Published in Soil Sci. Soc. Am. J. 68:185-193 (2004).
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
DIVISION S-5—PEDOLOGY
Composition, Fabric, and Porosity of an Arenic Haplustalf of Northeast Thailand
Relation to Penetration Resistance
Ary Bruand*,a,
Christian Hartmannb,
Santi Ratana-Anupapb,
Pramuanpong Sindhusenb,
Roland Possb and
Michel Hardyc
a ISTO, UMR 6113 CNRS-UO, Univ. d'Orléans, Géosciences, BP 6759, 45067 Orléans, Cedex 2, France
b Land Dev. Dep., Office of Sci. for Land Dev., Phaholyothin Road, Chatuchak, Bangkok 10900, Thailand
c Unité de Science du Sol, INRA Orléans, Domaine de Limère, Ardon, BP 20619, 45166 Olivet Cedex, France
* Corresponding author (ary.bruand{at}univ-orleans.fr).
 |
ABSTRACT
|
|---|
Sandy uplands of northeast Thailand have subsoils with high resistance to root penetration that varies even though there is no variation in bulk density (Db). The objective of this study was to determine how the resistance to penetration in these soils is related to the size, mineralogy, and arrangement of the elementary particles. We studied the tilled topsoil (Ap horizon), upper subsoil (E horizon), and lower subsoil (Bt horizon) of an Arenic Haplustalf. Results showed similar skeleton grain-size distribution in the three horizons. Results also showed similar Db in the E and Bt horizons, and higher resistance to penetration in the E horizon. Dispersion of the material indicated the absence of a cementing agent between elementary particles. The <2-µm fraction content was 37, 53, and 88 g kg–1 in the Ap, E, and Bt horizons, respectively, and that fraction was about one-third quartz. The clay minerals were kaolinite with a small amount of swelling 2:1 clays in the Ap and Bt horizons. Mercury porosimetry and scanning electron microscopy (SEM) showed a closer arrangement of the sand resulting in less interparticle porosity in the E horizon than in the Bt horizon. The smaller interparticle porosity in the E horizon was compensated for by greater macroporosity resulting from biological activity, thus explaining the similar Db in the E and Bt horizons. Despite similar Db, the greater penetration resistance in the E horizon results from a close packing of sand particles that restricts the displacement of soil particles during penetration by probe or roots.
Abbreviations: BESI, backscattered electron scanning image • Db, bulk density • De, equivalent diameter • SEM, scanning electron microscopy • XRD, X-ray diffraction • Vp,t, total pore volume
|
INTRODUCTION
|
|---|
GENERALLY, SANDY SOILS under native vegetation do not exhibit root growth restriction. However, when sandy soils are cultivated, and more frequently than in other soils, the upper subsoil is compacted, thus reducing root development in the whole subsoil (Jones, 1983; Vepraskas, 1994; Busscher et al., 2001). As a result, yields can be affected and deep tillage or subsoiling are used to reclaim subsoil structure (van Ouwerkerk and Raats, 1986; Vepraskas and Miner, 1986; Busscher et al., 2000, 2001).
In the uplands of northeast Thailand, sandy soils comprise 
62% of the land area (Kheoruenromne and Suddhiprakarn, 1984; Mitsuchi et al., 1986; Craig and Pisone, 1988; Kheoruenromne et al., 1998). In the last 50 yr, forests were cleared and replaced by crops like cassava (Manihot esculenta Crantz) and sugarcane (Saccharum officinarum L.). The topsoil of the sandy soils is acidic (pHCaCl2 < 5.0) with clay content < 50 g kg–1 and C content < 8 g kg–1 (average = 5 g kg–1) as a result of continuous cultivation since clearing (Ragland and Boonpuckdee, 1987; Singhatat, 1996; Poss et al., 1998). In these sandy soils, field experiments showed that root growth was restricted in the upper subsoil (Kheoruenromne and Suddhiprakarn, 1984; Vityakon and Keerati-Kasikorn, 1987; O'Donnell et al., 1994; Sakurai et al., 1996) where there was no apparent difference in compaction as indicated by no variation in subsoil Db (Hartmann et al., 1999).
An increase in the resistance to penetration without any variation in Db may result from the presence of cements that are responsible for the cohesion of the skeleton particles. Iron oxy-hydroxides and aluminum hydroxides are not responsible for variation in the resistance to penetration because they comprise only a small fraction of the upland sandy soils of northeast Thailand (Yoothong et al., 1997). On the other hand, amorphous silica may contribute significantly to an increase in the interparticle cohesion as shown in loamy and fine-silty soils (e.g., Lindbo and Veneman, 1989; Smeck et al., 1989; Duncan and Franzmeier, 1999). A variation in the clay fabric may also cause a variation of the cohesion in coarse-textured soils. Lamotte et al. (1997a)(b) showed in sandy-loam soils from Northern Cameroon that an increase in cohesion was related to the presence of clay wall-shaped bridges linking the skeleton grains. Miura et al. (1992) studied the clay mineralogy of sandy soils from northeast Thailand, but there is no information available about the arrangement of the elementary particles and resulting pore-size distribution. Thus, the causes of the increase in the resistance to penetration in these subsoils are still unknown.
