GB1587187A - Method of reinforcing a soil structure - Google Patents

Description

(54) METHOD OF REINFORCING A SOIL STRUCTURE (71) I, RICHARD AUCHINLECK JEWELL, a British Citizen, and residing at Stillwaters, Rolvenden, Kent, do hereby declare the invention, for which I pray that a patent may be granted to me, and the method by which it is to be performed, to be particularly described in and by the following statement: The invention relates to a method of reinforcing soil structures, such as embankments, retaining walls and dams.

The method of the invention may be used to strengthen soils, and combinations of soils.

Soils may typically be classified by their particle size: for example Boulders > 200mm; cobbles 200 > 60mm; gravel 60 > 2mm; sand 2 > 0.6mm; silt 0.06 > 0.002mm; clay < 0.002mm; Natural soils generally contain particles from at least two of the above divisions.

As a material, soil is an assemblage of particles. The general behaviour of a soil mass is complex and depends on several factors, two of which are the properties associated with individual particles, and the way that the particles interact with each other.

It should be noted that the description of soils by fundamental and consistent stress-strain laws, and the establishment of a proper understanding of their mechanical behaviour is complex and the subject of continuing research. Such soil descriptions or models require versatile analytic methods so that they may be applied in design; the most versatile method currently available is the finite element method. These analytic methods are very complex.

It is not necessary in this Specification to discuss the detail of reinforcement action and design using the complex concepts, and mathematics, associated with a description of soil stress-strain behaviour, and in the analytic solution of soil boundary value problems.

Instead, recourse is made to simple strength calculations: the soil is modelled with two components of strength, cohesion and friction, and the simple concepts of rupture surfaces and limit equilibrium are used for stability analysis. This simple analytic mode, which is commonly used in the stability analyses of many soil structures in present engineering practice, will be sufficient to give a simplified account of reinforcement action by the method of the invention, and a simple method of design.

As an engineering material two important considerations for a soil are deformation and failure, under natural and imposed stresses. Deformation is a stress-controlled system governed by the stress-strain properties of the soil.

In engineering practice failure is the primary design consideration. The ability of a soil to safely fulfil a structural function without failing may be checked by examining the stress state within the soil body, on potential rupture surfaces, and ensuring that it is within permissible limits defined for the soil. The overall ratio of Restraining Strength Activating Stress is generally used as a safety index called the factor of safety.

Having ensured the stability of the structure with suitable factors of safety against all possible failure mechanisms, the designer will usually consider the deformation of the structure.

In a soil individual particles are strong. As an assemblage of particles soil may sustain loading. Figure 1 of the accompanying drawings shows a sample of soil S under imposed stresses ol, a2 and n3 (for simplicity assume a triaxial state of stress where a1 > (a2 = a3)).

If the stress a1 is approximately equal to the stresses a2 and a3, most soils may sustain large loadings, the deformations being mostly dependent on the crushing of the constituent particles.

Soils cannot generally accept a large difference in the levels of the stresses a1 and (73.

There is generally some limiting difference between a1 and (73, which is typically a property of the soil and a function of the stress levels a1 and 03, above which the shearing strength of the soil is exceeded and shearing will occur within the soil. It is the process of shearing, sliding occuring on rupture surfaces, which is commonly observed when soils fail: the shearing resistance of a soil may therefore be regarded as a critical property.

It has been observed that many soils fail by rupturing into two or more separate blocks which slide relative to each other, remaining in contact, along certain critical rupture surfaces, until equilibrium is restored.

As discussed above a simple and commonly used model for a soil specifies two components of strength: cohesion (k) and friction (p). Both these components must be overcome before sliding along a potential rupture surface may occur.

Four typical types of rupture, comprising a failure mechanism, are shown in Figures 2 to 6 of the accompanying drawings, by way of example.

