Weak Layer Overview
The Weak Layer option allows you to define a surface with its own material properties. This can be used to model any type of joint, discontinuity or interface along which failure might occur.
There are two options available to define a Weak Layer surface:
See Weak Surface Tutorial and Multiple Weak Layers Tutorial for examples of how to use the Weak Layer options.
How weak layers work in Slide3
If there are weak layers in the model and they intersect a slip surface generated during the search (or user-defined surfaces), then the following options (available via Slip Surface Options > Weak Layer Handling) will determine the logic for clipping the slip surfaces to the weak layers:
- Always snap to highest layer – An envelope of weak layers will be used to clip the slip surface. More specifically, if an active weak layer exists above any given point on a slip surface, then the elevation of the slip surface will snap to the highest weak layer at that location.
- Automatic case generation – For a given slip surface and the weak layers that it touches, the program will determine all applicable combinations of toggling each weak layer on and off and attempt to evaluate the factor of safety for each case. Note that for models with many weak layers, this option may slow down the search significantly.
- Heuristic – Toggles the weak layers on and off during the random search process (available only for Particle Swarm Optimization and adapted from Kennedy and Eberhart (1997)). Weak layers producing lower factors of safety tend to remain toggled more often as the search progresses. It is much more efficient than the Automatic case generation for models with many weak layers.
See Multiple Weak Layers Tutorial for a demonstration of how these options can be used. Some additional details about the logic for clipping slip surfaces to the weak layers are discussed below.
Entire weak layer
Note that each weak layer defined in a model is assumed to exist in its entirety. Slip surfaces affected by the weak layer are either cut by the entire weak layer or, when considering the various cases during automatic case generation, not cut at all.
Suppress weak layer
It is possible to completely exclude a weak layer defined in the model from the analysis, without having to delete it. You can do this by checking off the “Suppress” checkbox in the weak layer properties.
Discontinuous layers
If a weak layer enters a given slip mass but does not fully exit the mass, either via the ground surface or through the slip surface, then it is discontinuous.
Discontinuous weak layers can significantly increase the difficulty of the search routine for finding the critical slip surface. For example, the image shows a blue weak layer that ends inside of a slip mass (delineated by the red line) near the ground surface.
The discretization process will initially snap the portion of the slip surface found below this weak layer up to the weak layer. In Slide2, the rest of the slip surface will not be changed, resulting in a slip surface with a local discontinuity which may render it invalid. However, for Slide3 an extrapolation scheme described in the following section will be attempted to produce a valid surface.
Horizontal extrapolation
For the above example, Slide2 will discard the slip surface with the local discontinuity shown, because it is relatively easy to generate a better slip surface such as the one shown in the following image. This is because in Slide2, slip surfaces can be drawn using simple polylines or circular arcs.
However, in Slide3, the geometrical parameters defining the slip surfaces are much more complicated. For this reason, it can sometimes be helpful to apply an extrapolation of the discontinuous layer to cut the slip surface beyond the extents of the weak layer, as shown in the figure below.
To help visualize this, a few examples of how this logic is applied are shown below.
Finally, please be advised that the extrapolation of discontinuous weak layers may cause material blocks beneath the extrapolation to be ignored in the analysis. For example, the grey layer in the example below will be missed due to the extrapolation.
The grey layer will only be analyzed for the case where the blue weak layer is not considered (i.e. during automatic case generation).
Results naming
For a given user-defined slip surface that intersects multiple weak layers, there will be several cases evaluated if Automatic Case Generation is selected. This will produce multiple slip surfaces, which if they converge, will show up in the Show All Surfaces interface with suffixes a, b, c, etc.
Vertical weak layers
Slide attempts to cut slip surfaces using the weak layers that they intersect. Sometimes, discretization issues can occur when these cuts are vertical or near-vertical.
Case a) Vertical cuts in tension zones
In many cases, vertical weak layers can produce valid slip surfaces for analysis. For example, the slip surface below (blue) is cut vertically in a zone of tension by the vertical face of a weak layer (red). The resulting slip surface and slices / columns for analysis is highlighted in yellow. This slip surface is valid because in the zone of tension, no reaction force is assumed to be exerted by the soil to the right side of the cut.
Case b) Vertical cuts in compression zones
Suppose the weak layer is modified such that it also cuts the slip surface in a zone of compression near the base of the slope, as shown below. The resulting slip surface is invalid, because the restoring lateral force exerted by the soil left of the cut in the zone of compression is non-zero, and cannot be determined using limit equilibrium methods.
If this case is ever encountered, Slide will report Error -149 and prompt a warning when viewing the results. However, if the desired analysis is for the slip surface to pass through (and effectively ignore) the weak layer in the zone of compression, then there are several ways to bypass the error:
- Replace the vertical faces of the weak layers with Tension Cracks, which vertically cut the slip surfaces only if the soil is in tension at those locations.
- Alter or delete the left vertical portion of the weak layer so that it does not vertically intersect the slip surface (like case (a)).
- Split up the weak layer geometry into separately defined weak layers, and use the Weak Layer Handling feature.
- Use more advanced techniques such as finite element analysis in RS2 & RS3, which consider the behavior of the entire slope.
Case c) Vertical cuts in compression zones
Near-vertical weak layers can also create columns with very steep base angles, which are known to induce numerical instability in limit equilibrium methods and should be avoided if possible.
If the base angle of a column, θ, exceeds the maximum allowable angle, θmax, then the slip surface will be discarded with Error -413. The default value of θmax is 80°, which can be changed via Project Settings > Advanced).
References
J. Kennedy and R. C. Eberhart, "A discrete binary version of the particle swarm algorithm," 1997, IEEE International Conference on Systems, Man, and Cybernetics. Computational Cybernetics and Simulation, 1997, pp. 4104-4108 vol.5, doi: 10.1109/ICSMC.1997.637339.