The objectives of this study were (i) to characterize the pore-size distribution and fabric, and (ii) to measure the resistance to penetration as a function of water content to determine how resistance to penetration is related to the size, mineralogy, and arrangement of the elementary particles.
|
MATERIALS AND METHODS
|
|---|
Site Characteristics and Soils
The study was performed 15 km north of Korat, Thailand. The soil (loamy, siliceous, isohyperthermic Arenic Haplustalf) developed in sandy alluvial materials coming from erosion of soils developed on sedimentary rocks. The soil belongs to the Nam Phong series (Moormann et al., 1964) (Table 1). Soils belonging to the Nam Phong series occupy 5274 km2 in northeast Thailand (Miura et al., 1992). Average annual precipitation is 1020 mm and highly irregular from year to year (min. = 599 mm and max. = 1446 mm for the 1971–1998 period). Monthly rainfall is characterized by high variation, with rainfall exceeding evaporation during the crop-growing season and at least three months of dry season each year. A water table is present at 50 cm during the rainy season. The plot area was in black cow-pea (Vigna unguiculata L. Walp) in 1997 and 1998 and Brazilian lucerne [Stylosanthes hamata (L.) Taub.] in 1999. The topsoil (5YR 4/3) was plowed to an 18- to 20-cm depth every year since clearing, either manually or by disk-plow. Ridges, 10 cm high and 30 cm apart, were made before sowing to prevent flooding during rainfall events. A subsoil (5YR 5/6) with a high resistance to penetration that decreases with depth underlies the tilled topsoil.
Field Measurements
We studied three pits during the rainy season and three others during the dry season. In every pit we measured the Db in the Ap horizon at a depth of 10 to 20 cm (Ap horizon), in the upper subsoil between 25 and 35 cm (E horizon), and in the lower subsoil between 40 and 50 cm (Bt horizon). We measured Db in duplicate in every horizon with cylinders 100 cm3 in volume. Penetrometer measurements were made during the rainy season and the dry season with a pocket penetrometer (Soiltest Inc. CL700, Chicago, IL). The piston (6.4 mm in diameter) was pushed horizontally into the soil on the vertical face of every pit at the 15-, 30-, and 45-cm depth, every measurement with a gap of 10 cm. As the piston needle was pushed into the soil, the maximum deformation of the spring was recorded and expressed as soil strength in MPa. We also measured the soil water content on samples collected from the face of the pit using the same sampling grid as for penetrometer measurements. The water content was expressed on a mass basis after oven-drying the sample at 105°C for 24 h.
Laboratory Measurements
Undisturbed samples (15 by 10 by 8 cm3) were collected during the dry season in the Ap, E, and Bt horizons of one of the three pits studied. Particle-size distribution was measured after pretreatment with H2O2 and sodium hexametaphosphate. Clay (<2 µm) and fine silt (2–5 and 5–20 µm) content was measured by the pipette method. Coarse particle distribution (50–200, 200–500, 500–1000, 1000–2000, and 2000–5000 µm) was determined by wet sieving the >50 µm material. The coarse-silt (20–50 µm) content was determined by computing the difference between 100 and the sum of the clay, fine-silt, and sand fractions. The clay-sized fraction (<2 µm) was extracted after pretreatment with H2O2 and NaOH at pH = 9, mechanical shaking for 4 h, and decantation following centrifugation. Clay was saturated with Mg2+ using MgCl2 (1N) and washed successively with water and 1:1 water:ethanol until free of chloride, as determined by the AgNO3 test (Hardy et al., 1999). Mineralogical analysis was performed by X-ray diffraction (XRD) on powder samples obtained by grinding <2-mm air-dried soil and on oriented samples obtained by sedimentation of the <2-µm fractions on a glass slide and air-drying. The air-dried Mg-saturated and ethylene-glycol-treated Mg-saturated <2-µm fractions were examined. The K-saturated, heated at 520°C <2-µm fraction was also examined. We used a Philips PW1730 X-ray generator and a PW1710 control panel (CoK
radiation, Fe filter, 40 kV and 30 mA). Quartz content of the clay-sized fraction was measured by XRD using ZnO as the internal standard (Hardy, 1992). Soil pH was measured at 25°C in a 1:5 soil:water suspension. Organic C content was measured by oxidation with an excess amount of potassium dichromate in sulfuric acid at 135°C. Cation exchange capacity was measured by the cobalt-hexamine trichloride method (Ciesielski and Sterckeman, 1997).
