Advertisement

The Role of the Coefficient of Permeability K

  • Luigi Coppola
Chapter

Abstract

In this chapter it will be analysed the downward movement of water due to gravity, favoured by a permeability (K) more or less large, whereas the ascending, generally superficial, movement will be treated in the end for particular cases where even on jungle-like, intensely forested slopes (see Fig.  5.23 in Chap. 5) it is possible that there are plastic-rotational landslides that evolve in flows.

6.1 Introduction

The seepage of water in cohesive soil is governed by two principles of different origin: the head determining the flow of water downwards under the action of gravity; the capillary force that is generated by the variation of the capillary potential from point to point. The water content of a certain area of the slope may decrease due to evaporation or to a network of tree roots, which results in increased suction force (pF) or to the migration of excess water in the wettest lithic horizons in the nearby. These conditions mean that the lower the hydraulic head in the ground, the more difficult the development of shearing processes due to landslides on a slope.

In this chapter it will be therefore analysed the downward movement of water due to gravity, favoured by a permeability (K) more or less large, whereas the ascending, generally superficial, movement will be treated in the end for particular cases where even on jungle-like, intensely forested slopes (see Fig.  5.23 in Chap. 5) it is possible that there are plastic-rotational landslides that evolve in flows.

6.2 The Effect of Ground Anisotropy on Permeability

The coefficient of permeability K indicates the capacity of seepage of water by gravity into the ground, which is much higher as the greater is the porosity of the soil (Baver 1948). Researches on this subject (Combeau and Monnier 1961; Vigneron and Desaunettes 1960) have shown for long time that permeability is subjected to the predominant influence of soil structure in relation to its fabric.

The value of the coefficient of permeability therefore depends on the characteristics of both the fluid and the soil.

Structurally chaotic and mineralogically heterogeneous clayey soils are much more permeable than those with a compact and homogeneous structure. It is also to be taken into account that the permeability of unaltered cohesive soil is influenced by the electrostatic forces present on the surface of clay platelets (see Chap.  1).

Taylor (1948), developed an equation that links the permeability coefficient K to both fluid and soil characteristics:
$$K = D_{s}^{2} \frac{\gamma }{\mu } \cdot \frac{{e^{3} }}{1 + e} \cdot c$$
where
K

Darcy’s coefficient of permeability

D s

diameter of soil particles

γ

specific weight of the fluid

μ

viscosity of the fluid

e

porosity index of the soil

c

shape factor of soil grains.

This equation is very useful for examining the factors affecting the permeability coefficient of which are first considered the ones dependent on the permanent fluid and subsequently those dependent on the soil (Table 6.1).
Table 6.1

Classification of soils based upon permeability

Permeability

Values of K (cm/s)

High

>10−1

Medium

10−1 to 10−3

Low

10−3 to 10−5

Very low

10−5 to 10−7

Impermeable

<10−7

From Terzaghi and Peck (1967)

On natural slopes, the lithic horizon of alteration, rather shallow, is generally poor in organic matter and the percentage of silt and sand is very small compared to that of the upper agricultural ground. In it, permeability decreases progressively in inverse proportion with depth. The clay layers of the underlying lithic unit, if they do not present generic discontinuities, behave as impermeable horizons that can lead the overlying ones to total saturation during wet periods. These ones, in lack of air become asphyxiating and reducing, iron partially passes to the ferrous state (marbling phenomenon), while the other chemical components take on particular properties defined as “hydromorphic”. There are two cases of temporary hydromorphism due to impermeability.
  1. (1)
    Surface hydromorphism , which occurs without a real impermeable substrate, resulting from the jamming of impermeable substances released by capillary water in very small interplatelet voids. This usually occurs at a depth of a few meters from ground level, not because of the gravity of infiltration water but because of the slow progression of capillary water with depth. When these conditions occur on a slope, the failure due to landslide of the lithic mass happens always at the base of the hydromorphic horizon and the maximum shear component is represented by gravity and not by pore-water pressure (u), which always has a very low value. In this case, a very shallow, flow type landslide is developed on the slope, but not less dangerous because of this (Fig. 6.1). The flow is due to the saturation of this horizon, with the piezometric surface at ground level, and occurs under very viscous fluid conditions. The landslide tends to be canalised into pre-existing erosional grooves that act as real riverbeds, downslope of which material accumulates in the fluid state and in the form of a conoid. This phenomenon of hydrogeological instability occurs only in very degraded soil, with a high index of voids when the platelets of clay component are detached from each other, i.e. when the frictional strength is equal to zero. Water infiltration is not the same at all the points of the shallow horizon and reaches different depths on the same slope, thus resulting in a discontinuous and wavy sliding surface.
    Fig. 6.1

    Flow instability due to the saturation of the shallow alteration horizon. The flowing material, widely heterogeneous, tends to canalise within pre-existing erosional grooves and deposits downslope in the form a viscous conoid (superficial hydromorphism)

     
  2. (2)
    Deep hydromorphism, it is always characterized by the presence of three overlapping lithic horizons (Fig. 6.2). The first is the horizon of agricultural soil, with very rough fabric and relatively good structure, due to the presence of organic material and characterised by very high non-capillary porosity (Fig. 6.2—Horizon Oe, A, E); the second horizon, underneath, is the lithic horizon of alteration (Fig. 6.2—Horizon Bt, Bw, C), composed of clayey-silty soil with a chaotic structure, in which the non-capillary porosity is virtually absent. The permeability of this horizon gradually decreases with depth; it retains free gravity water from the upper horizon which in wet periods becomes stagnant and forms a perched groundwater table if the underlying clayey soil, generally layered and light blue in colour, is impermeable (Fig. 6.2-R). This table disappears by evaporation during dry periods. The marbling process starts from the base of the permeable layers and develops progressively towards the surface. There, hydromorphism is often temporary, so tree roots are forced to penetrate deeply until they reach layers of low permeability and competent clay. The roots are then alternately subjected to asphyxiated conditions in periods of saturation and to desiccated conditions in dry periods. Slope stability factors depend essentially on the inclination of the contact surface between the horizon of alteration and the underlying unaltered lithic unit. This surface is very often wavy and discontinuous, with an average inclination similar to that of the slope, according to which, occasionally, oblique drainage can occur when the parent rock has an impermeable layer. In this case, the surface can facilitate the evacuation of gravity water and therefore local conditions of instability can hardly develop. When the ground is made up of predominantly clayey but tectonically fractured units, as almost all the units of the Earth’s mountain ranges basins are, perched groundwater tables in the horizon of alteration find a percolation pathway among the fractures tectonically originated. Under these conditions, the hydraulic head assumes a very important function in the mechanism of slopes deformation at failure.
    Fig. 6.2

