I don't know of any comprehensive treatment of the subjuct. You might find one that some professor has put together in a book, but I doubt it.

My opinion is more or less the following:

1. The intent of any specific FOS is to account for unknowns, variability in material properties, variability in groundwater conditions, variability in loads, and provide some degree of assurance that the slope is stable.
2. With short term conditions, the amount of expected variability in some of the conditions above is less, just do to less passage of time. Therefor a lower FOS results in the same degree of assurance.
3. With long term conditions, the amount of variability is much greater in most if not all of the variables listed above, therefore the assurance is less.

Mike Lambert

To: GeoPaveTraffic

Something tells me that variability in material properties, groundwater conditions, loads etc. can be applied more or less equally to both short-term and long-term conditions. May it be related to the probability of loading conditions and consequences of failure instead?

Actually the variability in parameters and conditions is different between short term and long term. This is particuarly true with water levels and changes in geometry.

Consequences of failure are generally the same with long and short term.

Mike Lambert

To add to the above, often when short term load conditions control design, such as construction of an embankment on soft foundations, factors of safety can control lateral displacement during and immediately after construction and then with time the foundation consolidates and gains strength so that stability will improve (i.e. if you can successfully build it, it will get better with time) so you might accept a lower FS at end of construction. There may also be situations where consequences do differ between short and long term conditions such as a newly constructed levee with no flood load vs. a levee with a flood load where you could want a higher reliability during loading if you have high consequences.

To: GeoPaveTraffic

Can you please elaborate a bit more on variability of material properties between short term and long term conditions?

Regarding the consequences of failure, you say that there is no difference between consequences of dam failing right after construction (before it fills up) or dam failing 2,3,4 years later with a full reservoir behind it?

It is more related to the risk and consequence of failure, though the material properties selection may contribute to it as well i.e. for short term stability you may use the undrained shear strength which can be measured quite easily in the laboratory, where as for the long term stability, it is very trouble some to estimate the pore pressure for full reservoir condition theoretically or in laboratory without knowing the actual water content of your embankment, hence the higher safety factor for the latter.

Slide during construction may not cause major loss of life or great property damage, so we are justified accepting lower safety factors than would be for when the dam is in operation i.e., when there is water in the reservoir.

For fine grained soil foundation/embankment, the short term stability is mainly governed by the dissipation of construction pore pressure. Since this pore pressures are likely to exist only during the first few years of the life of dam, it is desirable to use less conservative design than otherwise required. Though you may want to inform your client on the risk and the cost involved.