Mercury porosimetry involves measurement of the pressure required to force mercury into the pores of a dry sample and of the volume of intruded mercury at each pressure (Fiès and Bruand, 1998). Mercury intrusion was performed with a porosimeter (Micromeritics 9320, Mönchengladbach, Germany) which operated from a pressure of 4 kPa up to a maximum of 2000 kPa, enabling pore-size distribution study for pores with equivalent diameters (De) ranging from 360 down to 0.006 µm, respectively. Values for the surface tension of mercury and its contact angle on soil material were 0.484 N m–1 and 130°, respectively. Small clods (1–2 cm3 in volume) were selected and dried at 105°C for 24 h before mercury injection. Six clods were measured for each horizon.
For fabric analysis, undisturbed samples 5 x 5 x 8 cm3 were oven-dried at 40°C for a week and impregnated with polyester resin that was diluted with styrene monomer (33% by volume) at room temperature under a vacuum of 5kPa (Ukikapon S 2983V, Cray Valley, Drocourt, France). The hardener [Butanox M50, 2-Chloro-2',6'-diethyl-N-(butoxymethyl)acetanalide] was used at a rate of 0.23% by volume of base resin. The resin-impregnated samples were left for 4 wk at room temperature to ensure complete polymerization. Then small blocks measuring 45 x 60 x 1 mm were cut. They were polished with diamond grains of decreasing size (down to 0.25 µm) sprayed on polished sheets. The surface of the blocks was examined via SEM (90B, Cambridge Instruments Ltd., Cambridge, UK) with backscattered electron emission (Bruand et al., 1996). The intensity of backscattered emission is closely dependent on the atomic number of the target. This technique enables pores occupied by resin, which appear black, to be distinguished from the soil mineral particles, which appear lighter, with a resolution of 0.1 µm. The backscattered electron scanning images (BESIs) are like very thin sections of a few micrometers thickness because the investigated depth with backscattered electrons can be estimated as ranging from 2 µm for mineral grains to 5 µm for pores occupied by resin. Two thin sections were examined for each horizon studied and the BESIs were recorded as gray-level images at several magnifications to examine the fabric.
|
RESULTS
|
|---|
Field Observations
A few discontinuous channels from one to several millimeters in diameter and several centimeters long that are termite or root channels were observed in the E and Bt horizons. Few chambers one to several centimeters in size that result from termite activity were also observed in the E and Bt horizons.
Particle-Size Distribution and Mineralogy
Optical microscopy observations showed that the sand fractions collected after particle-size distribution measurements were composed of elementary sand particles without any clay or silt coating. Particles 50 to 500 µm in size represented 738, 730, and 671 g kg–1 in the Ap, E, and Bt horizons, respectively (Table 2). The silt and clay content ranged from 123 to 134 g kg–1 and from 37 to 88 g kg–1, respectively, with the highest clay content in the Bt horizon. The XRD patterns of the whole soil (<2-mm material) showed the presence of quartz in the three horizons studied and a small amount of rutile [d(110) = 0.325 nm] in the Bt horizon (Fig. 1)
. The XRD patterns of the clay fraction (<2 µm) showed the presence of phyllosilicates and quartz (Fig. 2)
. Phyllosilicates are mainly kaolinite (wide reflection, d(001) = 0.719–0.726 nm) with some swelling 2:1 clays (Fig. 2). Figure 2 shows clear evidence of the presence of 2:1 clays in the Bt horizon. The 2:1 clays are slightly visible in the Ap horizon and not visible in the E horizon (Fig. 2). In the Ap and Bt horizons, the 2:1 clays show a diffuse reflection between 1.0 and 1.1 nm after heating at 520°C (Fig. 2). The presence of a small amount of anatase is indicated by the reflection 0.351 nm (Fig. 1). The quartz content of the <2µm fraction ranged from 253 to 345 g kg–1 (Table 3).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 1. X-ray diffractograms (Co K, values of reflection in nm) of the grounded <2-mm fraction: tilled topsoil (Ap horizon), 25- to 35-cm subsoil (E horizon), and 40- to 50-cm subsoil (Bt horizon).
