20 40 60 80 100

Apg

Btcg

Btg1

Btg2

Btg3

Btg4

BCg1

BCg2

Toeslope (%)

Khon Buri catena

0 20 40 60 80 100

Ap1

Ap2

Bto1

Bto2

Bto3

Bto4

Bo1

Bo2

Bto5

Crest (%)

0 20 40 60 80 100

Ap1

Ap2

Bt1

Bt2

Bt3

BCr1

BCr2

Backslope (%)

0 20 40 60 80 100

Ap1

Ap2

Btc1

Btc2

Btc3

Btc4

Btc5

BCrt

Footslope (%)0 20 40 60 80 100

Apg

Bssg

Bg1

Bg2

BCg1

BCg2

Crg

Valley floor (%)

Figure 3 Profile distribution of sand, silt and clay for soils of the Nam Phong and Khon Buri catenae.

Sand

Clay

Silt


Nam Phong catena Khon Buri catena

020406080

100120140160180200220

0 1 2 3 4 5Mz

Soil

dept

h (c

m)

020406080

100120140160180200

2 3 4 5 6Mz

Soil

dept

h (c

m)

020406080

100120140160180200220

0 1 2 3σI

Soil

dept

h (c

m)

020406080

100120140160180200

0 1 2 3σI

Soil

dept

h (c

m)

020406080

100120140160180200220

-1.0 -0.5 0.0 0.5 1.0SKI

Soil

dept

h (c

m)

020406080100120140160180200

-0.8 -0.6 -0.4 -0.2 0.0SKI

Soil

dept

h (c

m)

020406080

100120140160180200220

0.0 0.5 1.0 1.5 2.0K'G

Soil

dept

h (c

m)

020406080

100120140160180200

0 1 2 3K'G

Soil

dept

h (c

m)

Figure 4 Grain size and sorting parameters (mean size (Mz), sorting (σI), skewness (SKI) and kurtosis (K'

G)) for the soils of the Nam Phong and Khon Buri catenae.

Crest

Backslope

Footslope

Valley floor

Summit

Shoulder

Upper midslope

Lower midslope

Footslope

Toeslope


from the A-horizon to B-horizon has taken place. The high clay content of the B-horizon may also indicate a high extent of transformation of the primary minerals in basalt to clay (Beckmann et al., 1974).

The depth function for graphical mean size (Ф) (Figure 4) shows that mean size is mostly small (Ф>4) and there is no systematic trend in mean size with depth apart from the upper part of the footslope soil being notably coarser (smaller Φ). There are systematic differences in sorting between the four profiles with the footslope profile being more poorly sorted, especially in the upper profile, which is again consistent with the admixture with the in situ materials of exogenous material such as the aeolian sediment proposed by Prone (2003).

Both skewness and kurtosis vary considerably for all four profiles with the footslope soil exhibiting greatest skewness which is consistent with the colluvial admixture of coarse sand to a predominantly fine textured material.

The depth function of the coarse/fine sand ratio (Figure 5) is constant for the soil on the crest, is larger and quite constant for the backslope soil and quite variable for the footslope and valley floor soils. Values of the ratio are high (>1.8) for the upper part of the footslope soil which may be indicative of the admixture of coarse sand from another rock type.

The major reason for the thorough study of particle size and sorting for these catenae was to establish if materials are simply weathered

residuum of underlying rocks or if there had been admixture of exogenous material. This knowledge is essential to the interpretation of surface features of quartz grains. If the sand particles in these soils include introduced grains then differences in surface features between profiles or horizons can not be ascribed to in situ chemical and physical alteration. The data presented in previous sections indicate that at least some profiles in both catenae are likely to include introduced sand grains.

A powerful statistical tool for comparing complex data sets is factor analysis. The data are reduced to a small number of variables (factors) that best discriminate data sets. The standardized particle size data (six size classes) were analysed in this way with outputs shown in Figure 6 being used to test for uniformity of materials. For the Nam Phong catena most of the materials are rather similar apart from B and C horizons of the lower midslope profile, and 2C horizons of the shallower upper midslope profiles which may therefore contain quartz from different sources. All of the toeslope materials are quite different from the other profiles in this catena. For the Khon Buri catena, all samples are quite similar apart from Ap1 and Ap2 horizons of the footslope soil which are distinctly different. This evidence indicates a uniformity of most materials within each catena and enables a more confident interpretation of the surface feature information discussed in the following section.