Figure 2 shows a homogeneous slope 1 experiencing failure along a rupture surface 2; Figure 3 shows a slope 1 experiencing distorted failure along a rupture surface 2 passing substantially along a weak layer 3; Figure 4 shows a slope 1 experiencing failure by planar rupture 2 due to a weak layer 3; Figure 5 shows the classic failure mechanism beneath a surface foundation slab 5, under an applied load 6, resting on level homogenous ground; Figure 6 shows a retaining wall experiencing failure along a rupture surface 2, the wall moving outward as indicated.

The potential failure mechanism that has the lowest factor of safety is termed the critical failure mechanism: its primary surface, that surface which separates the moving soil from the stationary soil, is termed a potential critical rupture surface. One or more zones may also be located within the cross-sections of the soil mass through which all possible potential rupture surfaces with factors of safety below a certain specified value will pass. This or these zones may be termed the critical zones, and represent the most severely stressed soil.

The invention provides a method of reinforcing a soil structure, the method comprising the steps of determining the path of potential critical rupture surfaces (as hereinbefore defined) through the structure, and then incorporating wholly within the structure a soil-shear-resisting, rigid reinforcement unconnected with any external surface of the structure in a position in which it extends across the path of the potential critical rupture surfaces so that the reinforcement crossing the rupture surfaces resists straining of the soil in the structure along the rupture surfaces.

The invention extends to a soil structure reinforced by such a method.

By inserting one, or more, rigid reinforcement members into the soil mass, in critical zone(s), across the path of the potential critical rupture surface(s), the ability of that soil mass to resist the development of that failure may be increased, the failure prevented, and the soil thereby strengthened.

For soil structures reinforced in accordance with the method of the invention the overall stability is provided for mutually between the natural strength of the soil and additional strength derived from the reinforcement in the critical zone(s) resisting the deformations and stresses which would have led to collapse in the unreinforced soil.

The principles by which the reinforcement functions may be visualised in different ways.

One simple concept, would focus on the interaction between the reinforcement and the soil where the reinforcement passes through the potential zone of highly localised shear straining. The presence of the reinforcement at this zone, interacting with the soil as it deforms, would be seen to inhibit the localised shear straining, strengthening the soil and altering the stress state and strain pattern. Loading of the reinforcement by the soil, as it strains, sets up stresses in the reinforcement: these stresses are redistributed by the reinforcement member(s) to the surrounding soil mass by a system of forces (bending, tension and shear) within the reinforcement member(s) in association with the surrounding soil/stress system.

Using the method of the invention many conventional soil structures may be improved, and many other structural problems may now be solved. Some examples of structures made in accordance with the invention are given later, by way of example.

A simplified approach to the strengthening by the method of the invention will be described with reference to Figures 7, 8 and 9 in which: Figure 7 shows a section of a rupture surface 2 where zone A may slide with respect to zone B. Normal effective stress and shear stress are shown.

Figure 8 shows in detail the relationship between X and o' the shear and normal stress.

The shear stress has components of friction 7 and cohesion 8.

Figure 9 shows the limits of shear stress (I ) which may be sustained by a natural soil sample on a potential critical rupture surface prior to sliding: after Coulomb, X = TI (a').

It is generally accepted that the available resistance to sliding along a given rupture surface may simply be calculated in terms of two properties of the soil, cohesion (k) and friction (p), and the normal effective stress on the given rupture surface (o') (Figure 7) where the normal effective stress is a measure of the particle to particle loading.

Resistance due to cohesion: (k) Resistance due to friction: (o' tan p) Total resistance T = k + ó' tan p (Figure 8) The rupture surface will remain stable if the activating shear stress (t) is less than the total available resistance (t ).

i.e. wall k + ó' tan p. (Figure 9) Now let a rigid reinforcement member 9 be integrated into the soil mass and across the potential rupture surface as shown in Figure 10. Any tendency, at the limit of stress, for sliding on the rupture surface so that zone A will slide as a rigid body with respect to zone B, will be resisted by the reinforcement member which is substantially fixed in both zone A and zone B.