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2. X-ray diffractograms (Co K, values of reflection in nm) of the oriented Mg-saturated <2-µm fraction (Mg N), ethylene-glycol-treated Mg-saturated <2-µm fraction (Mg EG) and 520°C heated K-saturated <2-µm fraction (K 520°C) in the tilled topsoil (Ap horizon), 25- to 35-cm subsoil (E horizon), and 40- to 50-cm subsoil (Bt horizon).
|
|
Scanning Electron Microscopy
The SEM observations of the three horizons studied showed that all sand and silt particles were quartz except for a few very bright grains in the silt fraction (Fig. 3)
. A BESI of the Ap horizon showed a loose arrangement of sand and silt particles with interparticle voids 50 to 500 µm in diameter (Fig. 3a and 4a,b)
. In the E horizon, a BESI showed a much closer arrangement of sand and silt particles than in the Ap horizon (Fig. 3b and 4c,d), resulting in smaller interparticle voids 50 to 100 µm in diameter (Fig. 4d). In the Bt horizon, a BESI showed a looser arrangement of the sand and silt particles than in the E horizon with many large voids 300 to 600 µm in diameter corresponding locally to a very loose arrangement of the sand and silt particles (Fig. 3c). Also, the BESI showed accumulations of very fine silt particles in very close arrangement forming clusters between sand grains in all horizons (Fig. 3 and 4c,d). In the Bt horizon, clay coatings were present between sand and silt particles, forming continuous coatings in some of the interparticle voids and clay bridges between sand and silt particles (Fig. 4e,f).
View larger version (86K):
[in this window]
[in a new window]
|
Fig. 3. Back-scattered electron scanning images: (a) topsoil (Ap horizon), (b) 25- to 35-cm subsoil (E horizon), and (c) 40- to 50-cm subsoil (Bt horizon). Voids occupied by resin are black, homogeneous gray areas are quartz grains, very light particles in the silt range size are much heavier mineral than quartz, and heterogeneous gray areas are fine silt and clay particles and associated porosity (bar scale = 1 mm).
|
|
View larger version (191K):
[in this window]
[in a new window]
|
Fig. 4. Back-scattered electron scanning images: (a,b) topsoil (Ap horizon), (c,d) 25- to 35-cm subsoil (E horizon), and (e,f) 40- to 50-cm subsoil (Bt horizon). Voids occupied by resin are black, homogeneous gray areas are quartz grains, and heterogeneous gray areas are fine silt and clay particles and associated porosity (bar scale = 100 µm).
|
|
Mercury Porosimetry
The total pore volume (Vp,t) measured by mercury porosimetry (Vp,m) ranged from 0.162 to 0.224 cm3 g–1, with those values corresponding to the E and Ap horizons, respectively (Table 4). Cumulative intrusion curves and their derivative curve showed that Vp,m resulted from the contribution of three classes of pores that were denominated Pores A, B, and C (Fig. 5) . The class of Pores A had a modal equivalent-pore diameter (De,A) of 42 to 65, 28, and 53 µm for the Ap, E, and Bt horizons, respectively (Table 4). This class of pores contributed 83 to 85% to Vp,m. Analysis of the curves also showed a secondary entry of mercury corresponding to the class of pores B with De,B ranging from 0.4 to 0.8 µm. This class of pores contributed for 10 to 13% to Vp,m. Finally, for pores with De < 0.03 µm there was a third entry of mercury corresponding to the class of pores C, but the derivative curve showed that the slope increased until De = 0.006 µm, which were the smallest pores investigated in mercury porosimetry.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 5. Mercury intrusion results recorded for the 25- to 35-cm subsoil (E horizon): (a) mean cumulated pore volume curve, (b) derivative curve.
|
|
Resistance to Penetration and Bulk Density
Soil strength ranged from 0.36 to 0.52 MPa in the E horizon, from 0.21 to 0.40 MPa in the Bt horizon, and from 0.29 to 0.37 MPa in the Ap horizon when the water content ranged from 0.085 to 0.030 g g–1, from 0.050 to 0.099 g g–1, and from 0.028 to 0.070 g g–1 in the same horizons, respectively (Fig. 6)
. There was a strong relationship between penetration resistance and water content in the E (R2 = 0.86; P = 0.001) and Bt (R2 = 0.82; P = 0.001) horizons and a weaker relationship in the Ap horizon (R2 = 0.59; P = 0.01). Within the common range of water content, penetration resistance was greater in the E horizon than in the Bt horizon at a given water content (Fig. 6). The Db was lower in the Ap horizon than in the subsoil, but results showed similar Db in the E and Bt horizons (Table 2).
