Structural collapse and strength of some Australian soils in relation to hard setting: I. Structural collapse on wetting and draining

The surface structure of many Australian red and red-brown earths frequently collapses (slakes) when dry, disturbed aggregates are wetted by rain or irrigation. The resulting fine matrix sets, on drying, to a strong, cohesive layer of up to 200 mm thick (hard setting). We investigated the mechanism of collapse and the extent to which the structure of aggregate beds Iron hard setting and non-hard setting soils collapsed when wetted by quick flooding or slowly with water at a suction of 200 mm, then drained in sequential steps of increasing suction and finally dried at 40°C. After flood wetting, but before draining, no collapse was observed due to the small effective stress prevalent in the flooded beds. After suction wetting, some collapse was measured owing to the effective stress (approximately 1.4 kPa) from the applied suction. On draining, flood-wetted beds collapsed extensively (volume strain >0.20), largely due to the disappearance of large pores (>75 μm diameter). Suction-wetted beds collapsed less (volume strain <0.16) and retained more large pores. Hard setting soils collapsed more following both flood and suction wetting (volume strain >0.20 and 0.10, respectively), while non-hard setting soils did not collapse as extensively (volume strain <0.16 and 0.09, respectively). Results indicate that the mechanism causing collapse was independent of wetting method and involved two steps: (i) slaking of aggregates on wetting, and (ii) collapse of the aggregate bed on draining as a result of development of effective stress within the beds.     

干湿效应下崩岗区岩土抗剪强度衰减非线性分析

发育于花岗岩的崩岗侵蚀区红土受干湿变化影响显著。通过室内直剪试验,研究了不同干湿效应对崩岗侵蚀区岩土抗剪强度衰减的影响。试验处理采用5种干湿效应水平(风干48h、风干24h、自然含水率、浸30s和浸60s)。结果显示:土壤黏聚力c和内摩擦角φ随干湿变化呈非线性衰减趋势,当土壤含水率13%左右时,对应的抗剪强度指标出现峰值;峰值强度前符合线性递增规律,峰值强度后符合一阶指数衰减规律。在风干阶段,抗剪强度主要受裂隙性影响,而在增湿阶段,基质吸力是影响抗剪强度的主要因素;探讨了干湿循环效应对崩岗侵蚀发育的影响。     

The creation of longitudinal cracks in shrinking soils to enhance seedling emergence. Part I. The effect of soil structure

Stand establishment in crusting soils is one of the most critical stages in the production of crops with delicate seedlings. This becomes more difficult in hard‐setting soils of arid lands where dispersion of aggregates due to irrigation results in the formation of a hard layer as the soil dries from the surface downwards and impedes seedling emergence. However, seeds planted near to natural cracks manage to emerge through them. The aim of this study was to improve seedling emergence of irrigated crops in arid and semi‐arid conditions by devising methods to create longitudinal cracks in the vicinity of the seed rows during the subsequent drying phase. Laboratory experiments were conducted in soil boxes containing air‐dried clay soils to investigate the influence of different mechanical effects on the cracking pattern after flooding. Promising results were achieved by compacting a 7–15‐mm wide strip along the seed row. Monitoring the effect of compaction on water release characteristics and tensile strength of soil revealed that the greater water loss of the surrounding non‐compacted soil caused a suction gradient towards the points of lower water content resulting in movement of water and particles towards the drier zones. The compacted soil with a greater tensile strength did not permit the flow of water and particles to the loose soil and a discontinuity occurred. A field experiment in central Iran proved the feasibility of the technique in a semi‐arid area. This was achieved by lightly compacting a narrow strip of soil by applying 17–35 kg on a 22‐mm wide, 558‐mm diameter wheel covered by a layer of flexible rubber which ran over the pre‐compacted soil.     

THE EFFECT OF HYSTERESIS OF PORE-WATER ON THE HYDRAULIC CONDUCTIVITY

It is pointed out with the help of the Independent Domain Theory, that a fraction of the pore-water constituting a volume ΔV is contained in pores not common to both the wetting and drying states and that this volume in a wetting state is contained in pores whose draining radii are greater than those of the pores containing the water volume ΔV in the drying state. On the above basis it is suggested that hysteresis may affect the hydraulic conductivity: firstly, through differences of the radii of the pores containing the water volume ΔV, which tend to endow a wetting state with a conductivity greater than that of an equally wet but drying state; secondly, through differences of the coordinates of the above-mentioned pores, which may influence the paths of flow and consequently the hydraulic conductivity, and will have either a similar or opposite effect. It is concluded, therefore, that the conductivity in the wetting state can be greater, equal, or even less than that in the drying state, depending on the particular porous body.     