Num Phong catena

020406080

100120140160180200220

0.0 0.5 1.0 1.5 2.0 2.5

C/F ratio

Soil

dept

h (c

m)

Summit ShoulderUpper midslope Lower midslopeFootslope Toeslope

Khon Buri catena

0

20

40

60

80

100

120

140

160

180

200

0.0 0.5 1.0 1.5 2.0 2.5

C/F ratio

Soil

dept

h (c

m)

Crest BackslopeFootslope Valley floor

Figure 5 Depth functions for the coarse/fine sand (C/F) ratio for the Nam Phong and Khon Buri catenae.


Num Phong catena VCS CS

MS

FS VFS silt

clay

-1.0 -0.5 0.0 0.5 1.0Factor 1 : 45.63%

-1.0

-0.5

0.0

0.5

1.0

Fact

or 2

: 35

.41%

Khon Buri catena

VC C M F VF

Silt

Clay

-1.0 -0.5 0.0 0.5 1.0Factor 1 : 67.94%

-1.0

-0.5

0.0

0.5

1.0

Fact

or 2

: 14

.90%

Num Phong catena

-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4Factor 1: 45.63%

-6-5-4-3-2-101234567

Fact

or 2

: 35.

41%

Toeslope

Lowermidslope

Shoulder

Footslope

Summit Upper midslope

Summit Shoulder Upper midslope Lower midslope Footslope Toeslope

Khon Buri catena

-10 -8 -6 -4 -2 0 2 4 6Factor 1: 67.94%

-3

-2

-1

0

1

2

3

4

5

Fact

or 2

: 14.

90%

Footslope

Crest

Valley floor

Backslope

Crest Backslope Footslope Valley floor

Figure 6 Factor analysis based on soil particle size distribution for the Nam Phong and Khon Buri catenae. For the Nam Phong profiles some samples from the lower midslope and all samples from the toeslope are different from the remaining samples. For the Khon Buri catena two footslope samples and one crest sample are very different from the remaining samples.


Quartz Sand Grain Surface Features Scanning electron micrographs of representative

surface features of quartz sand gains are given in Figures 7, 8 and 9. The nomenclature and interpretation of quartz surface features are based on the work of several researchers (Krinsley and Donahue, 1968; Krinsley and Doornkamp, 1973;

Hurst, 1981; Asumadu et al., 1987). Three types of feature were recognized. Features characteristic of mechanical damage include surface roughness (Figure 7A), angular edges (Figure 7B), rounded edges (Figure 7C), conchoidal fracture (Figure 7D), cleavage plates (Figure 7E) and cracks (Figure 7F). Dissolution features include irregular etch pits

A B

F E

D C

Figure 7 SEM secondary electron micrographs of quartz grains showing (A) rough surface; (B) angular edge; (C) rounded edge; (D) conchodial fracture; (E) cleavage plate and (F) crack (the arrows show specific features).


A

D C

B

Figure 8 SEM secondary electron micrographs of quartz grains showing (A) etch pits; (B) v-shape pits; (C) triangular pits; (D) striae (the arrows show specific features).

(Figure 8A), v-shape pits (Figure 8B), triangular pits (Figure 8C) and solution striae (Figure 8D). Precipitation features include silica precipitation (Figure 9A), silica plate and sheets (Figure 9B), plastered amorphous silica (Figure 9C), adhering particles (Figure 9D), crystal growth (Figure 9E) and authigenic quartz (Figure 9F). Together these features reflect the various transport and chemical weathering processes that take place in soils. Provided the feature can be distinguished from features inheritance from previous environments they can be used to differentiate the extent of grain surface modification and dominant alteration processes at each position on the catena. All of the above surface features occur in soils in Northeast, Thailand and have been interpreted by Prone (2003) as indicating a complex provenance for soil material. The types and abundance of surface features on fine sand grains of selected horizons on

Nam Phong and Khon Buri catenae are summarized in Tables 2 and 3.