|
DISCUSSION AND CONCLUSION
|
|---|
Characteristics of the Mineral Phase
Results showed similar skeleton grain-size distribution between the Ap, E, and Bt horizons, and thus homogeneity of the material in which the soil developed (Table 2). The XRD (Fig. 1) and BESI (Fig. 3 and 4) showed that the sand and silt particles were predominantly quartz with very small amounts of rutile and anatase. The XRD diffractograms showed also that a significant part of the <2-µm fraction was quartz (Table 3). The high quartz content of the <2-µm fraction was not reported by Yoothong et al. (1997) in their study of the clay mineralogy of Thai soils, but they did not sample sandy soils because they were considered as clay-free soils. However, the presence of quartz in the <2-µm fraction was reported by several authors in sandy soils of Northern Thailand, but it was not quantified (Miura et al., 1992; Kheoruenromne et al., 1998). Miura et al. (1992) suggested that a decrease in the quartz clay-sized content with depth might result from the vertical translocation of the clays downwards but that does not explain the origin of the quartz clay-sized particles (Table 3). A high quartz content in the <2-µm fraction was possibly a result of intense weathering in silty soils developed in the Red River valley in Vietnam (Hardy, 1992), but it remains unclear whether the quartz clay-sized particles were inherited from the parent material or if they formed in the soil by breakdown of the silt and sand particles. The high quartz content in the <2-µm fraction might also result from a surface process like wind transport. Observations via SEM of sand and silt grains collected in sandy soils from northeast Thailand showed evidences of wind transport (A. Prone, 2001, personal communication).
Calculation of the clay mineral content gave 24, 36, and 66 g of clay minerals kg–1 oven-dried soil in the Ap, E, and Bt horizons, respectively. The greater clay mineral content in horizon Bt is consistent with the clay coatings shown on the BESI (Fig. 4e,f). The presence of these coatings could indicate vertical illuviation of clay as earlier shown by Miura et al. (1992) in other sandy soils from northeast Thailand. The presence of smectite in very small quantities with kaolinite was recorded by Miura et al. (1992) in sandy soils of northeast Thailand. Our results show that because of the diffuse reflection between 1.0 and 1.1 nm after heating at 520°C, the 2:1 clays appear to have some octahedral material partially filling the interlayer space and thus preventing its collapse after heating (Fig. 2). The swelling 2:1 clays might be hydroxy-aluminous smectites resulting from neoformation in the lower subsoil, possibly in relation to fluctuations of the water table. The low pH (4.8–5.6) and high exchangeable aluminum content reported for the Nam Phong Series are conditions that favor the formation of hydroxy-aluminous smectites when water concentrates during the dry season (Miura et al., 1992).
Porosity and Fabric
We calculated the Vp,t (in cm3 g–1) in every horizon with the Db measured in the field and 2.65 g cm–3 as particle density (Table 2). Thus Vp,t was 0.299, 0.216, and 0.220 cm3 g–1 in the Ap, E, and Bt horizons, respectively. In the three horizons studied, Vp,t was higher than Vp,m and the difference decreased with depth: 0.075, 0.054, and 0.028 cm3 g–1 for the Ap, E, and Bt horizons, respectively (Table 4). This difference was attributed to large voids related to biological activity (mainly termites and root channels). Such voids were present within the core samples used for Db determination, but were poorly represented within the smaller clods used for mercury porosimetry.
In mercury porosimetry, two classes of pores were usually recorded for sandy soils and they correspond to different levels of fabric (Fiès, 1984; Coulon and Bruand, 1989): (i) a class of pores that results from the arrangement of the sand particles and having 5 < De < 250 µm, and (ii) a class of pores with De < 0.05 µm that results from the arrangement of the clay particles. Our results showed three classes of pores A, B, and C for the sandy soil studied. The class of pores A results from the arrangement of the sand particles alone as earlier reported for sandy soils (Fiès, 1984; Coulon and Bruand, 1989; Fiès and Bruand, 1998). Indeed, the sharp peak on the derivative curve evidences the lack of biological and interaggregate pores in Vp,A (Fig. 5b). The class of pores B was never encountered in sandy soils but only in artificial clay-silt-sand mixtures (Fiès and Bruand, 1998). According to the results reported with artificial mixtures, the class of pores B would correspond to the pores within the clusters of fine material that are present between the sand particles (Fig. 4a–d). The clusters are made up of the very fine, silt-sized quartz particles shown on the BESI and would also incorporate the quartz clay-sized particles revealed by the XRD of the <2-µm fraction. Finally, in agreement with the results reported on soils varying in texture (Fiès, 1984, 1992; Richard et al., 2001; Balbino et al., 2002), the class of pores C results from the packing of the clay particles within clay coatings on sand particles. However, clay particles were too small to enable measurement of De,C by use of mercury porosimetry alone.