    Lithic horizons

    (according to Fitz Patrick 1988, modified)

    Notably, the cause of plastic-rotational landslides (Varnes 1978) is to be attributed to a condition of pore-water pressure developed at the boundary of a saturated sliding surface. Moreover, in predominantly clayey soils, different hydraulic gradients between contiguous areas are easily developed, while the slope above acquires more potential energy than the one downslope in accordance with the assumptions introduced by Tison (1953), for the motion of a flow on an inclined impermeable plane (Fig. 6.3a–c).
    Fig. 6.3

    ac Cylindrical groundwater flow (two-dimensional)

     
Assuming that the groundwater table has an impermeable substrate inclined in the same direction of flow, Tison (1953), starting from Darcy’s Law, derives (Fig. 6.3c):
$$dy = dz\,{\text{cos}}\,i + dx\,\sin\,i$$
where
$$Q = Kz = \left( {\frac{dz}{dx}\cos \,i + \sin\,i} \right)$$
therefore
$$x = \frac{1}{tg\,i}\left[ {z_{o} - z + \frac{Q}{K\sin\,i}ln\frac{{K\,z_{o} \sin\,i - Q}}{K\,z\,\sin\,\,i - Q}} \right]$$
x

is the distance between the piezometer and the axis of the pumped well;

i

is the hydraulic gradient at point x;

K

the coefficient of permeability;

Q

flow capacity in m3/s;

ln

Neperian logarithm.

The application of this formula shows that the curve of Dupuit is less accentuated for the inclined substrate than for the horizontal one. On the contrary, for a reverse inclination, the curve is more pronounced.

However, the depression curve of the piezometric surface of groundwater does not always reach a profile of equilibrium linked to a constant radius of action, as if it was characterized by a permanent regime, since anisotropy of the ground prevents that the equipotential flow lines become perfectly parallel to each other. This could only happen where the subsoil presents an isotropic and homogeneous distribution of permeability, that soils actually do not have. The heterogeneity of these occurs for a number of reasons; for example: a different nature of the lithic components of the soil; a variation in the particle size distribution during the sedimentation process; different interaction of the electrostatic bonds between the platelets; chemical effects of the precipitation of substances in soil pores; the consolidation pressure, variable both horizontally and vertically; fissuring of rocks; stratification; cementation; etc. Such heterogeneity causes a variation in the physical characteristics of the soils and a curved and asymmetric pattern of flow lines that are predisposing for the variability of the permeability coefficient K from one area to another. Bromhead (1986), pointed out that the effect of the anisotropy of the soil lowers or raises neutral pressure (u); “Which one of these two cases actually occurs depends on the length of the drainage path”. There is a sudden spacing of flow lines if the soil has distinct areas with variable permeability. Furthermore, if the slope is subjected to load increase, such as for the overlapping of glacial masses, of earth dams, of artificial water basins, etc. or is subjected to tectonic compression forces, the volume of fractures or generic discontinuities decreases and thereby also decreases the permeability of the ground along the infiltration pathways, especially of those of the lesser depths. In these cases, inevitably increases the neutral pressure, which is directly proportional to the extension of the fractures. Moreover, during the triggering of a new landslide, extension fractures are initially discontinuous and occur at different depths below the ground level, within a zone where the tensile stress, due to gravity water, is greater. Therefore, for predictive purposes, it is needed to be cautious in identifying the depth of the sliding surface of a first-generation landslide, since this surface will manifest at failure as corrugated even if obtaining surface continuity.

In highly permeable soils (K ≥ 10−4 cm/s), instead, the excess of neutral pressure hardly establishes due to the absence of a response in terms of pore-water over-pressure due to the flow of fluid among soil pores. In that case, the soil tends to consolidation.

6.3 Consequence of Slopes Erosion on the Variations of the Coefficient of Permeability K

Geological and geotechnical literature have often highlighted the consequences of the erosion of a slope on soil structure and fabric variations as well as on the regime of neutral pressures and on the stability of natural slopes.

Matheson and Thomson (1973) analysed swelling phenomena developing in some valleys in Alberta, Canada, due to the effect of river erosion (Fig.  2.1). In the example considered, the stratigraphic sequence is composed of overconsolidated grey clays underlying sandstone layers from the Pleistocene. The erosion of the arenaceous horizon produced a kind of unloading of the clays, so that the latter underwent a continuous modification of the consolidation state and a reduction in the effective strength over time. In the outcropping clay, cyclic oscillations of the neutral pressure regime gradually took place, with the development of softening and swelling phenomena on the scarps as well as of sliding and collapsing of the latter. In fact, the total swelling of the clay in the eroded valleys was estimated at approximately 12% of the depth of the excavation carried out by the watercourses.

Similar situations have also been observed by Pasek (1974), which presented a deformational model of some valleys subject to fluvial erosion, whose slopes are composed of fractured competent layers (e.g. sandstone, conglomerates, limestones, etc.) laying on a substrate of deformable cohesive soils such as consolidated clays, marly-silty clays, pelitic flyschs, etc. From Fig. 6.4a, b it can be noted that the deepening of the river initially results in the swelling of the clay soil (phase 4) and then in the instability of the competent rock masses (phases 5–6–7). The dynamics of the disaster is characterized by an initial and progressive opening of the vertical fractures producing to the overturning (toppling) of the blocks on the slope due to the movement of the clays at plastic state (phases 6 and 7). This mechanism of slope deformation and failure, already influenced by the incisiveness of water, occurs only when the outcropping cohesive soils undergo a decay in strength due to softening (Fig. 6.5). It should be noted that the moving blocks are just the front ones, projecting towards the valley under erosion. Indeed, it is precisely towards this front that the clay may undergo a state variation for softening. Subsequent studies (Di Maio 1996a) showed that the phenomenon of state variation is particularly important in the case of overconsolidated clays and very competent and fractured argillites of high plasticity. The phenomenon is attributed to the alteration of the physical-chemical characteristics of zeolitic water due to the infiltration of rainwater, characterised by lower salt content. However, the shear strength of these types of cohesive soil also depends on empirical parameters that are function of time, such as oscillations of the piezometric level (Yoshida 1990).
Fig. 6.4

Mechanisms of deformation and failure for rock slopes on clayey substrate: a model by Pasek 1974; b model by Cancelli and Pellegrini (1987)

Fig. 6.5

Hypothetic evolution over time of the factor of safety of a natural slope subjected to oscillations of the groundwater table (from Picarelli 1999)

Cerere and Lembo Fazio (1986), by means of a finite element numerical model, examined the stress states induced for two different hydraulic conditions of the slope corresponding to the presence or absence of a deep permeable layer and assuming elastic-plastic behaviour of the clay material. The results have shown the emergence of traction efforts in the rock mass, if the clay formation extends to a considerable depth from the bottom of the valley. The increase in tensile stress near the vertical wall of the rock mass becomes greater, if it is assumed the presence of a band of softened material at the base of the mass and along the slope.