Structure properties and pore dynamics in aggregate beds due to wetting–drying cycles

Aggregate hierarchy and porosity changes in aggregate beds as a consequence of wetting–drying cycles were studied in two Andisols and one Mollisol from Chile, collected at two depths. Bulk density and indirect tensile strength were measured in aggregates of different sizes. Aggregate beds were prepared in cylinders with two size classes. Six wetting–drying cycles between 0 and –60 hPa were applied. Bulk density (Db) of soil matrix was controlled after each cycle, and the macroporosity was calculated. A repellency index was measured in one of the Andisols. In addition, also the air permeability was measured after the sixth cycle. It could be proofed, that the aggregate strength is an appropriate parameter to evaluate the aggregate hierarchy, and this parameter is also more sensitive than aggregate Db to discriminate between the effects of land‐use intensity. Aggregate strength is furthermore well correlated with changes in pore water pressure and can be applied to relate strength values with aggregate development level. Only if the predrying exceeds pF > 3.0, aggregate strength correlates with Db. The more pronounced is the land‐use, the higher is the increase of Db values for aggregate beds. The decrease of coarse porosity during wetting–drying cycles can be explained by mass differences between saturated and equilibrated water conditions that considers the water around aggregates and within the contact area. Nevertheless, the relation of relative macroporosity change, calculated by P_exped where D_agg is the Db measured by clod method, and the relative Db change, is useful to explain possible presence of coarse pores inside the aggregates. The newly formed porosity prevents the water repellency, but after six cycles of drying, the repellency index increased in the topsoil while we could detect a decrease in the subsoil samples (under defined conditions in the laboratory) which we assume to be caused by microbial activity. The approaching of aggregates by drying cycles generates in Andisols a reduced area to air fluxes, with low values of air permeability.     

Hysterese der Wasserspannungskurve in organogenen Böden

Soil water hysteresis in organic soils In 20 undisturbed samples of peat, muck, peaty soil and humic sand, the hysteresis of the moisture characteristic by soil suction from 0 to –9,8 kPa and from 0 to –29,4 kPa was measured. The greatest hysteresis was found in peat. Suction values from –1 to –3 kPa caused differences of 0,076 cm³/cm³ in water content during drying and wetting. In muck with low ash content and in humic sand the hysteresis was smaller. In peaty soil and muck with high ash content (> 40 %) it diminished to twice or three times lower values in comparison with those in peat. In organic soils the hysteresis decreases clearly by soil suction higher than –6 kPa. Repetition of drying and wetting gave a diminution (ca. 30 %) of the hysteresis loop in peat and displaces it towards the field of lower water content. The hysteresis of muck did not change considerably. The change of the porosity structure (pore ϕ > 30 μm) during the repetition of drying and wetting indicate that some soil shrinkage has occured.     

Soil mechanical properties as influenced by exchangeable cations

The tensile strengths of dried, undisturbed aggregates and of dried, artificial aggregates and the shear strengths of freshly-moulded, moist soils were investigated as functions of their contents of exchangeable cations. Tensile strength increased with increasing exchangeable sodium and decreased with increasing exchangeable calcium. Shear strength of moist soil, as assessed through the Atterberg limits, increased with increasing exchangeable calcium, potassium and sodium. The influence of exchangeable magnesium was variable in that it tended to increase the tensile strength of the soils when dry, whereas it tended to decrease the shear strength of one of the sets of soils in the moist state. It is concluded that the exchangeable cations which give rise to greater repulsion between, and dispersion of, clay particles in water also give rise to greater strength in the dried soil. It is proposed that this is because greater particle repulsion in the soil water during drying enables particle rearrangements to take place more readily and this results in denser packing arrangements and increased strength.     

基于模糊综合评判的崩岗重力侵蚀评价方法

发育于花岗岩出露区的典型崩岗岩土受干湿效应影响,致使坡体在重力作用下产生崩坍,这是崩岗形成后最主要的侵蚀形式。通过对原状土进行5种干湿水平处理,采用直剪试验研究干湿效应下崩岗岩土抗剪强度及其指标变化规律。结果显示：从风干到增湿过程,抗剪强度及其指标衰减明显,粘聚力C下降幅度最大为92％,而内摩擦角9最大降幅为48％。随着垂直压力递增,抗剪强度增大。此外,基于不同干湿效应下岩土抗剪强度指标变化规律,运用模糊综合评判法对不同干湿水平、干湿阶段崩岗坡体稳定性等级（重力侵蚀）进行评价,并讨论了完善该模型的途径,可为崩岗危害性分级、针对性治理提供技术支撑。     