The smaller Vp,A in the E horizon than in the Bt horizon despite similar particle-size distributions (Table 2) indicates a closer arrangement of the sand particles in the E horizon. This was also shown by a BESI (Fig. 3). This is consistent with the value of De,A in the E horizon that was approximately one-half of De,A in the Bt horizon. In the Ap horizon, Vp,A was greater than in the E and Bt horizons, indicating a looser arrangement of the sand particles than in the subsoil. This was consistent with observations by SEM (Fig. 3). Results also showed that Vp,B was similar in the three horizons, indicating similar packing of the silt particles among sand particles. Finally, the Ap and E horizons had smaller Vp,C than in the Bt horizon, which is consistent with the lower clay content of those horizons (Table 2). Thus, despite similar Db in the E and Bt horizons, there was a closer arrangement of the sand and silt particles in the E horizon.
Consequence for Resistance to Penetration
The greater penetration resistance in the E horizon than in the Bt horizon was not related to the presence of cements causing the cohesion of sand and silt particles. Indeed, observation in optical microscopy showed that treatment with sodium hexametaphosphate alone resulted in complete dispersion of the horizons studied. Greater penetration resistance in the E horizon was not related to the clay mineral content because clay content was less in the E horizon (36 g kg–1) than in the Bt horizon (66 g kg–1). Penetration resistance was not related to a variation in the clay fabric since clay coatings were not present in the E horizon as indicated by SEM. The resistance to penetration may vary independently of the Db because the latter integrates a pore volume that differs from the pore volume of the effective matrix actually affecting the resistance to penetration (Sojka et al., 2001). Indeed, mercury porosimetry showed different porosity resulting from the arrangement of elementary particles between the E and Bt horizons. There was a closer packing of sand particles in the E horizon than in the Bt horizon as indicated by Vp,A; Vp,A was 16% smaller in the E horizon compared with the Bt horizon. The closer arrangement of sand particles in the E horizon than in the Bt horizon was also evident from SEM (Fig. 3b,c). The Db measured in the field included pores with De > 360 µm that were not inferred with mercury porosimetry, thus explaining the difference between the porosity that was deduced from Db and the porosity measured with mercury porosimetry. These pores with De > 360 µm were the discontinuous channels and chambers seen in the field. Our results showed that they were more numerous in the E horizon than in the Bt horizon, thus compensating for the difference in porosity for pores De 
360 µm observed between the two horizons.
We computed the Db of the undisturbed samples that were studied in mercury porosimetry using 2.65 g cm–3 as particle density. We obtained 1.66, 1.86, and 1.76 g cm–3 for the Ap, E, and Bt horizons, respectively. These values are more representative of the physical properties measured by penetration resistance and, consequently, those experienced by roots when they explore the soil between the discontinuous channels and chambers seen in the field. The Db computed from mercury porosimetry measurements may be regarded as effective Db. This effective Db better explains the different sensitivity of the resistance to penetration to the influence of water between horizons than the Db that is measured on the whole horizon. Similar to work by Sojka et al. (2001) for a coarse-silty soil, our results also show the need to group values of resistance to penetration by depth intervals even when there is no or small variation of soil composition. By grouping the results by depth intervals, there would be less variation of the elementary fabric and then less variation of the sensitivity of the resistance to penetration to the influence of water.
The close arrangement of sand particles in the E horizon could result from compaction by traffic because of the high sensitivity of sandy soils to rearrangement of the elementary fabric (Coulon and Bruand, 1989). Thus, the E horizon can be interpreted as tillage pan that is a common feature in sandy soils (Vepraskas, 1994). Compaction by traffic that was responsible for the elementary fabric in the E horizon should have also affected the macroscopic pores, that is, discontinuous channels and chambers, that were observed in the field and quantified by comparing porosity in cores and clods. As for the small number of biological pores recorded in the E horizon, this condition may have resulted from a dynamic equilibrium between processes responsible for their destruction and their formation.
|
ACKNOWLEDGMENTS
|
|---|
This work formed part of a research program funded by the Department of Technical and Economical Co-operation (DTEC), the Land Development Department (LDD), and the Institut de Recherche pour le Développement (IRD). The authors gratefully acknowledge Nijaporn Kunklang and Emmanuel Bourdon (IRD, Bangkok) for their help in the field and Olivier Josière, Hervé Gaillard, and Christian Le Lay (INRA, Orléans) for their help in the laboratory.
Received for publication June 4, 2001.
|
REFERENCES
|
|---|
- Balbino, L.C., A. Bruand, M. Brossard, M. Grimaldi, M. Hajnos, and M.F. Guimaraes. 2002. Changes in porosity and microaggregation in clayey Ferralsols of the Brazilian Cerrado on clearing for pasture. Eur. J. Soil Sci. 53:219–230.