Cancelli and Pellegrini (1987) pointed out that among the joints of an arenaceous plate laying on argillites, a very common lithologic condition on the hills of the central and northern Apennines (Fig. 6.4b), differential slides take place among the arenaceous blocks due to the variation in state of the underlying argillites, which induce the central blocks to settle more than the lateral ones. The blocks projecting toward the valley under river erosion, however, are involved in sliding and/or flow phenomena that mainly affect cohesive soils.

Last but not least, Lefebvre (1987) described a typical stratified structure found in some valleys of Canada, generated by the erosive action of the watercourses. This is a sequence of layers, with different permeability, which strongly influence the pressure regime in the ground. In fact, when such layers outcrop in the valley for the incisiveness of the watercourse, they act as drainage pipes for the surrounding layers, thus altering the regime of neutral pressures in favour of slope stability.

Picarelli (1993) studied the theme of excavations in cohesive soils with particular reference to the distribution of short and long-term pore-water pressures and the stability of the slopes. A neutral pressure regime which is depressed due to drainage may also be induced by natural excavations, such as those due to erosion. This situation was observed in the area of the municipality of Bisaccia, southern Apennines, where for several years surveys and monitoring have been carried out by Fenelli et al. (1992) (Fig. 6.6). Starting from such observations, a numerical analysis was carried out that fully confirmed the relationship between erosion and the neutral pressure regime and revealed that the geomorphological situation in the area itself depends on the erosive phenomena in progress.
Fig. 6.6

Map of Colle di Bisaccia hill and investigation locations (from Fenelli et al. 1992)

Despite of the considerable interest in these studies, it should be noted that none of the Authors has ever been concerned by variations in the permeability coefficient K as a consequence of the stability of slopes, unless implicitly. In this sense, Bromhead (1986) in his book “The stability of slopes” performed, in specific terms, an analysis on the effects of variations in the permeability coefficient K on the stability of slopes.

From the 1950s, wrote Bromhead (1986), “techniques based on flow-nets for solving problems of steady seepage were introduced in Soil Mechanics books. However, the flow net technique requires an isotropic and homogeneous distribution of permeability, which is not observable in natural soils. Intrinsically, soils are variably heterogeneous and anisotropic in all the natural slopes of the Earth’s globe. Consequently, also the permeability varies from place to place in the ground as well as pore-water stresses that develop into it. Therefore, flow nets based techniques can be used only to solve seepage problems for small areas characterised by low thickness, where the soil is homogeneous and isotropic. In effects, even in this case, the flow net technique is lacking in concreteness and it is very difficult to be applied”.

The effect of anisotropy on permeability is to prevent that equipotential and flow lines could cross at right angles (except in the particular case of water flow parallel to one of the anisotropy axes) while this is the condition for the development of square mesh flow nets (Fig. 6.7). This distraction can be sudden if the ground is made up of a succession of layers or beds with different permeability. Both of these effects can be defined using simple numerical models. Finite element techniques are particularly suitable for such demonstrations.
Fig. 6.7

Effect of variations in permeability. Even in a steady seepage analysis, slight alterations in the parameters ruling infiltration have a disproportionate effect. As an example, subterranean drainage is important and where it is present, the influence of reduction in permeability with depth is remarkable (Bromhead 1986)

Soil layering can therefore be the source of a significant anisotropy of permeability. Morphological and structural variations of a slope may also cause anisotropy of permeability. Soil compaction can induce a reduction in porosity and therefore of permeability. The effect of the permeability variation is therefore to increase or decrease the value of pore-water pressure or neutral pressure.

The heterogeneity in particle size distribution of a soil as well as chemical processes in the ground are also reasons for variations in permeability.

An important case in the change in permeability is the presence in the ground of tectonic fractures caused by compression and/or distension, especially for what regards the activation of pre-existing landslides. If the ground is subjected to an increase in effective stresses, for natural or artificial causes, the volume of fractures or joints decreases and therefore permeability reduces. Whatever is the cause for the reduction in soil permeability, a consequent increase in neutral pressures and in shear stresses takes place, which can destabilize a cohesive slope.

There are soil permeability conditions for which the stress variation is null or negligible. This occurs when the water flow in the ground is intermediate between the maximum flow and the minimum during the yearly season cycle.

In slightly compressible and, therefore, low permeability soils (e.g. overconsolidated clay) the effect of cyclical variations in stress is negligible; these are generally deep soils, still subjected to important lithostatic loading. This also occurs in heavily fractured soils for tectonic reasons, such as Varicoloured Clay, Red Flysch, Flysch of Crete Nere, etc., so that each slope has always a limit depth where the seepage regime is stationary and in balance with the average conditions of the groundwater table in the above ground; this global pathway of the seepage of surface water is the one that controls the overall stability of a slope in cohesive soils. It is possible, however, that on a slope a more superficial shearing surface develops with respect to the value of its limit depth; this in fact depends on short-term variations in the infiltration path, that are induced by hydraulic conditions at the boundaries of the surface. For example, as it happens in the hydrogeological basin of the Comarca de Aguas Negras Calilegua, in the province of Jujuy in northern Argentina (Fig.  5.23). In this context the hydrogeological instability of the slopes is predominantly superficial and affects only the lithological horizon of alteration, that abrupt and abundant tropical rains lead to saturation in a short time (a few hours). The inability to rapidly dissipate interstitial water pressure causes an immediate mass swelling and constant stress changes due to variations in water content. Starting from the moment of failure, the behaviour of the cohesive soil, initially assimilated to that of a saturated continuous body, becomes definitively conditioned by neutral pressure, which transforms the lithic mass from the plastic state to that of a growing mud flow (Figs.  5.28 and 6.1—Surface hydromorphism).

A very different situation is the one where the underground drainage of a slope is influenced by a single permeable layer or by a succession of permeable layers at depth. By changing the flow conditions in the latter, varies also the total head of groundwater and, therefore, also the infiltration paths as well as the neutral pressure conditions. If, for example, by pumping water out of wells is obtained a reduction in the piezometric level within the slope and a reduction in neutral pressures, with consequent consolidation of the soils.

This function certainly improves the stability of the slope itself. Very dangerous, instead, is a rise in the piezometric level within the wells themselves, as it can change the underground drainage path from the previous one, potentially to the opposite direction. The effects may be those of a destabilization of the slope. Therefore, any variation in the hydraulic conditions of a slope occurring faster than the correspondent variation in neutral pressures, will produce unbalanced stresses in the ground.