Microbial biomass response to a rapid increase in water potential when dry soil is wetted

The proportion of microbial biomass-C released into the soil environment following rapid water potential increase was quantified in two soils using a modified chloroform-fumigation biomass assay. Dry samples were isopiestically equilibrated to −2.8 and − 6.9 M Pa and then wetted to field capacity with either H₂O or KCl solutions. The KC1 solutions wetted the soils without altering total soil water potential. The biomass-C released by water potential increase ranged from 17 to 70% of total, depending on the soil, the magnitude of the increase, and the method of calculation. In both soils, a greater proportion of biomass-C was released following a 6.9 MPa than a 2.8 M Pa increase. Biomass-C release was also demonstrated by an increase in soluble organic C in leachates of soils subjected to rapid wetting. Respiration of biomass-C mobilized by water potential increase exceeded respiration of biomass-C made available by preceding desiccation, thereby comprising a significant component of the pulse of respiration observed following wetting of dry soil. Water potential increases associated with the wetting of dry soil may be a major catalyst for soil C turnover.     

Structural change induced by wetting and drying in seedbeds of a hardsetting soil with contrasting aggregate size distribution

Hardsetting has been related to two main processes: (i) development of strength when the soil is still moist (–100 kPa) due to matric suction acting within interparticle and interaggregate bridges and (ii) temporary cementation of dry soil by poorly ordered silica and aluminosilicates. In both cases, hardsetting of a seedbed should depend on geometric aspects of macro- and micro-structure. This study deals with aggregate breakdown and/or deformation on wetting and with the structural changes which follow on drying. Repacked seedbeds of a hardsetting red-brown earth were wetted by capillary rise. Seedbeds with a coarse and a fine aggregate size distribution were examined. Before and after wetting, the bulk density of the seedbeds was measured at 5-mm increments using a gamma probe. Replicate samples were air dried, gamma scanned and impregnated. Binary images of pore space (> 107 μm) of vertical faces were used to generate depth functions of structure attributes, including macroporosity. Bulk density measurements combined with image analysis resulted in successful structure characterization. Thin-section observations were used to interpret the quantified changes in terms of physical processes. In both coarse and fine seedbeds the physical processes which determined structural change occurred upon wetting rather than on drying. Coalescence of aggregates under plastic conditions, partly due to overburden pressure and enhanced by microcracking and partial slaking, occurred at the bottom of the coarse seedbed. In the fine seedbed, the fine material agglomerated due to matric suction. These different processes led to rather similar, microporous microstructures which exhibited similar strength properties after drying.     

Shrinkage properties of differently managed clay soils in Finland

Soil cracking is a well-known phenomenon, also seen in clay soils in the boreal climatic zone. This study was carried out to quantify soil shrinkage properties in six differently managed clay soils in Finland (Vertic Cambisols, 51% clay). Cylinder samples (100 cm³) were taken in spring from two depths (0–5 and 5–10 cm), then saturated with water and dried as a function of applied suction. The heights of the sample were measured after each drying step and the volume of soil was calculated assuming isotropic shrinkage. The volume loss by shrinkage at a suction of −50 kPa was 1.6–3.8% and the total shrinkage was 5.2–10.5% of the total soil volume, respectively. All shrinkage curves showed structural shrinkage which occurred in the matric potential range from saturation to around −6 kPa. The shrinkage curves were characterized by minor proportional and wide residual shrinkage zones. Eight of twelve sites showed a steeper shrinkage in the proportional shrinkage zone than the theoretical 1:1 line. Large slope values, up to 3.0, reflect the collapse of inter-aggregate pore space due to shrinkage pressure. The results indicate significant particle rearrangement and structural changes, e.g. structural collapse and changes in inter-aggregate pore space due to shrinkage pressure. Continuous water saturation and variable periods of freezing between spring and autumn are mostly responsible for soil weakness against increasing effective stress as soil dries. It is presumed that shrinkage behaviour will change substantially with increases in drying and wetting cycles.     