- Bruand, A., I. Cousin, B. Nicoullaud, O. Duval, and J.C. Bégon. 1996. Backscattered electron scanning images of soil porosity for analysing soil compaction around roots. Soil Sci. Soc. Am. J. 60:895–901.[Abstract/Free Full Text]
- Busscher, W.J., J.R. Frederick, and P.J. Bauer. 2000. Timing effects of deep tillage on penetration resistance and wheat and soybean yield. Soil Sci. Soc. Am. J. 64:999–1003.[Abstract/Free Full Text]
- Busscher, W.J., J.R. Frederick, and P.J. Bauer. 2001. Effect of penetration resistance and timing of rain on grain yield of a narrow-row corn in a coastal plain loamy sand. Soil Tillage Res. 63:15–24.
- Ciesielski, H.R., and T. Sterckeman. 1997. Determination of cation exchange capacity and exchangeable cations in soils by means of cobalt hexamine trichloride. Effects of experimental conditions. Agronomie 17:1–7.
- Coulon, E., and A. Bruand. 1989. Effects of compaction on the pore space geometry in sandy soils. Soil Tillage Res. 15:137–152.
- Craig, I.A., and U. Pisone. 1988. Overview of rainfed agriculture in north-east Thailand. p. 24–37. In C. Pairintira et al. (ed.) Soil water and crop management systems for rainfed agriculture in northeast Thailand. Khon Kaen Univ., Thailand.
- Duncan, M.M., and D.P. Franzmeier. 1999. Role of free silicon, aluminium, and iron in fragipan formation. Soil Sci. Soc. Am. J. 63:923–929.[Abstract/Free Full Text]
- Fiès, J.C. 1984. Analyse de la répartition du volume des pores dans les assemblages argile-squelettes. Comparaison entre un modèle d'espace textural et les données fournies par la porosimétrie au mercure. Agronomie 4:891–899.
- Fiès, J.C. 1992. Analysis of soil textural porosity relative to skeleton particle size, using mercury porosimetry. Soil Sci. Soc. Am. J. 56:1062–1067.
- Fiès, J.C., and A. Bruand. 1998. Particle-size distribution and organisation of the textural porosity: Study of clay-silt-sand mixtures. Eur. J. Soil Sci. 49:557–567.
- Hardy, M. 1992. X-Ray diffraction measurement of the quartz content of clay and silt fractions in soils. Clay Miner. 27:47–55.[Abstract]
- Hardy, M., M. Jamagne, F. Elsass, M. Robert, and D. Chesneau. 1999. Evolution of phyllosilicates of silty fractions of a Luvisol developed on loessial material of the Paris basin (France). Eur. J. Soil Sci. 50:443–456.
- Hartmann, C., R. Poss, and V. Singhatat. 1999. Soil compaction and plant growth in Northeast Thailand: The case of the Nam Phong soil series. Proc. Workshop Ecoregional Approaches to Natural Resource Management in the Korat Basin. 26–29 Oct. 1999. IRRI and Khon Kaen Univ., Khon Kaen, Thailand.
- Jones, C.A. 1983. Effect of soil texture on critical bulk densities for root growth. Soil Sci. Soc. Am. J. 47:1208–1211.[Abstract/Free Full Text]
- Kheoruenromne, I., and A. Suddhiprakarn. 1984. Ecology, classification and effect of management on selected sandy soils in Thailand. p. 208–222. In Ecology and management of problem soils in Asia. Food and Fertilizer Technology Center for the Asian and Pacific Region, Taiwan.
- Kheoruenromne, I., A. Suddhiprakarn, and P. Kanghae. 1998. Properties, environment and fertility capability of sandy soils in northeast plateau, Thailand. Kasetsart J. 32:355–373.
- Lamotte, M., A. Bruand, F.X. Humbel, A.J. Herbillon, and M. Rieu. 1997a. A hard sandy-loam soil from Northern Cameroon. I. Fabric of the groundmass. Eur. J. Soil Sci. 48:213–225.
- Lamotte, M., A. Bruand, D. Ohnenstetter, P. Ildefonse, and G. Pédro. 1997b. A hard sandy-loam soil from Northern Cameroon. II. Geochemistry and mineralogy of the bonding agent. Eur. J. Soil Sci. 48:227–237.
- Lindbo, D.L., and P.L.M. Veneman. 1989. Fragipans in the northeastern United States. p. 11–31. In N.E. Smeck and E.J. Ciolkosz (ed.) Fragipans: Their occurrence, classification, and genesis. SSSA Spec. Publ. 24. SSSA, Madison, WI.