The strengthening that may be derived from such reinforcement in a soil depends on many factors of which the following are the most important and should be considered: dimensions, form (for example grids or bars), orientation, bending stiffness, longitudinal stiffness cross-sectional area, and surface properties of the reinforcement members, and size and grading density, stress level and drainage conditions of the soil.

Consideration of these factors should take account of the stress-strain behaviour of the soil, in particular the strain behaviour under the relevant loading. It is the strain behaviour which primarily governs the selection of the reinforcement material form, dimensions and disposition.

For present purposes a strength analysis using a suitable analytic method such as the limit equilibrium approach may be used. By making assumptions (based on test evidence) about the strain pattern and magnitudes of strain to failure for the given soil at a given initial state and stress level, information about the optimum reinforcement orientation, as well as design dimensions cross-sectional area and material modulus may be derived.

In general when reinforcing coarse granular materials where vigorous dilatancy and volumetric strain are a pre-requisite to failure then reinforcement oriented along the minor principal strain direction will be most effective in resisting this strain behaviour, and strengthening the soil against failure. In fine grained and cohesive materials the strain pattern and optimum orientations will vary depending on the soil state, loading, loading rate and drainage conditions; the stiffness of the reinforcement will be of great importance.

The reinforcement need not be at right angles to the rupture surface.

A suitable analytic approach would be (a) Do a series of limit equilibrium analyses on the unreinforced structure of the required geometric form to identify the most severely stressed soil elements (i.e. the critical zones) (b) Add in reinforcement members at suitably selected locations and orientations and reanalyse for stability adding in extra shear resistance to be derived in the soil by virtue of the reinforcement at the location of the reinforcement.

Repeat the process making design adjustments as required until all factors of safety are of the required value. Now re-check that all possible modes of failure have been considered.

The extra shear resistance derived in the soil by virtue of any reinforcement is a function of the parameters listed above. The form and dimensions should be such that the stresses in the reinforcement may be redistributed to the soil mass without bringing that mass to the limit state; the reinforcement must also be able to transmit the required stresses without yield or fracture. The reinforcement acts like a skeleton. A favourable reinforcement form is a grid or mesh which exhibits a high degree of fixity in the soil.

The reinforcement can be of any suitable form or configuration, and can be made out of any material or composition of materials that possesses sufficient strength and rigidity. The strength and rigidity should be such that the member should normally be able to sustain the stress system exerted by the soil, at the relevant location of this reinforcement member within the structural form, without undergoing deformations of a magnitude which might adversely effect the function of that reinforcement member in inhibiting the shear failure of the soil, and hence in strengthening the soil.

The reinforcement form may be of various different kinds, for example grids, meshes, panels, beams or bars. Suitable materials may be for example metals, plastics, polymers, composite synthetic plastics, wood, pre- or post-stressed concrete, simple reinforced concrete.

The reinforcement members should be fully integrated into the body of soil in such a way as to ensure the continuity of the soil mass: this installation may be likened to a skeleton which strengthens without destroying continuity.

On some structures the reinforcement may most conveniently be pre-formed units. These would most simply be installed during the filling operation, care being taken with the compaction in and around the reinforcement members. On other structures and in particular in cohesive material it may be convenient to form or cast the reinforcement insitu; for example a construction lift may be completed, compacted and then the "mould" for the reinforcement excavated - the soil acting as shuttering, so to speak.

It would also be possible to insert reinforcements in an existing structure, as a remedial measure.

When casting the reinforcement insitu, great care must be taken not to disturb the insitu state of the soil unduly. Also in some special instances it may be beneficial to bed the reinforcement, whether it is cast insitu or not, in a material differing from the main body of soil in the structure. The provision of suitable filter layers for drainage should be ensured where necessary.

Construction when using the method of the invention may be proceeded with conventionally. In fill structures, at the relevant levels, the reinforcement configurations should be inserted, the soil being carefully compacted in and around the reinforcement members to ensure continuity within the soil mass. When the reinforcement is to be installed into natural ground, great importance should be attached to avoiding any undue disturbance of the ground during installation. When excavation is being used the back filling and compaction must be thorough.