Einfluss des Porenwasserdruckes auf die Zugfestigkeit von Bodenproben

Effect of pore water pressure on tensile strength Direct tensile testing with measurements of the pore water suction was used to investigate the relationship between tensile strength and suction. The tests were conducted on a till and a clayey soil, both homogenized. A closer view is focused on the relationship between material strain and the development of suction. Beyond, the factor χ of the effective stress equation for unsaturated soils by Bishop (1959), which was calculated by the data of tensile strength and corresponding matric suction is compared to the volumetric χ of the tested soil specimens. It could be shown, that the pore water pressure changes with strain. Therefore, not the initial suction of a soil is relevant for its failure but the actual one that can be measured in the failure zone at the moment of fracture. In addition the application of the volumetric χ in the effective stress equation compared to the χ derived from tensile testing leads to an 1.6 to 2.8 fold overestimation of the contribution of matric suction to soil tensile strength.     

The effect of texture on strength of southeastern coastal plain soils

The effect of texture on soil strength was analyzed for 17 soils, mainly from the Southeastern Coastal Plains. All samples were tested with a 5-mm, flat-tipped probe after equilibration at 100 kPa of soil-water tension. Both mechanical compression and water consolidation (compaction by wetting and drying) were used to compact the soils. Probe resistance of water-consolidated samples was significantly affected by texture. In fact, bulk density or probe resistance of soils compacted by methods of constant compactive force or constant methodology correlated well with texture for Coastal Plain soils.

A secondary effect of texture involved a significant, positive correlation of silt with probe resistance for sandy soils low in organic matter. Bulk densities at root-restricting conditions (2 MPa) compared well with bulk density values obtained by water consolidation and the Random Packing Model of Gupta and Larson, all of which were significantly correlated to texture.     

Modelling the effect of water distribution and hysteresis on air-filled pore space

Soil pore networks have a complex geometry, which is challenging to model in three dimensions. We use a Boolean model of pore space that has proved useful in modelling gas diffusion in dry structures to investigate the distribution of water in this pore space and to quantify the effects on pore connectivity to the soil surface. We first show how total porosity in dry soil influences connectivity via the percolation threshold. Then we show that our model simulation of the ‘ink-bottle effect’ can account for much of the hysteresis of the soil water. The differences in distribution of water between wetting and drying result in maintaining greater connectivity of the air-filled pore space during drying than during wetting. Hysteresis is large at small total porosities and slowly declines as porosity increases. During wetting much pore space is blocked when more than 40% of the pore space is filled with water, although during drying all non-isolated air-filled pores are connected to the surface. Even when soil is allowed to wet to near saturation, there are rapid increases in pore connectivity during drying, which may explain, for example, rapid increases in production and emission of nitrous oxide in soils near saturation.     

Residual effects of additions of calcium compounds on soil structure and strength

Soil structure was measured from the roughness of soil fracture surfaces created by breaking unsaturated soil clods under tensile stress. Soil clods were collected in 1986 and 1987 from field plots to which calcium compounds had been applied in 1980 (to improve degraded structure) at rates of 0, 2, 4, 10, 14 and 20 t ha⁻¹, and which had not been disturbed since the initial establishment of pasture. A residual effect on the soil structure was sought. In 1987, samples were equilibrated at different water suctions in the laboratory, and physical and chemical measurements including tensile strength, penetrometer resistance, exchangeable cations and dispersible clay were made.

Exchangeable cation and dispersible clay values did not correspond with the amounts of calcium added in 1980, indicating that most of the applied calcium had leached by the time measurements were made in 1986 and 1987. The water content at which measurements were made was the dominant factor controlling the tensile strength and penetrometer resistance of the soil clods. The water content also had a major influence on the fracture surface roughness; wetter clods had rougher tensile fracture surfaces. Because most of the applied calcium was leached by the time measurements were made, the residual effect of the calcium amendments was only detectable in the microstructure (undisturbed since 1980). Soils to which greater amounts of calcium had been added tended to have clods with smoother tensile fracture surfaces. This was attributed to stabilization by calcium of aggregates with diameters up to 0.1 mm in the soil structural hierarchy. The importance of the scale at which fracture surface roughness measurements are made is emphasized.     

PORE SIZE DISTRIBUTION IN CRITICAL POINT AND FREEZE DRIED AGGREGATES FROM CLAY SUBSOILS

Using aggregates from 2 clay soils over a range of water contents from pF 1 to oven dry, shrinkage and water release curves were measured and pore size distributions found by mercury porosimetry after critical point and after freeze drying. Freeze drying caused less shrinkage, the maximum being 6 cm³ 100 g^?1 from pF 1, but gave a large increase, up to 10 cm³ 100 g^?1, of pores in the 0.1–10 μm size range. Critical point drying produced more shrinkage from pF 1, more than half of which was attributed to loss of interlamellar water and the rest to the collapse of pores larger than 10 μm. In these clays the volume of pores of over 3 μm diameter was very small (<3 cm³ 100 g^?1) and most of the plant available water was released by collapse of narrower pores and not by pore emptying.     