- Mitsuchi, M., P. Wichaidit, and S. Jeungnijnirund. 1986. Outline of soils of the Northeast Plateau, Thailand: Their characteristics and constraints. Tech. Paper No. 1. Agric. Dev. Res. Center in Northeast Thailand, Khon Kaen.
- Miura, K.T., T. Tulaphitak, and K. Kyuma. 1992. Pedogenic studies on some selected soils in northeast Thailand. I. General soil characteristics. Soil Sci. Plant Nutr. (Tokyo) 38:485–493.
- Moorman, F.R., S. Montrakun, and S. Panichapong. 1964. Soils of northeastern Thailand: A key to their identification and survey. Soil Survey Div., Dep. of Land Dev., Ministry of Natl. Dev., Bangkok, Thailand.
- O'Donnell, A.G., J.K. Syers, P. Vichiensanth, P. Vityakon, M.A. Adey, P. Nannipiery, W. Sriboonlue, and A. Suwanarit. 1994. Improving the agricultural productivity of the soils of north-east Thailand through soil organic matter management. p. 192–205. In J.K. Syers and D.L. Rimmer (ed.) Soil science and sustainable land management in the tropics. Cab Int. and British Soc. of Soil Sci., Wallingford, UK.
- van Ouwerkerk, C., and P.A.C. Raats. 1986. Experiences with deep tillage in The Netherlands. Soil Tillage Res. 7:273–283.
- Poss, R., P. Dreshel, D. Brunet, V. Singhatat, and S. Krutkun. 1998. Soil organic matter as main factor for corn yield on a fertilized Arenic Acrisol of Northeast Thailand. In Proc. Int. Soil Sci. Congr., Montpellier, France. [CD-ROM]. Assoc. Française pour l'Etude du Sol, Montpellier, France.
- Ragland, J., and L. Boonpuckdee. 1987. Fertilizer responses in Northeast Thailand: Literature review and rationale. Thai J. Soils Fert. 9:81–92.
- Richard, G., I. Cousin, J.F. Sillon, A. Bruand, and J. Guérif. 2001. Effect of compaction on the porosity of a silty soil: Consequences on unsaturated hydraulic properties. Eur. J. Soil Sci. 52:49–58.
- Sakurai, K., S. Kozasa, T. Yuasa, B. Puriyakorn, P. Preechapanya, V. Tanpibal, K. Muangnil, and B. Prachaiyo. 1996. Changes in soil properties after land degradation associated with various human activities in Thailand. Soil Sci. Plant Nutr. (Tokyo) 42:81–92.
- Singhatat, V. 1996. Management of acid soils for sustainable upland cropping at Natkhon Ratchasima, Northeast Thailand. p. 59–62. In The management of acid soils in Southeast Asia (IBSRAM/ASIALAND). IBSRAM Network Doc. 16. Int. Board for Soil Res. and Manage., Thailand.
- Smeck, N.E., M.L. Thompson, L.D. Norton, and M.J. Shipatalo. 1989. Weahering discontinuities: A key to fragipan formation. p. 99–112. In N.E. Smeck and E.J. Ciolkosz (ed.) Fragipans: Their occurrence, classification, and genesis. SSSA Spec. Publ. 24. SSSA, Madison, WI.
- Sojka, R.E., W.J. Busscher, and G.A. Lehrsch. 2001. In situ strength, bulk density, and water content relationships of a Durinodic Xeric Haplocalcid soil. Soil Sci. 166:520–529.
- Vepraskas, M.J. 1994. Plant response mechanisms to soil compaction. p. 263–287. In R.E. Wilkinson (ed.) Plant–Environnement Interactions. Marcel Dekker, New York.
- Vepraskas, M.J., and G.S. Miner. 1986. Effects of subsoiling and mechanical impedance on tobacco root growth. Soil Sci. Soc. Am. J. 50:423–427.[Abstract/Free Full Text]
- Vityakon, P., and P. Keerati-Kasikorn. 1987. Farming systems and soils in Northeast Thailand. p. 361–383. In Rep. of the first Regional Seminar on Soil Management under Humid Conditions in Asia, Khon Kaen, Phitsanulok, Thailand. 13–20 Oct. 1986. Int. Board for Soil Res. and Management, Bangkok, Thailand.
- Yoothong, K., L. Moncharoen, P. Vijarnson, and H. Eswaran. 1997. Clay mineralogy of Thai soils. Appl. Clay Sci. 11:357–371.
This article has been cited by other articles:
|
|
|
|
 
E. V. Vazquez, J. P. Ferreiro, J. G. V. Miranda, and A. P. Gonzalez
Multifractal Analysis of Pore Size Distributions as Affected by Simulated Rainfall
Vadose Zone J.,
May 27, 2008;
7(2):
500 - 511.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
), E (
), and Bt (
) horizons.