In certain structures, sealing of the surface of the soil, e.g. by a skin, will be required to secure the structure 1 from external effects such as impact loads or erosion, 2 from internal effects such as local loss of material through local surface instability or loss from seepage, (3) for aesthetic, architectural or environmental effects - e.g. facades for walls and abutments or skins for seeding grass on embankments.

If a sealing skin is required or desirable; the type of skin selected will depend on the structural form and function. In some structures the skin may be required to prevent the loss of material at the face under static conditions. The stresses which the skin may have to resist are generally small when compared to stesses within the soil body. Depending on structural form and function some anchoring of the skin to the body of soil will be advantageous in ensuring structural integrity. This may simply be accomplished by suitably spaced anchored extending into the soil mass.

The skin may be rigid or flexible, comprising preformed units or units formed insitu. In all cases the skin material must be capable of resisting attack due to external or internal erosive forces or, where necessary, aggresive chemicals. Provision for expansion joints, drainage (filter layers behind the skin for example) should normally be made.

Suitable materials for the skin are, for example bitumens, asphalts, plastics, polymers, woods, metals, composites, pre- or post-stressed concrete, simple reinforced concrete and mixtures of these and other materials where desirable.

A composite skin may also be formed by the addition of chemicals or bonding agents to the soil to form a coherent mass capable of withstanding the required stresses. Soil-cement is an example.

The skin preferred for a retaining wall with a vertical face may often be one comprising many pre-fabricated units, of rigid structure, which will fit and interlock with each other to form a homogeneous skin. Often a thickness of filter may be placed behind the face connected to some drainage arrangement in the fill and to a series of weep holes or outlets at the face. With such a skin, allowance for exansion joints may often be dispensed with due to the inherent flexibility of inter-unit connections. The pre-cast units may be anchored to the soil by conventional anchors or by a method detailed here.

Figure 11 shows one possible facing unit 22 of the simplest form by way of example. On two edges there is a female junction 20, and on the other two edges a male junction 21 to provide some interconnection.

Figure 12 shows a side elevation of one unit 22. An anchor 22a of length T may be attached at 23 to the face unit to anchor the unit to the body of the soil and to stabilise the immediate vicinity of the face from localised collapse. This length T need not normally exceed three times the height of the face unit (u), depending on the soil, the filter layer chosen and location on the face. Depending on the position of a particular unit 22, its anchor 22a may extend across the path of a potential critical rupture surface in the soil structure, to provide additional reinforcement.

Figure 13 shows a front elevation of the skin as seen on a section of the wall.

When the structure requires a skin, and the local face stability depends on that skin, then construction should proceed with the sequential placing of free units, suitably anchored, and the main reinforcement as described above. An example is a retaining wall 14 as in Figure 22, with facing units as in Figure 12; the facing units should be erected in phase with the construction, suitable attention being paid to their installation, positioning and compacting around the reinforcement, and the main reinforcement installed as described above.

By way of example some details of one specific reinforcement type, a grid, are given below with details of its use in the construction of an embankment and retaining wall: some general remarks relevant to the design and calculation method are given.

Figure 14 shows a rigid reinforcement grid 9. The grid has principal length Y and width X.

Principal grid members N are spaced at intervals m. Secondary members M are spaced at intervals n which may vary over the width of the grid.

Figure 15 shows cross-sections of a member M or N. The member will usually be formed from a single material such as a metal or polymer, or it may be formed from reinforced concrete. The cross section may be circular, rectangular, I-shaped or T-shapd. The surface may be deformed to facilitate good fixity/bond with the soil, for example like the deformed bars used in reinforced concrete.

For simplicity in this particular example, the grid apertures may be assumed to be square and of constant size over the whole grid area, the members M and N may be taken as the same in strength and dimension. In practice the members N which cross or traverse the rupture surface have the main function of resisting that rupture. The members M which extend roughly parallel to the rupture surface act mainly to give fixity and to bond with the soil, and hence to redistribute stresses. These two may therefore have two different optimum cross-section.