Summarising, it can be said that neutral pressures in a slope may not be in equilibrium with the hydraulic conditions at the boundary and this for two main reasons. The former considers that there may be excessive pore-water pressure in the slope due to phases of loading or unloading under undrained conditions; the latter considers that there may be a recent variation of the hydraulic conditions themselves at the boundary, which, however, has not yet reached its maximum potential.

An example showing the application of both of these conditions is on the north-eastern slope of the hill of Tricarico, southern Apennines, province of Matera, Italy.

6.4 The North-Eastern Slope of Tricarico

6.4.1 Location

The small town of Tricarico is located to the south of the Apennines range in the territory of the Basilicata region (Fig. 6.8). It is an ancient medieval town (Fig. 6.9a, b) whose residential area extends across an organogenic calcarenitic plate oriented along the Apennine direction (NW-SE), at between 620 and 680 m above the average sea level (Fig. 6.10).
Fig. 6.8

Schematic map of Southern Appenine (Italy) with location of the town of Tricarico

Fig. 6.9

a Medieval Tricarico of 17th century. b Current Tricarico

Fig. 6.10

Areal distribution of the town of Tricarico

6.4.2 Lithology of the Slope

At the north-eastern boundary of the town, calcarenite forms a 30 m thick steep slope, while it gradually reduces in thickness towards SE, where it gets 10 m thickness at its outer boundary. The calcarenitic mass is clinostratified, dipping of about 30° to SW, not clearly distinguishable due to amalgamation phenomena of the sediments. It rests locally in stratimetric continuity on grey-light blue marly clays of the Lower Clayey Member and belongs to a short, superficial, supralittoral sedimentation episode of the basin before it collapsed during Upper Pliocene (Calabrian), giving place to the deposition of the Upper Clayey Member of the Unit of the Bradanic Foredeep (Fig. 6.11 by Coppola 1993).
Fig. 6.11

Geological scheme of the north-east slope of Tricarico

The clayey-marly layers have a thickness of 30–50 cm and are highlighted by millimetric interbedments of grey silty sand.

On the north-eastern side of the hill there is an extensive colluvial layer that rests directly on the marly clays of the Lower Clayey Member (Figs. 6.11 and 6.12); these are grey and grey-light blue silty-clayey deposits, with, to the upslope, fragments and stone blocks coming from the calcarenitic mass at the top of the slope. The silty-clayey material is heterogeneous, chaotic, sometimes altered. Lithological and granulometric variability develops both horizontally and vertically within the deposit. The thickness of this varies between 35 and 20 m respectively from the upslope to the downslope. The two lithologies, overlying and underlying, are separated by a discontinuity surface that acts as a sliding surface only under certain physical-mechanical conditions of the colluvial mass.
Fig. 6.12

Stratigraphic sketch of the area of Tricarico. The Upper Clay Member outcrops to the west of Tricarico (partially reported on the geological map)

6.4.3 Morphological Processes in Place

At the northern edge of the colluvial mass flows the torrent Milo (Figs. 6.11 and 6.13), whose riverbed, heavily engraved within the Lower Clayey Member, follows a pre-existing sinistral strike-slip fault surface that acts as the main collector for the flows and runoff coming from the surrounding slopes. It exerts a decisive action on the mechanisms of deformation and failure of colluvial deposits.
Fig. 6.13

North-east slope of Tricarico. Colluvial deposits in hydrogeological instability laying on marly clays from the Lower Member

The north-eastern slope of Tricarico hill can be represented by two sections corresponding to sections A-A, B-B (Figs. 6.14 and 6.15).
Fig. 6.14

Zone 1 corresponds to the zone of depletion; zones 2a and 2b instead are zones of accumulation. Boreholes P1, P3, P4 were provided with piezometers; P2, K2 with piezometers and inclinometers

Fig. 6.15

Zone 1 corresponds to the zone of depletion; zones 2a and 2b instead are zones of accumulation. Boreholes P5, K1 e P7, K3 were equipped with piezometers and inclinometers; P6 with piezometer

  • Depletion zone (1) upslope, home of landslides in stratified marly clays and of falls of calcarenite blocks whose debris feed the accumulation zone downslope. The average acclivity is about 30°;

  • Accumulation zone, intermediate zone and valley areas, where the colluvial mass extends to; is home to the most important rotational type mass movements. The upslope ground surface (2a) has an average inclination of = 20° while the average of the downslope Section (2b) is = 22°. Furthermore, the accumulation zone (2) is typically bilinear due to the different nature of the deposited lithic components. In 2a the colluvial mass has a ground surface conditioned by the presence of a sandy-arenaceous fraction mixed within the clay component, in greater percentage than that of zone 2b below. In fact, from the particle size distribution of some colluvial soil samples, it resulted that the silty-sandy fraction increases from 10 to 30% of the total within deposit 2a, especially at ground level. The difference in slope between 2a and 2b is minimal and irrelevant to the slope failure mechanisms; however, a different deformation process develops between the two zones. In 2a, during rainfall, an excess in the seepage of rainwater is obtained in the ground that corresponds to a load increment in undrained conditions. There, the deposit deforms by compressing at the foot, being confined from the downslope (zone 2b), not necessarily moving. In 2a, therefore, undulations develop tending to expand upwards, well visible at ground surface. The upslope part (zone 1) undergoes tensile stresses and hence the plastic-rotational failure process of the clay beneath the foot of the calcarenitic mass, which has already undergone softening. These are neutralised by applying to the base of the calcarenitic mass itself a tensile reaction with the protruding blocks upwards opening according to an inverted V (∧) shape. The blocks are involved in the stress process of the clays under conditions of passive resistance of the wall (Fig. 6.16). In practice, the calcarenitic front undergoes initially vertical deformations parallel to the front itself, with a maximum opening at the base that tends to close upward (Figs. 6.14 and 6.15; Sect. 6.2). The blocks, therefore, settle and sink into the ground at the plastic state, toppling to the upslope.
    Fig. 6.16

    Stress-path applied to the colluvium in passive limit state conditions (from Lambe and Marr 1979)

Mass movements, with a plastic-rotational behaviour in 2a, occur only when the roto-traslational kinematic process in zone 2b, of impulsive type, frees up the upslope mass from the lithostatic counterthrust. They, both in 2a and 2b, are caused by direct seepage of meteoric waters into the colluvial mass, with no percolation from the overlying calcarenitic mass. There, water outflow follows a westerly path imposed not only by the dipping of the layers but above all by the separation surface between the calcarenitic cluster and the underlying marly clays, acting as an impermeable bed dipping 25° to SW. Here, in fact, no softening phenomena develop, as the marly clay is confined in every direction. The calcareous cluster, in fact, remains stable throughout its extension except to the eastern edge, projecting on the valley of torrent Milo, where the bulging of the clay soil at the base induces differential settlements in unstable calcarenitic blocks, as already described.