It is evident from Table 2 that all the above mentioned surface features occur on quartz grains from sites on the Nam Phong transect, with the exception of authigenic quartz which was present in minor amounts in only one sample from the summit. The various features indicate that physical change, dissolution and precipitation of silica have affected quartz grains from the entire catena. However there are no distinct trends in the abundance of features along the catena or down profiles. Consequently firm conclusions on pedogenesis can not be determined. There may be a weak trend towards the more common occurrence of etch pits in the upper part of the catena which in consistent with the stronger leaching condition at the higher sites.


A B

F E

D C

Figure 9 SEM secondary electron micrographs of quartz grains showing (A) silica precipitation; (B) silica plate; (C) plastered amorphous silica; (D) adhering particle; (E) crystal growth and (F) authigenic quartz (the arrows show specific features).

The quartz sand grain surfaces for the selected horizons of the Khon Buri catena are given in Table 3. The features are similar to those for the Nam Phong catena. Features indicative of the dissolution of silica (i.e. irregular etch pit, V-shape pit, triangular pit) are dominant in soils from all position, however the soil on the valley floor has the lowest frequency of silica dissolution features which is consistent with the less aggressive

chemical conditions in this soil. The abundant rounded and sub-rounded edges observed in soils for all landscape positions indicate that transportation rounding has taken place. This degree of rounding can not be attributed to short-distance colluvial transport but is consistent with weathering dissolving sharp edges and also the introduction of exogenous rounded grains.


Table 2 Summary of quartz sand grain surface features in soils on Nam Phong catena.

Summit Shoulder Upper midslope Lower midslope Footslope Toeslope Feature Profile horizon E Bt6 E Bt3 2C E Bt5 2C Bt2 2BC 2C Bw Btg1 Apg BCg1

Conchoidal Rough surface Angular edges Rounded edges Cleavage plate Crack

Tran

spor

tatio

n

Etch pit V-shape pit Triangular pit Solution line and striae

Dis

solu

tion

Silica precipitation Silica plates and sheets

Plastered amorphous silica

Adhering particle Crystal growth Authigenic quartz

Wea

ther

ing

Prec

ipita

tion

Absent Few Present Common Abundant

56 S. Thanachit et al

.Thai Journal of Agricultural Science


Table 3 Summary of quartz sand grain surface features in soils on Khon Buri catena.

Crest Backslope Footslope Valley floor Feature Profile horizon Bto1 Bo1 Ap2 Bt2 Bt2 BCrt Bssg Crg

Conchoidal

Rough surface

Angular edges

Rounded edges

Sub-rounded edges

Crack

Tran

spor

tatio

n

Etch pit

V-shape pit

Triangular pit Solution lines and

striae

Dis

solu

tion

Silica precipitation

Silica plate and sheets Plastered amorphous

silica

Adhering particle

Crystal growth

Authigenic quartz

Wea

ther

ing

Prec

ipita

tion

Absent Few Present Common Abundant

Descriptions of grain roundness used in this paper follow the nomenclature of Powers (1953). The roundness classes for quartz from selected horizons at each position are shown in Figure 10. Sub-rounded to sub-angular grains are dominant but there are no systematic differences between soil profiles or horizons for either catena.

Conclusions Quartz sand grains from the Nam Phong and

Khon Buri catenae exhibit a large range of surface features due to precipitation, dissolution and

mechanical damage. Precipitation and dissolution features occur on quartz grains in soils at all position on the catenae. Soils on the Khon Buri catena have more dissolution features than those on the Nam Phong catena. The hypothesis that differences in parent material and differences in weathering environment for the catena would be associated with clear, systematic differences in surface features has not been sustained. Differences in particle size distribution within and between profiles are consistent with differences in parent materials, with differences due to particle size sorting colluviation and authigenesis of clay in


Num Phong catena

0 10 20 30 40 50 60 70 80 90 100

SU_ESU_Bt6

SH_ESH_Bt3

SH_2cUM_E

UM_Bt5UM_2C

LM_Bt2LM_2BC

LM_2CFT_Ap2

FT_Btg1TS_Apg

TS_BCg1

Soil_

Hor

izon

Grain (%)