干湿循环条件下重庆地区三种土壤抗剪强度的动态变化

选择广泛分布于重庆丘陵山区的黄壤、钙质紫色土和中性紫色土3种土壤,通过对室内三轴剪切试验,测定含水率和干密度交互作用对土壤抗剪强度指标的影响,在含水率和干密度对土壤抗剪强度影响分析的基础上,3土壤按各自最优含水率和干密度制作干湿循环试验土样,进行干湿循环条件下土壤抗剪强度的动态变化分析。试验结果显示：(1)在相同干密度情况下,3种土壤粘聚力c值随着含水率的增加呈现出先增加后减小的趋势,在相同土壤含水率水平下,土壤粘聚力c值随干密度增大而增大,3种土壤内摩擦角φ值在各干密度条件下均随着含水率增加呈明显减小的趋势。(2)含水率和干密度的交互作用对土壤粘聚力c值有显著影响,粘聚力c值在1.3-1.7g/cm₃干密度范围内随着干密度的增大而增大,且每一个干密度都有一个含水率与之对应,在这样一个交互作用下粘聚力c值达到最大,含水率和干密度的交互作用对内摩擦角φ值影响相对较小,同一干密度下,其φ值差异不大,随干密度的增大缓慢增大。(3)3种土壤的粘聚力c值均随干湿循环次数的增加均呈减小趋势,且前两次循环c值衰减幅度都很大,从第三次干湿循环到第五次干湿循环粘聚力c值衰减幅度很小,趋于稳定。(4)3种土壤在干湿循环后内摩擦角φ值总体呈减小趋势,但不同土壤类型间存在差异,第五次循环结束后,黄壤为24.6?,中性紫色土为22.6?,钙质紫色土为19.3?。     

HYSTERESIS IN THE MOISTURE CHARACTERISTICS OF CLAY SOIL

The relation between soil suction, degree of saturation, and the air and water permeabilities is experimentally investigated for two clay soils. Marked hysteresis can exist between saturation and permeability, but does not follow a unique trend. Thus permeability can be greater on a wetting or drying cycle depending on the structure of the clay soil. This behaviour is considered in terms of the cluster concept of soil structure, and it is concluded that any general theory of hysteresis must recognize both the macrostructure and microstructure of a soil.     

降雨条件下土壤入渗的规律研究

本文从描述非饱和土壤中水分流动的基本方程出发,对4种不同典型土壤在各种假定的均匀降雨速率下的入渗速率进行了数值模拟。结果表明,在降雨初期,由于降雨率低于入渗率,故全部雨量入渗,但当雨率大于入渗速率之后,入渗流动可以比拟为饱和水柱向非饱和区推进的柱塞流,其速率与饱和水柱长度成反比,而与分隔非饱和区与饱和水柱的湿润阵面上吸力大小成正比。如果对实际计算的吸力与Mein及Larson所推荐的理论计算吸力分别按各自最大吸力进行无量纲化,发现二者值吻合得很好。根据柱塞流这样一个概念,本文推导了决定入渗率等于雨率时的时间及该时的湿润锋面位置,并继而推导了在此以后,如果雨率永远大于入渗率情况下(并且不考虑地表有积水)入渗速率与湿润锋面位置的解析解,大大地简化了入渗率的工作量,具有实际应用的意义。     

THE UNIQUENESS OF THE MOISTURE CHARACTERISTICS

Experiments, designed to test the non-uniqueness which has been reported in the relationship between moisture content and suction in porous materials during transient flow conditions, are escribed. When a vertical column of sand was brought to uniform moisture content and uniform suction in a draining condition by percolating a constant flow through it until apparent equilibrium conditions ensued and then placed horizontally at the same time as the flow was stopped, the suction was found to decrease with time over a period of a day, indicating that the sand was now wetting up. This decrease was smaller the drier the sand. However, when the column was left in a vertical position with the water percolating through it for three hours after apparent equilibrium conditions were reached, a small adjustment of moisture content and suction was observed; when it was then placed horizontally as in the previous experiments little decrease of suction was observed. No change in suction was observed in either case if the equilibrium condition was reached in a wetting porous material. These experimental results and those of previous workers are explained by the hypothesis that, during drainage, some water is left in pores from which the air-water interface has already retreated to drain through film flow, more slowly, subsequently to the continuous water body. Additional experiments showed that the release of entrapped air may have a small effect on the uniqueness of the moisture characteristics.