Landslides are triggered predominantly in the winter season or in the months when abundant meteoric precipitations occur; in that period torrent Milo has a primary function: once removed the debris accumulated in the riverbed, the action of the flow is to deepen the riverbed itself and, consequently, to induce limit equilibrium conditions in the colluvium. Thus, begins the succession of landslides that grow backwards along the slope. This dynamic of landslides is typical of all the soils whose geomorphological model involves valleys originated by river erosion within deposits made up of more or less competent rock clusters resting on cohesive materials. Consequently, the evaluation of the conditions for the triggering of landslides on slopes formed by these soils requires adequate knowledge of the groundwater regime, as a quick triggering is often observed as the result of substantial variations in the values of pore-water pressures in conjunction with intense and/or prolonged rainfall events.

In the intermediate accumulation zone (2a), sometimes shallow movements occur affecting the altered and decompressed colluvium that moves in the form of mud flow when the piezometric level is at ground surface. There is no clear landslide front; the accumulation at the foot develops only for the downslope landslide mass (N. 1 in the cross section) where the debris from the sliding zone is pushed into the riverbed of torrent Milo.

6.4.4 Instrumental Monitoring

Instrumental monitoring was performed through n° 7 boreholes (p), aligned along sections A-A and B-B (Figs. 6.14 and 6.15), some equipped with multi-parametric columns (K) for 2D inclinometric and piezometric monitoring.

Boreholes drilling started on Monday 8th September 2008, while continuous monitoring began on Tuesday 1st October 2008, and ended on Monday 30th May 2011. It has been determined via mechanical borehole drilling that colluvium thickness decreases to the downslope (Table 6.2); it is 34.3 m in the upper part of the slope (P5) and 45.7 m in the middle (P6), while downstream it is about 31.5 m (P4).
Table 6.2

Referring to Figs. 6.16 and 6.17

Geological sections

Boreholes depth (m)

Parametric columns depth (m)

Colluvium depth (m)

Rupture depth (m)

A-A

P1 = 27

P2 = 50

K2 = 50

42.5

28.2

P3 = 48

44.3

P4 = 36

31.5

B-B

P5 = 43

K1 = 43

34.3

34.3

P6 = 54

45.7

P7 = 38

K3 = 38

31.5

The same measurements, of course, concern the depth of the discontinuity surface between the colluvium and the underlying stratified marly clays. However, this surface does not uniformly act as a sliding surface along the path of the moving mass. This, in fact, depends on the position the sloping mass assumes on the slope. This means that the shearing surface of colluvium in the medium and upper part of the slope is not conditioned by the pre-existing lithological discontinuity surface, but that activation can develop by means of a progressive deformation mechanism induced by pore-water pressure at a certain depth of the colluvial deposit. On the other hand, to the downslope such deposit has a reduced thickness, often insufficient for the hydraulic head to be equivalent to that of collapse and therefore the landslide moves along the pre-existing surface of lithological discontinuity where the resistance to be mobilized is certainly the residual strength, subjected to the hydraulic head of the torrent riverbed. Therefore, while to the upslope the failure occurs due to the rise of neutral pressures (Figs.  1.2 and 6.17), to the downslope the colluvial mass moves for the strong hydraulic head of the torrent and for the active thrust of the upstream soils.
Fig. 6.17

1 and 2 Elastic-viscous-plastic deformation measured within landslide units n. 3 Sez. B-B amd Sez. A-A, respectively in K1 and K2 up to failure

In structured cohesive soils, the pre-failure phenomena are identified by means of tension cracks of centimetre and/or decimetre length that appear at a certain depth under the effect of the neutral pressures induced by the cyclic variations of the piezometric level. They are irreversible deformations that define the initial state of the sliding surface. Tension cracks therefore represent one of the most important predictors of landslides, as they contribute to the definition of the incipient shearing area in propagation prior to collapse.

In colluvial soils, instead, although cohesive, the pre-failure deformation occurs through an elastic-visco-plastic deformation process along a very irregular continuous surface, as shown on Figs.  1.2 and 6.17. The colluvial mass of the medium to upper slope towards the east of Tricarico therefore deforms without sliding. However, as landslide masses 2 and 3 on sections AA and BB undergo the release of the lithostatic counterthrust of the immediately downslope colluvium, progressive failure propagates exactly along the pre-existing surface of visco-plastic deformation and not along the underlying surface of lithologic discontinuity. The piezometric level at failure is at 2.3 m below ground level (mean of the piezometric values obtained in the different landslide masses mentioned).

6.4.5 Discussion

In cohesive, chaotic, heterogeneous and sometimes altered soils of colluvial accumulation, widely spread on the Plio-Pleistocenic hills of the Apennines range, the cause of landslides is to be attributed to an interstitial water pressure condition developed at the boundary of the potential sliding surface at saturation state. Consequently, the determination of the conditions for the activation of landslides on the slopes formed by these soils requires adequate knowledge of groundwater regime, as there is often quick mobilization as a result of substantial variations in pore-water pressures in conjunction with intense and/or prolonged rainfall events.

The deformation related to the distinctive features of landslides in colluvial soils has been reconstructed in this explorative context by checking the progress of field experiments obtained from the complementarity between measurement apparatuses.

The morphological analysis immediately pointed out a peculiarity of colluvial soils, due to the fact that the landslide masses are well marked on the slope even before the triggering of their movement along the sliding surface. This is due to the fact that following to meteoric precipitation the colluvial mass is subjected in the short-term to a load increase in undrained conditions and deforms slowly by differentiating itself into several units of landslide, in succession on the slope, without triggering their sliding.

The drilling of boreholes showed a further characteristic feature of colluvial deposits; not all landslides use the pre-existing surface of lithological discontinuity, obtained by overlapping the colluvial mass on the underlying stratified marly clays (surface of separation between colluvial deposits and marly clay of the Lower Member), which has a low frictional resistance. Upslope landslide masses 2 and 3 in the sections reveal a pre-rupture surface at depths of 28–34 m below ground level, while the lithological discontinuity is at around 36 m. Furthermore, it has been found by means of the boreholes themselves that the thickness of the colluvial mass decreases to the downslope as well as the depth of the surface of lithological discontinuity, which passes from 36 m below ground level from to the upslope to 16 m below ground level to the downslope.