Khon Buri catena

0 10 20 30 40 50 60 70 80 90 100

CT_Bto1

CT_Bo1

BS_Ap2

FS_Bt2

FS_Btc2

FS_BCr1

VF_Bssg

VF_Crg

Soil_

Hor

izon

Grain (%)

round sub-round sub-angular angular

Figure 10 The sand roundness classes for selected horizons of the Nam Pong and Khon Buri catenae. the lowest position on the catenary landscape together with the possible introduction of exogenous material, possibly the aeolian deposits proposed by Prone (2003). This discontinuity of parent materials at shoulder, upper midslope and toeslope positions is consistent with the field morphology, the presence of a stone line and the particle size distributions. The principal component analysis also clearly indicates a discontinuity in soil materials at the shoulder and midslope positions on the Nam Phong catena. For the soils on the Khon Buri catena, the principal component analysis, graphical sorting parameters and particle size distributions show the near uniformity of the parent

materials within most of the profiles. Thus the Khon Buri profiles provide a suitable situation for investigation differences in quartz grain morphology due to in situ process.

For the Nam Phong catena, the soils on the summit, shoulder, midslope and footslope have formed in colluvium over chemically altered sandstone while authigenesis of clay and carbonate has occurred in the toeslope soil. All soils on the Khon Buri catena have been mostly derived from basalt but colluvial materials exist over in situ materials in this catena so differences in quartz grain morphology may reflect differences in provenance..


Acknowledgments This work is supported by the Royal Golden

Jubilee Programme under The Thailand Research Fund. We are grateful to the Center for Microscopy and Microanalysis, The University of Western Australia for providing the SEM.

References

Acquaye, D.K., G.N. Dowuona, A.R. Mermut and R.J. St.

Arnaud. 1992. Micromorphology and mineralogy of cracking soils from the Accra Plains of Ghana. Soil Sci. Soc Am. J. 56: 193-201.

Asumadu, K., R.J. Gilkes, H.M. Churchward and T.M. Armitage. 1987. Detailed characterization of quartz grains in two sandy soils, Western Australia. Geoderma 41: 29-47.

Beckmann, G.G., C.H. Thompson and G.D. Hubble. 1974. Genesis of red and black soil on basalt on the Darling Downs, Queensland, Australia. J. Soil Sci. 25: 265-281.

Boonsener, M. 1991. The geomorphology of Quaternary stratigraphy of northeast Thailand. J. Thai Geosci. 1: 23-32.

Darmody, R.G. 1985. Weathering assessment of quartz grains: a semiquantitative approch. Soil Sci. Soc. Am. J. 49: 1322-1324.

Day, D.R. 1965. Particle fraction and particle size analysis, pp. 545-566. In C.A. Black ed., Methods of Soil Analysis, Part I. Agronomy No. 9. Am. Soc. of Agron. Inc., Madison, Wisconsin, USA.

Doornkamp, J.C. and D. Krinsley. 1971. Electron microscopy applied to quartz grains from a tropical environment. Sedimentology 17: 89-101.

Douglas, L.A. and D.W. Platt. 1977. Surface morphology of quartz and age of soils. Soil Sci. Soc. Am. J. 41: 641-645.

Eswaran, H. and Stoops, G., 1979. Surface texture of quartz in tropical soils. Soil Sci. Soc. Am. J. 34, 420-424.

Folk, R.W. and W.C. Ward. 1957. Brazos River bar: a study in the significance of grain size parameters. J. Sed. Petrol. 27: 3-16.

Helland, P.E., Huang, Pei-Hua and R.F. Diffendal Jr. 1997. SEM Analysis of quartz sand grain surface textures indicates alluvial/colluvial origin of the quaternary ‘Glacial’ Boulder Clays at Huangshan (Yellow Mountain), East-Central China. Quaternary Res. 48: 177-186.

Hoang Ngoc Ky. 1989. The Thu Duc loess formation, a typical aeolian deposit of tropical regions, pp. 100-104. In ESCAP Secretary, Quaternary Stratigraphy of Asia and Pacific. IGCP 296 (1989). ESCAP Atlas of Stractigraphy X (Mineral Resources Development series 60).

Hurst, A.R. 1981. A scale of dissolution for quartz and its implications for diagenetic processes in sandstones. Sedimentology 28: 451-459.

