An Updated Two-Pager on the Limitations of Simplified Methods for Evaluating Earthquake-Induced Liquefaction and its Consequences

I recently posted “Yet Another Update of Limitations of Simplified Methods for Evaluating Earthquake-Induced Liquefaction and its Consequences” on LinkedIn and elsewhere. The latest update added more detail about the limitations of simplified methods for predicting the magnitude of lateral spreading displacements and emphasizes the fact that if the initial evaluation for triggering of liquefaction is conservative, then the predicted settlements and the predicted lateral spreading displacements will also be conservative. However, while you should carefully read the twenty-page version, I believe that a two-page summary might also be helpful. This update simply adds a paragraph to the conclusions regarding risk-informed decision making.

The original simplified method was a brilliant way of making it possible for engineers to better understand and to make preliminary evaluations of the potential for liquefaction at a time when there were only a few people who could run even equivalent linear site response analyses. However, times change, and fifty years later, there are many more geotechnical engineers with graduate education who are capable of running nonlinear effective stress site response analyses in order to obtain better estimates of the cyclic shear stresses and strains and excess pore pressures in critical layers.

In the meantime, the extensions of the simplified method have become too complicated and contain too many poorly understood assumptions. Just as an example, although using energy measures as an indicator of the duration of loading is a valid idea, to add this, or probabilistic approaches, on to simplified methods is to put lipstick on a pig.

If asked to give the two main limitations of simplified methods, I would say they are: (1) working backwards from a peak ground surface acceleration that is inconsistent with the particular soil profile; and (2) using cyclic stress ratios that trigger initial liquefaction that are based on penetration resistance and collections of case histories in which sites that did not liquefy are greatly under-represented. This results in the effects of fines, the method of deposition, and ageing being underrepresented. See Bray (2022) and Kayen (2024) for two good examples of the significant effects of fines and the method of deposition.

How do I know that simplified methods tend to be very conservative? Because in my consulting practice I am frequently asked to look at sites where simplified methods have predicted clearly excessive settlements and lateral spreading relative to what would be expected based on the geology, that is the depositional history, of the site and the recorded history of liquefaction with engineering consequences on similar sites in areas with similar tectonics and seismicity. Intelligent use of bi-directional nonlinear effective stress site response analyses on these sites gives results which are much more consistent with our knowledge of the observed responses of similar profiles in historic events. Although as a matter of good practice engineers should understand the methods they are using and their limitations, it is not necessary to understand in detail all the limitations of Ishihara and Yoshimine (1992) on predicting the settlement of saturated sands or of Zhang et al. (2004) on predicting lateral spreading deformations to know that the results are frequently not consistent with reality.

The simplified methods for evaluating liquefaction, settlement and lateral spreading are particularly inappropriate in the context of risk-informed decision making, which is increasingly being adopted worldwide. It makes no sense to conduct quantitative risk analyses using analytical methods that are both not very accurate and have a built-in conservative bias.

So, what is the role for simplified analyses today? At best they are appropriate for screening analyses. If they indicate that there is no problem, then likely there is no problem. But a better screening analysis is simply to ask the question, “has there ever been liquefaction with engineering consequences in a similar soil profile in an area with similar tectonics and seismicity.” If the answer is no, then there is no problem. If there is a history of similar soil profiles exhibiting problems, then you should conduct a proper site investigation with paired borings with SPTs (so that you can run PIs and full particle size distributions) and SCPTs and conduct bi-directional nonlinear effective stress site response analysis. In the case of particularly high value projects, conduct an even more complete field investigation with triplets that include a second boring to obtain relatively undisturbed samples for laboratory testing. See Crawford et al. (2019). [See the longer article, “Yet Another Update …” for these references.]

Finally, it doesn’t require any analysis to conclude that it is not a good idea to build apartment blocks with shallow or no foundations on recently deposited loose sands or silts where there is a high water table. Predicting the exact consequences of doing this is very difficult, so it is best just not to do it!