Instrumental monitoring, in particular by means of the multi-parametric columns for inclinometric and piezometric monitoring, carried out to the upslope, revealed that the pre-failure deformation in colluvial soils is of elastic-viscous-plastic type.

There, at a minimum depth of 28 m from ground level, the colluvial deposit behaves like a closed system without seepage motions; the effect of the variations in the stress state therefore largely depend on the oscillations of the piezometric level, which results in a substantial change in the effective stress field able to cause the elastic-viscous-plastic deformation without sliding of the lithic mass. The sliding occurs, instead, only when ceases the lithostatic counterthrust of the landslide mass immediately downslope. Therefore, the deformational dynamics of landslides in colluvial soils explicate via a deformative process of regressive-sequential type.

The downslope landslide mass (n. 1), close to torrent Milo, does not undergo pre-failure deformations. It moves along the pre-existing discontinuity surface only when the debris at the foot of the landslide have been removed from the watercourse and in conjunction with an appropriate hydraulic head existing within the torrent riverbed itself, under the thrust of the upslope soils.

In summary, what characterizes the activation of landslide movements in silty-clayey soils of colluvial accumulation is:
  1. (1)

    the destabilizing action of colluvial soils is due to the oscillation of the piezometric level, which at failure reaches the value of 2 m below ground level;

     
  2. (2)

    in colluvial soils landslide masses are identified on the slope in terms of shape, size and sequence even before their movement by collapse;

     
  3. (3)

    the pre-failure deformation in colluvial soils is elastic-viscous-plastic;

     
  4. (4)

    the pre-existing discontinuity surface between the colluvial mass and the underlying structured soils does not always work as a sliding surface, especially where the colluvial thickness exceeds 30 m;

     
  5. (5)

    the mechanics of the collapse of landslide masses in colluvial soil are influenced by the passive resistance applied by the landslide mass immediately to the downslope;

     
  6. (6)

    the deformational action of a slope in colluvial soils is of regressive-sequential type for subsequent collapse of the landslide masses.

     

6.5 Hydraulic Conductivity and Hydraulic Potential at Failure

Based upon field experiences by means of monitoring equipment it has been observed that in saturated soils under undrained conditions (φ′ = 0) the depth of the sliding surface depends solely on the permeability coefficient K, whatever the structural and mineralogical conditions of the lithotype. In fact, K decreases progressively with depth until it reaches the value of K = E × 10−7 cm/s where the substrate becomes almost impermeable. There, the excesses of interstitial water over-pressures, that do not dissipate in a short time and have a significant influence on soil behaviour at failure, tend to stabilise. Flow lines, fed by the infiltration of meteoric waters, are initially directed vertically downwards because of the high permeability value (K = E × 10−3 to E × 10−4 cm/s) within the colluvium and/or the alteration layer, then change gradually direction at greater depth, describing a typical arched shape, becoming parallel to the slope (Fig. 6.18) and finally coming out to surface at the bottom of the slope. This experimental observation is explained as follows: starting from a dry soil, the infiltration velocity (Vi) is directly proportional to the permeability coefficient (K) and reduces approximately according to the inverse of the square root of time (t) during which the process takes place.
$$V_{i} = \, K/t$$
Fig. 6.18

Schematic representation of the path of water flowlines in cohesive soils

Given that K reduces with depth, in parallel reduces also the infiltration velocity V i .

The amount of water (I) seeping through the ground during time (t) instead is proportional to the square root of time
$$I = K \cdot \sqrt t$$
When the soil is partially saturated, the infiltration velocity V i tends to remain constant and the previous formulas, in practice, become
$$\begin{array}{*{20}c} {V_{i} = \, K} & {I = K \cdot t} \\ \end{array}$$

As it can be observed, the infiltration (I) is strongly influenced by the hydraulic conductivity (K) of the soil and by the hydraulic potential already existing in the soil, which reduces enormously the infiltration velocity (V i ).

The depth at which flow lines become horizontal is the maximum depth of infiltration of water into the ground, where the water potential reaches its maximum (Fig. 6.19). In a cohesive soil, therefore, this potential is maximum at the depth where K = E × 10−7 cm/s. At this depth develops the shear strain which takes place, initially, through tension cracks (Fig. 6.20). However, even before the shear strain of the slope there are some cases that affect the processes of decay of the mechanical properties of the soils.
Fig. 6.19

Diagram of total vertical \(\left( {\sigma_{vo} } \right)\), of effective \(\left( {\sigma_{vo}^{\prime } } \right)\) and of pore-water stress \(\left( {u_{o} } \right)\) with groundwater at ground level with reference to the stratigraphy of 2nd member (A2) of the Formation of Serra Palazzo. In C the soils has a value of k = E × 10−7 cm/s. Along the same line, \(u_{o}\) has its maximum value

Fig. 6.20

Mesostructural detail of the shearing zone of Guildford landslide (from Skempton and Petley 1967—modified)

Picarelli (1999) highlights the destabilising action of the oscillations of the free surface of groundwater table on a slope in cohesive soils. The ground is in fact subjected to cyclic stresses due to the continuous variations of the “mean normal effective stress and to minor changes in deviatoric stresses ”.

The seasonal oscillations of the piezometric line in the ground induces, over time, a yielding in soil resistance conditions. The stress field within which do move the corresponding paths of the maximum oscillations is delimited by that soil band where the permeability coefficient K assumes values ranging between E × 10−6 and E × 10−7, i.e. where the neutral stress remains the maximum during its variations.

A second effect induced by the fluctuations in neutral pressure regime is that observed by Eigenbrod et al. (1992), who carried out specific laboratory investigations on the behaviour of slightly overconsolidated clays of medium plasticity and, by means of these, they noted that the excursions of groundwater table produce, in the long-term, viscous and softening effects, with the triggering of gradually increasing deformations, favoured by the reduction of soil stiffness characteristics. Even this second aspect of the reduction of soil strength has been detected on site, by means of the monitoring equipment, at the depth where the value of K ranged between E × 10−6 and E × 10−7, whatever is the nature of the clay deposit.

The importance of undertaking detailed experimental investigations to identify any shearing areas present in the subsoil is therefore of utmost importance in order to achieve the prediction and prevention of hydrogeological instability, since the shearing phenomena present in the ground are all dependent on the time factor, usually not defined through numerical analysis.

It was observed (Fig. 6.19) that at the depth where K = 10−7 cm/s the hydraulic head is maximum, if the overlying cohesive soil is saturated. By determining the coefficient K in situ along the vertical then the shearing zone can be identified, described by Skempton and Petley (1967) (Fig. 6.20), which manifested precisely where the coefficient of permeability reaches the value of K = E × 10−7 cm/s.