Kheoruenromne, I. 1999. Soil Survey Laboratory Manual. Dept. of Soil Science, Faculty of Agricultural, Kasetsart University, Bangkok, Thailand.

Krinsley, D.H. and J. Donahue. 1968. Environmental interpretation of sand grain surface textures by electron microscopy. Geo. Soc. Am. Bull. 79: 743-748.

Krinsley, D.H. and J.C. Doornkamp. 1973. Atlas of Quartz Sand Surface Textures. Cambridge University Press, Cambridge.

Little, I.P., T.M. Armitage and R.J. Gilkes. 1978. Weathering of quartz in dune sands under subtropical conditions in eastern Australia. Geoderma 20: 225-237.

Löffler, E. and J. Kubiniok. 1991. The age and origin of the Yasothon soils and associated gravel deposits. J. Geo. Soc. Thailand 1: 69-74.

Löffler, E. and J. Kubiniok. 1996. Landform development and bioturbation on the Khorat Plateau, Northeast Thailand. Nat. His. Bull. Siam Soc. 44: 199-216.

Marcelino, V., G. Mussche and G. Stoops. 1999. Surface morphology of quartz grains from tropical soils and its significance for assessing soil weathering. European J. Soil Sci. 50: 1-8.

Mazzullo, J. and S. Magenheimer. 1986. The original shapes of quartz sand grains. J. Sed. Petrol. 57: 479-487.

Mehra, O.P. and M.L. Jackson. 1960. Iron oxide removal from soils and clays in a dithionite-citrate-bicarbonate system buffered with sodium. Clays Clay Miner. 7: 317-327.

Parry, J.T. 1996. The high terrace gravels, Northeast Thailand: a re-evaluation and an integrated theory of their origin. Z. Geomorph. N. F. 40: 145-175.

Penny, D. 2001. A 40,000 year palynological record from north-east Thailand: implications for biogeography and paleo-environmental reconstruction. Paleogeography, Paleoclimatology, Paleoecology 171: 97-128.

Powers, M.C. 1953. A new roundness scale for sedimentary particles. J. Sed. Petrol. 23: 117-119.

Prone, A. 2003. L'analyse Texturale et Microstructurate des Sols: Exemple P'edologique du Nord-Est de la Thailande. Publications DE L' Universit'e de Provence. (in French).

Pye, K and J. Mazzullo. 1994. Effects of tropical weathering on quartz grain shape: an example from Northeastern Australia. J. Sed. Res. A. Sed. Pet. Pro. 64: 500-507.

Sahu, B.K. 1964. Depositional mechanisms for the size analysis of clastic sediments. J. Sed. Petrol. 31: 73-83.

Sanderson, D.C.W., P. Bishop, I. Houston and M. Boonsener. 2001. Luminescence characterization of quartz-rich cover sands from NE Thailand. Quaternary Science Reviews 20: 893-900.

Soil Survey Staff. 1999. Soil Taxonomy: A Basic System of Soil Classification for Making and Interpreting Soil Surveys. 2nd. USDA-SCS Agric. Handbook. United State Government Printing Office, Washington, D.C.


Suwanasing, A. 1972. General Geology of Thailand. Department of Mineral Resource, Thailand.

Udomchoke, V. 1989. Quaternary stratigraphy of the Khorat Plateau area northeast Thailand, pp. 69-74. In N. Thiramongkol, ed., Proceedings of the Workshop on Correlations of Quaternary Succession in South, East and Southeast Asia. UNESCO Regional Office for Science and Technology SE Asia, ESCAP.

Xiao, J., S.C. Porter, Z. An, H. Kumai and S. Yoshikawa. 1995. Grain size of quartz as an indicator of winter monsoon strength on the loess Plateau of central China during the last 130,000 yr. Quaternary Research 43: 22-29.

Manuscript received 5 September 2008, accepted 3 March 2009

Now online at http://www.thaiagj.org

Documents

Quartz Sand Grain Size and Surface Morphology in some ... · Quartz Sand Grain Size and Surface Morphology in some Tropical Soils S. Thanachit1, A. Suddhiprakarn1,*, I. Kheoruenromne1