Harmonizing engineering geology with rock engineering for assessing slope stability
The construction, design, remediation and maintenance of rock slopes have always been an important area in geo-engineering. Particularly, in the last two decades, increasing demand for ultra-deep open pits reaching up to depths greater than 1000 m and large civil engineering constructions in rocks such as expressways, highways, railways and dams, and the effects of earthquake-triggered landslides on settlements located in mountainous regions resulted in more attention to be paid to rock slope stability. Progresses in understanding, analysis and control of rock slope movements have been the result of interdisciplinary efforts mainly involving engineering geologists and rock engineers.
Engineering geology is the discipline of applying geologic data, techniques and principles to the study of rock and soil materials, surface and subsurface fluids, and interactions of introduced materials and processes with the environment so that geologic factors are adequately recognized, interpreted and presented for use in engineering and related practice. The engineering geologist, as a predictor, translates the scientific facts, observed or measured, into engineering data to identify areas of significant physical constraint that will adversely affect the design, construction and maintenance of any intended engineering project (Figure 1).
Most of the major practitioners in applied geology in the 19th and 20th centuries have contributed significantly to understanding of slope movement types and processes. Particularly, the experiences gained from the slope failures during the construction of Panama Canal (Figure 2), and at Malpasset (France) and Vaiont (Italy) (Figure 3) dams are the important milestones in terms of engineering geology. These cases showed that engineering geological models mainly including the effect of discontinuities and identification of accurate failure mode have vital importance in rock slope stability assessments. At the end of 19th and in the first half of 20th centuries, contributions from engineering geology mainly were directed toward developing unified theories of slope formation and evolution and preliminary slope movement classifications. Then the more detailed slope classification, which is the most popular one in geo-engineering, has been developed by Varnes. Development of Schmidt net and stereographic projection technique and description of discontinuity properties and their quantitative determination are the other contributions provided by engineering geology.
In the 1960s, with the beginning of world-wide use of “rock mechanics” as a young science and engineering discipline, existing uncertainties associated with rock slopes have been clarified and some significant developments on theoretical, experimental and numerical aspects for behaviour, analysis and stabilization of rock slopes have been achieved. Method of kinematic analysis of rock slopes and 2-D limit equilibrium methods of analysis and assessment of remedial measures for structurally-controlled rock slope failures large scale field and laboratory shear testing of rock discontinuities, the use of computerised test data collection, development of rock mass classification systems, such as RMR and Q, and the empirical shear failure criteria for rough-undulating discontinuities and for jointed rock masses are the main contributions of rock mechanics and rock engineering to rock slope engineering. In addition, monitoring the performance of rock slopes had been carried out for many years and had become an integral part of rock slope engineering. More recently, integration of the experimental and theoretical issues in rock mechanics and rock engineering with the computer technology well developed in the second half of 20th century and the use of numerical methods became very popular in rock engineering.
Numerical methods have also been one of the major methods applied to rock slopes.
Engineering geological inputs in slope stability assessments
The input from engineering geology is a pre-requisite in all stages of rock slope engineering. Failure to take into account engineering geological factors or inadequate inputs or considerations with respect to geological features with a particular slope can lead to slope failure with accompanying serious consequences. A comprehensive engineering geological model, based on four main factors/inputs such as lithology (rock type), structure (discontinuities), state of degradation and hydrogeological conditions prevailing in rock slopes, is fundamental in rock slope stability assessments. In addition, external forces, such as dynamic loading due to earthquakes are also important for slopes located at earthquake-prone regions. Correct selection of geomechanical parameters of discontinuities and rock masses is another issue having vital importance on the stability assessments.
Due to the different nature and origin of the rock types, their inherent geological features are also different. In softer or weaker rocks such as schist and shale, the material itself can be a predominant controlling factor in the slope instability, whereas, in the hard rocks such as granites and limestones, the major discontinuities control the stability.
In addition, limestones have solution features, which are called karst features originated from solution along discontinuities, may also contribute to trigger failures, particularly in sub-vertical or overhanging cliff slopes.
On the other hand, another important rock slope instability problem associated with rock type is commonly experienced in block-in-matrix rocks (BIMROCKS), which are a mixture of rocks composed of geotechnically significant blocks within a bonded matrix of finer texture such as melanges, fault rocks and other complex geological mixtures.
Bimrocks maybe sometimes mischaracterized due to their considerable spatial, lithological and mechanical variability, and therefore, their correct characterization is necessary to reduce expansive and inconvenient surprises during slope construction.
The most important factor controlling the stability of slopes in jointed rock masses is discontinuities such as bedding, fault, joint, schistosity surfaces etc. and the adverse interaction of their orientations with those of slopes is the greatest contributing factor to rock slope instability. Therefore, the success of rock slope stability assessments depends on the level of understanding of the characteristics of discontinuities, such as orientation, spacing, persistence, roughness, infilling etc., as described by ISRM. Today geological and geotechnical data collection techniques are well developed. However, good quality data collection from discontinuities and the efficient use of the data are important. From the practical point of view, it is the best to measure the most of the characteristics of critical geologic structures from surface exposures using different techniques such as scan-line survey, window mapping, photogrammetric method or laser scanning technique. However, in cases where it is usually limited surface area that is exposed and accessible for surface structure, the orientation of discontinuities should be determined from orientated cores and/or telewiever logs in conjunction, if possible.
Depending on the control of discontinuities on the slope, two main types of rock slope failure may occur.