These are cracks that indicate the state of incipient shear deformation of the ground. It should be noted that at this depth the cohesive rock is subjected to a triaxial compression state σ1, due to the hydraulic head, and the confinement pressure σ3. The behaviour at failure depends on the values of σ1 and σ3 of the local stress field; σ3 plays a very important role on the formation of pre-failure cracks (Fig. 6.20). Indeed, since the laboratory data indicate that the intermediate principal stress σ2 has less influence on the peak strength than the minimum stress σ3, the criteria of more general use reduce to the form:
$$\upsigma_{1} = {\text{ f }}(\upsigma_{3} )$$
Mohr’s circles can easily represent this stress situation. Their construction comes from the failure experience (Chap. 2, Figs.  2.16 and  2.17). Putting τ and σ on the Cartesian axes, the values of σ1 and σ3 from the test are reported on the abscissa and the circle of Mohr (Fig. 6.21a) is traced. The angle θ subsequently defines the inclination of the failure plane with respect to σ1 (Fig. 6.21b). Expressing this relationship in terms of shear stress τ and of normal stress σn is obtained
$$\uptau = {\text{f}}(\upsigma_{\text{n}} ).$$
Fig. 6.21

​Representation in terms of Mohr’s circles

See—Theoretical failure in presence of σ3—Chap.  2.

It is worth specifying that tension cracks are not the ones considered as of pre-failure. The latter are neoformation cracks generated by interstitial water pressure and lithostatic load. The failure process takes place as follows: the interstitial water pressure is maximum at the depth where K = E × 10−7 cm/s. At that depth, the cohesive material absorbs water up to saturation; it becomes therefore completely impermeable in the absence of shear strain, since its expansion is prevented by the confinement in all directions. There, the percolation water stagnates or flows out slowly along a surface path characterized by K = E × 10−7 cm/s. In the overlying lithic horizons, in the meanwhile, the clayey soil undergoes a process of hydromorphism resulting from the jamming of impermeable substances released by capillary water within very small interplatelet voids (see Sect. 6.2). It thus forms an alteration horizon that contains groundwater within itself. At the base of this horizon, the electrostatic effect of clay platelets tends to create a local stress field which is not perfectly isotropic at all points but reveals a privileged direction along which the resistance to loading is lower. The direction of the deformational axis at failure depends not only on the orientation of τ but also on the brittleness of the reaction of the soil which is greater along the surface of discontinuity between the horizon of alteration and the intact formation, i.e. at the depth where K = 10−7 cm/s. From Figs. 6.2 and 6.5 , it can be observed that the factor of safety of a natural slope subject to oscillations of the groundwater table decreases over time. Furthermore, the failure is influenced by the factor of vertical oscillation of groundwater table (σ1), even getting that the orientation of the stress axes remains constant: σ1 = vertical, σ3 = perpendicular to the slope and τmax oriented at 45° from the vertical, parallel to the surface of the impermeable layer.

In the very common case, where the intact cohesive soil formation is made up of tectonically fractured units, the groundwater suspended in the horizon of alteration finds a percolation pathway among tectonic fractures and penetrates to a depth where the permeability of the ground is characterised by a value of the permeability coefficient K = E × 10−7 cm/s. In this situation, the hydraulic head can reach a very high value and in the long-term, with a piezometric surface often at the ground level, the deformation mechanism at failure occurs by explosion of the cohesive rock, in the close proximity of very steep slopes, similar to vertical walls, projecting towards the river valley. There, indeed, the confinement pressure is very low (Fig. 6.22a–c).
Fig. 6.22

ac Explosion of marly rock, very fractured, happened live on 9.5.2010 after three days of persistent rainfall on the southern slope near Quebrata del Toro—ancient road—Yuyui Province, Argentinian Pre-Andes

Otherwise, if the slope has medium or low acclivity, the confining pressure can exert a value greater than the hydraulic head. In this case, for composite landslides, the masses to the upslope are blocked by the lithostatic counterthrust of the downslope ones. They can develop pre-failure deformations only when they are subjected to movements induced by the stress-path (Fig. 6.16).

Referring to Fig.  4.3a–d in Chap. 4, the appearance of unstable fractures subjected to triaxial compression can develop in several ways.
  • If at the depth where K = 10−7 cm/s there is a pre-existing stable crack (Fig.  4.3a) this is subjected initially to a closure if the vertical load \(\sigma_{1}^{b}\) is greater than the initial \(\sigma_{1}^{a}.\)  If the load value increases, such that \(\sigma_{1}^{c} > \sigma_{1}^{b}\), the faces of the crack undergo sliding (Fig.  4.3c). When the fracture plane is subjected to a further vertical load, where \(\sigma_{1}^{d} > \sigma_{1}^{c}\), extensional cracks develop at the two ends of the fissure due to traction parallel to σ1 (Fig.  4.3d). The failure occurs if the value of σ 1 is furtherly increased becoming greater than \(\sigma_{1}^{d}\) (Fig.  4.3e).

  • Changes can take place both in the stress field (because of the dissipation of σ1) and in the properties of the material (\({\varphi} \gg 0\)) during the development of cracking; this can produce a stabilization of the crack. In this case, an increase of σ1 is required for inducing the growth of the crack.

These stress conditions concerning the mechanics of failure of natural slopes heavily influenced by pre-failure deformation phenomena occurred in December 1982 on the northern slope of Montagnolo hill in Ancona, Central Apennines, giving place to the Great Ancona Landslide, which will be discussed in Chap.  7. Not only the progressive failure but creep and softening as well there played an important role, resulting in a continuous modification of the stress states and a on the reduction in the available shear strength.