Figure 16 shows an elevation of a critical zone with a potential critical rupture surface 2 crossed by the rigid reinforcement 9.

Figure 17 shows a plan of Figure 16. The reinforcement 9 extending from zone A to zone B across the rupture surface 2, may be seen. The rupture plane intersects principal members N.

The main function of the rigid reinforcement is to inhibit the formation of the rupture surfaces which lead to failure of the structure. The reinforcement must be of sufficient dimensions to ensure a reasonable degree of fixity within the soil mass under the expected soil/stress system, so that it may redistribute loading as required by its structural function, without unduly disrupting the continuity of the soil. Before a structure may be declared stable an adequate factor of safety must be ensured against any possible mode of collapse of the structural whole.

The location of the reinforcement within the structure, its inclination, strength and number may be detailed with regard to: (1) Providing sufficient extra shear strength in the soil at the required locations to inhibit all potential rupture modes of collapse with a suitable factor of safety.

(2) Relative cost of different arrangements.

(3) The final structural form and dimensions.

(4) Considerations of stiffness and deformation desirable in fulfilling a function of the structure.

(1) above: All possible modes of collapse must be investigated and prevented. The use of small scale models may be useful in checking this factor, (especially in centrifuge tests).

(2) above: In a fill structure constraints as the construction proceeds may often preclude a free choice of reinforcement orientation, which is one significant factor in the possible stengthening from any given reinforcement. Fortunately the horizontal position for reinforcement in an embankment or retaining wall is close to the optimum orientation.

However the cost benefit of placing reinforcement at the optimum angle should always be considered, especially when near foundation level or when reinforcing undisturbed ground insitu.

(3) above: For example, for a small structure one strong principal plane of reinforcement may be optimal; for larger structures several principal planes would yield a better solution.

(4) above: For example the deformations along the top of the structure may be critical in the function of the structure. (superstructure, roadway) Reinforcement near the surface will stiffen that surface zone against undue deformation.

Figures 18, 19 and 20 illustrate the stabilisation of an embankment. In particular, Figure 18 shows an embankment in which the critical zone 13 contains all potential critical rupture surfaces that have a factor of safety below a certain specified value. All areas outside the critical zone have higher factors of safety.

Figure 19 shows the embankment reinforced by the method of the invention with one principal plane of reinforcement 9 ensuring a higher factor of safety throughout the critical zone.

Figure 20 shows the embankment reinforced by the method of the invention with several planes of reinforcement inhibiting failure.

Depending on the function of the embankment it may be desirable to seal with a skin. A surface filter layer overlaid by a bituminous material to protect against erosion may be suitable: grass may be seeded on the surface for environmental reasons.

Figures 21, 22 and 23 illustrate the invention with respect to its use in a retaining wall. In particular Figure 21 shows a retaining wall in which the overall stability of the wall is ensured by one plane of reinforcement. A skin is a required feature of this wall, it may be anchored into the soil with light conventional anchors, or by the method described earlier with reference to Figures 11 and 12.

Figure 22 shows a retaining wall where the stability is ensured by two principal planes of reinforcement 9.

Figure 23 shows a retaining wall where the stability is ensured by several principal horizontal planes of reinforcement 9 and one vertical plane near the toe of the wall.

Each of three possible solutions, details of which are given above, have their own merits depending on the size of the wall, the type of soil used and construction method, the reinforcement material (the form and strength of the members), the facing adopted and the foundation conditions, as the most important considerations.

When using grid reinforcement considerations of design, construction, convenience, most and structural function may make the use of one single continuous reinforcement section impractical or undesirable.

Figure 24 shows a plan of a plane 18 to be reinforced.

A reinforcement 9 continues substantially over the whole of area 17 to reinforce the plane.

In many instances the reinforcement may be sub-divided into suitable smaller units which may be distributed over the area 18. The general scale of failure allows spaces 19 to be left if found convenient.

