The Changing Face of Soil Mechanics

by Delwyn G. Fredlund

Geotechnical engineers often comment on the youthfulness of Soil Mechanics as a science. I was only 3 years old when Karl Terzaghi published his English version of Theoretical Soil Mechanics, and I still like to consider myself "young".

Many things have changed in geotechnical engineering since the 1940s. The way we perform our subsurface investigations has remained quite similar. Boreholes are still drilled with disturbed and undisturbed soil samples taken at intervals for later laboratory testing. However, the manner in which we obtain our engineering solutions to problems has changed dramatically. Terzaghi and his contemporaries assembled the context for soil mechanics at a time when the tools for solving mathematical problems were significantly different than what is available today.

In the 1940s, the writers of soil mechanics' textbooks attempted to take complex three-dimensional, real-world problems and reduce them to simplified, closed-form solutions. Flownets provided a solution for the movement of water through an isotropic, homogeneous, two-dimensional porous continuum. Methods of (Vertical) Slices provided a solution for calculating the factor of safety of a two-dimensional slope. Method of (horizontal) Layers provided a solution for the calculation of settlement below a one-dimensional, compressible clay soil. The soil mechanics world contained a series of soil property constants (e.g., k, c', and φ'), and those soil properties that were not constants were linearized to become constants (e.g., Cc and Cs).

In the 1960s and 1970s it became clear from research that unsaturated soil properties would have to be defined as "nonlinear unsaturated soil property functions", USPF. Unsaturated Soil Mechanics became a vibrant area of geotechnical research and it was apparent that we were entering a new era that required a new "applications" paradigm. If unsaturated soil mechanics was to find its way into geotechnical engineering practice there needed to be reliable methodologies for obtaining the unsaturated soil property functions at reasonable cost and effort. Consequently, a wide variety of "estimation procedures" have emerged that rely upon the saturated soil properties and an understanding of the "soil-water characteristic curve", SWCC (i.e., water content versus soil suction).

The 1960s and 1970s were also decades that witnessed rapid growth in our ability to solve complex mathematical formulations. This led to a new mathematical framework for the visualization of saturated-unsaturated soil mechanics problems. Numerical methods of solution emerged for all areas of material behavior; areas that spanned well beyond soil mechanics. Soil mechanics problems were visualized as "boundary-value" problems with the following conditions defined; namely: i.) geometry and stratigraphy, ii.) initial and boundary conditions, and, iii.) soil properties. The physics of soil behavior was defined for a "Referential Elemental Volume", REV, of the continuum and the formulation took the form of a "partial differential equation", PDE. And so the equations that many of us so strongly disliked became the heart of problem solving. Fortunately, we can, to a large extent, hide the PDE in a black-box we call "computer software".

Geotechnical engineers need to also be grateful there was another science growth area during the 1960s and 1970s; namely, the field of computer technology (i.e., computer hardware and software). The stage was set for solving saturated-unsaturated soil mechanics problems through use of numerical modeling.

It is an understatement to say that the digital computer has revolutionized the way that soil mechanics is put into engineering practice. In fact, it is safe to say that it would not be possible to simultaneously handle saturated-unsaturated soil mechanics problems within a science framework without the power of the digital computer.

And so, Geotechnical Engineering has moved into a "New Paradigm"; a problem-solving world involving SWCCs, USPFs and PDEs! It is a world in which the challenge becomes "convergence" and "uniqueness" of the soil mechanics solution. It is a world in which computer software is no longer a luxury but a necessity for sound engineering practice.

Delwyn G. Fredlund
Senior Geotechnical Engineering Specialist,
Golder Associates Ltd.,
145 – 1st Avenue North,
Saskatoon, SK., Canada, S7K 1W6