References

  1. Baligh, M. M. & Levadoux, J. N. (1980). Pore pressure dissipation after cone penetration. Research Report R. 80–11, Mit, Cambridge, MA. Google Scholar
  2. Baver, L. D. (1948). Soil Psysic (p. 398). Chapman and Hall Ltd.: Londres.Google Scholar
  3. Bishop, W. A. (1955). The use of the slip circle in the stability analysis of slopes. Geotechnique, 5(1).Google Scholar
  4. Bromhead, E. N. (1986). The stability of slopes. Blackie & Son Ltd.Google Scholar
  5. Bromhead, E. N., Coppola, L., & Rendell, H. M. (1994). Geotechnical background to problems of conservation of the medieval centre of Tricarico, southern Italy. Quarterly Journal of Engineering Geology, 27, 293–307.CrossRefGoogle Scholar
  6. Cancelli, A., Pellegrini, M. (1987). Deep-seated gravitational deformations in the Northern Apennines, Italy. In Proceedings of the 5a International Conference and Field Workshop on Landslides (pp. 171–178). Christchurch.Google Scholar
  7. Castany, G. (1967). Traité pratique des eaux souterraines (duexième ed.). Paris: Dunod.Google Scholar
  8. Cello, G. & Coppola, L. (1984). Assetto Geologico-Strutturale dell’Area Anconetana e sua Evoluzione Plio-Quaternaria. Bollettino della Società Geologica Italiana, 103, 97–109, 6 ff., 2 tavv.Google Scholar
  9. Cerere, V., Lembo Fazio, A. (1986). Condizioni di sollecitazione indotte dalla presenza di una placca lapidea su un substrato deformabile. In Atti del XVI Convegno Italiano di Geotecnica (Vol. I, pp. 191–202). Bologna.Google Scholar
  10. Combeau, A. & Monnier, G. (1961). Sols africains, 6(1), 4–32.Google Scholar
  11. Coppola, L. (1993). Evoluzione tettonica e meccanismi deformativi della media Valle del Basento, Bollettino della Società Geologica Italiana, 112, 159–179, 20 ff., 1 tav. f.t.Google Scholar
  12. Coulomb, C. A. (1773). Essai sur une Application de Regles de Maximis et Minimis a Quelques Problemes de Statique Relatifs a l’Architecture. In Mémoires de Mathématique & de Physique, présentés à l’Académie Royale des Sciences par divers Savans, & lus* dans ses Assemblées (Vol. 7).Google Scholar
  13. Delpont, G., Deramont, J., Guchereau, J. Y., Joseph, J., Majeste-Menjoulas, C. L., Soula, J. C., et al. (1973). Ruptures extensives et cisaillantes dans des series rythmiques sédimentaires (Montagne Noire et Pyrénées). Revue de Géographie Physique et de Géologie Dynamique, France, 15(5), 567–574.Google Scholar
  14. Di Maio, C. (1996a). Exposure of bentonite to salt solution: Osmotic and mechanical effects. Géotechnique, 695–707.Google Scholar
  15. Di Maio, C. (1996b). The influence of pore fluid composition on the residual shear strength of some natural clayey soils. In Atti del VII International Symposium on Landslides (Vol. 2, pp. 1189–1194). Trondheim.Google Scholar
  16. Dupuit, J (1863) Etudes théoriques et pratiques sur le movement des eaux dans le canaux découverts et à travers les terrains perméables, 2e édit. Paris: Dunod.Google Scholar
  17. Eigenbrod, K. D., Graham, J., Burak, J. P. (1992). Influence of cycling porewater pressures and principal stress ratios on drained deformations in clay. Canadian Geotechnical Journal, 326–333.Google Scholar
  18. Fellenius, W. (1992). Erdstatisce Berechnungen. Berlin: W. Ernst.Google Scholar
  19. Fenelli, G. B., Picarelli, L., Silvestri, F. (1992). Deformation process of a hill shaken by the Irpinia earthquake in 1980. In Atti del Conv. Italo-FranceseStabilità dei Pendii in Zona Sismica”. Bordighera.Google Scholar
  20. Fitz Patrick, E. A. (1988). Soil horizon designation and classification. Technical Paper 17 Isric, Wageningeu.Google Scholar
  21. Janbu, N. (1973). Slope stability computations. The Embankment dam engineering: Casagrande Volume (pp. 47–86). Wiley.Google Scholar
  22. Lambe, T. W. & Marr, W. A. (1979). Stress path method (2nd ed., 724–738) JGED, ASCE, June 1979.Google Scholar
  23. Lefebvre, G. (1987). Slope instability and valley formation in Canadian soft clay deposits. Canadian Geotechnical Journal, XXIV(3), 261–270.Google Scholar
  24. Matheson, D. S., & Thomson, S. (1973). Geological implications of valley rebound. Canadian Journal Earth Sciences, X, 961–978.Google Scholar
  25. Morgenstern, N. R., & Price, V. E. (1965). The analysis of the stability of general slip surfaces. Geotechnique, 15, 79–93.CrossRefGoogle Scholar
  26. Pasek, J. (1974). Gravitational block-type slope movements. In Proceedings of the 2th International Congress (Vol. II, th. V-PC 1). Sao Paulo: IAEG.Google Scholar
  27. Picarelli, L. (1993). Structure and properties of clay shales involved in earthflows. In Atti dell’International SymposiumThe Geotechnical Engineering of Hard Soils-Soft Rocks” (Vol. 3, pp. 2009–2019). Atene.Google Scholar
  28. Picarelli, L. (1999). Meccanismi di Deformazione e rottura dei pendii. In Argomenti di Ingegneria Geotecnica (Vol. 14). Hevelius Edizioni.Google Scholar
  29. Sarma, S. H. (1979). Stability analysis of embankments and slopes. Journal of the Geotechnical Engineering Division, American Society of Civil Engineers, 105(GT 12).Google Scholar
  30. Skempton, A. W. (1948). The ϕ = 0 analysis of stability and its theoretical basis. In 2nd ICSM. Rotterdam.Google Scholar
  31. Skempton, A. W., & Petley, D. J. (1967). The strength alone discontinuities in stiff clays. In Atti della Geotechnical Conference (Vol. 2, pp. 29–46). Oslo.Google Scholar
  32. Taylor, D. W. (1948). Fundamentals of soil mechanics. New York: Wiley.Google Scholar
  33. Terzaghi, K., & Peck, R. B. (1967). Soil mechanics in engineering practice (2nd ed.). New York: Wiley. The first edition was published in 1948.Google Scholar
  34. Tison, L. J. (1953). Cours d’hydraulique, 2e parties (pp. 265–430).Google Scholar
  35. Torstensson, B. A. (1975). Pore pressure sounding instrument. In Proceedings of the ASCE Specialty Conference, Measurement of Soil Properties. Raleigh: North Carolina State University.Google Scholar
  36. Torstensson, B. A., & Petsonk, A. (1985). Bat water monitoring system. In ASTM Symposium of Field Methods for Ground Water Contamination Studies and their Standardisation. Cocoa Beach, FL, USA, February 2–7, 1985.Google Scholar
  37. Varnes, D. J. (1978). Slope movement types and processes. In R. L. Schuster & R. J. Krizek (Eds.), Landslides: Analysis and control. Washington, DC.Google Scholar
  38. Vigneron, J., & Desaunettes, J. R. (1960). VII congrès international de science du sol (Vols. 1, I, 5, pp. 114–121). Hadison.Google Scholar
  39. Yoshida, N. (1990). Time-dependent instability in fissured overconsolidated clays and mudstones (Ph.D. thesis). University of Alberta, Edmonton.Google Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.University of BasilicataPotenzaItaly

Personalised recommendations