One of them is the kinematically possible structurally controlled failures occurring along discontinuities or line of intersection of two discontinuities or along columnar structures adversely dipping to the slope face (Figure 4a-c).
The second type occurs when the rock mass involves closely spaced discontinuities, which fracture intact rock over small distances, and linear failure planes approaching a circular surface along the whole failure surface, as observed in soil slopes, develop partly through intact rock and partly following discontinuities, resulting in non-structurally controlled rock mass failure (Figure 4d).
Limited or unrealistic assessment of discontinuity characteristics, particularly of orientation, spacing and persistence may result in misconceptions on the modes of failure and on block sizes used in analyses generally different than that of expected in reality, and consequently unrealistic engineering geological models be established for the rock slopes. Depending on increase in slope height, rock slope failures may involve several composite mechanisms.
As illustrated in Figure 5, sliding may occur along relatively short adversely dipping discontinuities accompanied by dilation of the mass as the failure path steps up on other sub-vertical discontinuity sets within the rock mass and by shearing through and/or tensile failure of the intact rock and/or rock mass bridges between the discontinuities.
Effect of Degradation
It is well known that rocks may undergo degradation when they are exposed to atmospheric conditions and/or hydrothermal fluids through rock mass. The degradation process is commonly known as deterioration in the forms of weathering or alteration depending upon the physical and chemical process involved. Due to weathering hard rocks transform into soft rocks which maintain the structure of the intact rocks, but are characterized by higher void ratios and reduced bond strengths. Soft rocks are transformed into thick granular soil mantles generally called residual soils.
The degradation process also influences the joint spacing and discontinuity filling material in the form of clay.
Significant progressive deterioration of rock slopes may occur in engineering time, often giving rise to the need for unplanned maintenance and constituting a safety hazard. Superimposed on the lithology and structures, the physical and chemical weathering effects can be predominant in controlling the modes of rock slope failure. For example, open sheet/exfoliation joints, which are formed by stress relief as a result of physical weathering of granitic rocks, tend to run sub-parallel to the rock head with a tendency to daylight in the cut slopes and may result in a structurally controlled instability such as planar failure (Figure 6a). On the contrary to this, if the slope forming rock mass has fully transformed into soil due to intense chemical weathering (weathering grades V or VI), the slope will fail in the form of circular sliding as commonly observed in soil slopes (Figure 6b).
Another type of weathering, which is a common source of rock slope instability particularly along steep slopes, is differential weathering, which generally occurs in cases where slopes consist of inter-layered hard (e.g. sandstones, limestones) and soft rock units (e.g. shales, claystones and mudstones or tuffs), and may not be recognized sometimes.
As differential weathering of softer rocks (highly prone to weathering) progresses, undercutting can lead to loss of support and collapse (falls or topples) of the overlying material (Figure 6c). Large-scale, sub-vertical discontinuities and over-break fractures may promote failure by intersecting undercut surfaces and/or reducing block size within the rock mass.
Deterioration, however, is rarely given much attention at the design stage; emphasis is on the avoidance of deep-seated failures as seen in Figure 6d. But the above given examples emphasize the importance of type, degree, product and depth of the deterioration, which are the main inputs provided by engineering geology, on the mode of slope failure anticipated and on drastic changes in strength and deformability characteristics of the slope forming the rock mass.
Hydrogeological Conditions and Conceptual Model
Water pressure in the discontinuities in a rock mass reduces the effective stresses on such discontinuities with a consequent reduction in shear strength. Therefore, the presence of groundwater in a rock slope is an important factor affecting its stability. The procedure to define an engineering geological modelling involves the definition of a conceptual hydrogeological model that must be very close to the reality and reproduce groundwater flow patterns (fitting computed and measured groundwater heads) with consistent hydraulic parameters.
In a hydrogeological conceptual model, three major phases to be accomplished can be defined: a baseline study as a first phase, conceptualization and characterization phase and prediction as the final phase. Following the baseline hydrogeological study, performed to assess the occurrence and potential of groundwater at and around the slope site, a
detailed study program is developed and achieved to construct a hydrogeological conceptual model of the site.
The conceptual model construction is essentially based on delineation and characterization of the hydrostratigraphic units in the area and acquisition and analysis of hydrometeorological data. This phase requires extensive fieldwork such as surface and subsurface mapping, drilling, in-situ groundwater level and quality observations and field tests.
Upon development of a conceptual hydrogeological model of the site, the third main phase is accomplished to predict the Spatio-temporal groundwater head and flow rate distribution. This enables to assess not only the type and extent of the threat of groundwater on excavation activities in terms of inflow to excavation and slope-stability but also the impacts of excavation on groundwater resources at and around the site. A flowchart of hydrogeological studies integrated to rock slopes is given in Figure 7 as an example.
Effect of Earthquakes on Slope Stability
In natural and engineered slopes, earthquake motions can generate significant horizontal and vertical dynamic forces and increase the shear stress on potential failure surfaces. Unlike gravity and other static forces, earthquake forces are not likely to influence a slope during its design lifetime. However, depending on the characteristics of the slopes and strong ground motion, slope instabilities in different forms may occur. The first step in assessing the influence of seismic forces on slopes is to obtain geologic inputs including the prehistoric seismicity, anticipated magnitude based on the active faults in the vicinity and seismic hazard analysis and choosing suitable attenuation function for the rock to estimate the peak ground acceleration to be used in slope stability analyses.
Resat Ulusay; DGT BiH Geotehnika