Figure 25 shows this subdivision with a plan of a plane 18 being reinforced by smaller units 17 with spaces 19. The smaller units of reinforcement may be distributed above and below the principal plane 16 (Figure 26). This may be found desirable especially during construction.

Some other applications of the invention will be briefly described by way of example.

1. A foundation reinforced by the method of the invention. Figure 5 shows a typical failure pattern for a foundation. By placing reinforcement across the rupture surfaces, for example by placing reinforcement members 9 vertically as shown in Figure 27, or horizontally as shown in Figure 28, the bearing capacity may be increased.

Further planes of reinforcement may be placed on either side of the two shown in Figure 27 for extra strengthening.

2. A rock fill dam with a clay core.

Figure 29 shows a dam with rockfill shoulders 24 and a clay core 25. The conventional upstream profile is shown dotted 26. By strengthening the core by the insertion of reinforcement members 9, the failure characteristics of the core may be matched to that of the rockfill, thus allowing a new profile 27.

The merits of the invention when construction of embankments, walls, or dams is being undertaken on weak foundations should be especially emphasised. Failure of these structures often involves at least part of the failure surface passing through the weak foundation material. By reinforcing the critical zones within the structure to a sufficient degree, constraining them to fail rigidly, a constraint has been placed on the number and type of possible overall structural failure mechanisms. For example, if there is a specific zone of foundation weakness suitable reinforcement within the main fill structure will make the foundation weakness less critical. (In some cases it may be considered beneficial to strengthen the foundation weakness itself by reinforcement insitu).

WHAT I CLAIM IS: 1. A method of reinforcing a soil structure, the method comprising the steps of determining the path of potential critical rupture surfaces (as hereinbefore defined) through the structure, and then incorporating wholly within the structure a soil-shear-resisting, rigid reinforcement unconnected with any external surface of the structure in a position in which it extends across the path of the potential critical rupture surfaces so that the reinforcement crossing the rupture surfaces resists straining of the soil in the structure along the rupture surfaces.

2. A method as claimed in claim 1, wherein the soil is coarse grained and the

**WARNING** end of DESC field may overlap start of CLMS **.

Claims (16)

**WARNING** start of CLMS field may overlap end of DESC **. areas outside the critical zone have higher factors of safety. Figure 19 shows the embankment reinforced by the method of the invention with one principal plane of reinforcement 9 ensuring a higher factor of safety throughout the critical zone. Figure 20 shows the embankment reinforced by the method of the invention with several planes of reinforcement inhibiting failure. Depending on the function of the embankment it may be desirable to seal with a skin. A surface filter layer overlaid by a bituminous material to protect against erosion may be suitable: grass may be seeded on the surface for environmental reasons. Figures 21, 22 and 23 illustrate the invention with respect to its use in a retaining wall. In particular Figure 21 shows a retaining wall in which the overall stability of the wall is ensured by one plane of reinforcement. A skin is a required feature of this wall, it may be anchored into the soil with light conventional anchors, or by the method described earlier with reference to Figures 11 and 12. Figure 22 shows a retaining wall where the stability is ensured by two principal planes of reinforcement 9. Figure 23 shows a retaining wall where the stability is ensured by several principal horizontal planes of reinforcement 9 and one vertical plane near the toe of the wall. Each of three possible solutions, details of which are given above, have their own merits depending on the size of the wall, the type of soil used and construction method, the reinforcement material (the form and strength of the members), the facing adopted and the foundation conditions, as the most important considerations. When using grid reinforcement considerations of design, construction, convenience, most and structural function may make the use of one single continuous reinforcement section impractical or undesirable. Figure 24 shows a plan of a plane 18 to be reinforced. A reinforcement 9 continues substantially over the whole of area 17 to reinforce the plane. In many instances the reinforcement may be sub-divided into suitable smaller units which may be distributed over the area 18. The general scale of failure allows spaces 19 to be left if found convenient. Figure 25 shows this subdivision with a plan of a plane 18 being reinforced by smaller units 17 with spaces 19. The smaller units of reinforcement may be distributed above and below the principal plane 16 (Figure 26). This may be found desirable especially during construction. Some other applications of the invention will be briefly described by way of example. 1. A foundation reinforced by the method of the invention. Figure 5 shows a typical failure pattern for a foundation. By placing reinforcement across the rupture surfaces, for example by placing reinforcement members 9 vertically as shown in Figure 27, or horizontally as shown in Figure 28, the bearing capacity may be increased. Further planes of reinforcement may be placed on either side of the two shown in Figure 27 for extra strengthening. 2. A rock fill dam with a clay core. Figure 29 shows a dam with rockfill shoulders 24 and a clay core 25. The conventional upstream profile is shown dotted 26. By strengthening the core by the insertion of reinforcement members 9, the failure characteristics of the core may be matched to that of the rockfill, thus allowing a new profile 27. The merits of the invention when construction of embankments, walls, or dams is being undertaken on weak foundations should be especially emphasised. Failure of these structures often involves at least part of the failure surface passing through the weak foundation material. By reinforcing the critical zones within the structure to a sufficient degree, constraining them to fail rigidly, a constraint has been placed on the number and type of possible overall structural failure mechanisms. For example, if there is a specific zone of foundation weakness suitable reinforcement within the main fill structure will make the foundation weakness less critical. (In some cases it may be considered beneficial to strengthen the foundation weakness itself by reinforcement insitu). WHAT I CLAIM IS:

1. A method of reinforcing a soil structure, the method comprising the steps of determining the path of potential critical rupture surfaces (as hereinbefore defined) through the structure, and then incorporating wholly within the structure a soil-shear-resisting, rigid reinforcement unconnected with any external surface of the structure in a position in which it extends across the path of the potential critical rupture surfaces so that the reinforcement crossing the rupture surfaces resists straining of the soil in the structure along the rupture surfaces.

2. A method as claimed in claim 1, wherein the soil is coarse grained and the

reinforcement is positioned substantially along the minor principal strain direction of the soil at that location.

3. A method as clamed in claim 1 or claim 2, wherein the paths of the potential critical rupture surfaces are determined from a series of limit equilibrium analyses on the unreinforced structure.

4. A method as claimed in any preceding claim, wherein a plurality of reinforcement units are used, spaced along the critical rupture surfaces.

5. A method as claimed in any preceding claim, wherein the reinforcement is in the form of a grid or mesh.

6. A method as claimed in any of claims 1 to 3, wherein the reinforcement is in the form of a bar.

7. A method as claimed in any preceding claim, wherein the reinforcements are incorporated in the structure as it is erected.

8. A method as claimed in any of claims 1 to 6, wherein reinforcements are incorporated in the structure after the structure has been erected.

9. A method as claimed in any preceding claim for reinforcing a soil structure with a sealing skin, wherein the skin comprises a plurality of face units, with anchors attached to the rear faces of at least some of the face units.

10. A method as claimed in claim 9, wherein the length of the anchors, in a direction extending away from the rear faces of the skin units, is not greater than three times the height of the respective face unit.

11. A method as claimed in any of claims 1 to 8 when used for reinforcing a foundation, wherein reinforcements are positioned substantially vertically in the soil around the load-bearing foundation.

12. A method as claimed in any of claims 1 to 8 when used for reinforcing a foundation, wherein reinforcements are positioned substantially horizontally in the soil underneath the load-bearing foundation.

13. A method as claimed in any of claims 1 to 8 when used for reinforcing the clay core of a dam with rockfill shoulders and a clay core, wherein reinforcements are positioned substantially horizontally across the clay core.

14. A reinforced soil structure produced by the method of any one of the preceding claims.

15. A method of reinforcing a soil structure substantially as herein described with reference to the accompanying drawings.

16. A reinforced soil structure substantially as herein described with reference to any one of Figures 19 to 23, 27, 28 or 29 of the accompanying drawings.