Two Identical Slope/W Models Returning Different Factors of Safety

RyanGoldsmith
RyanGoldsmith Posts: 2
in GeoStudio

I have two approximately identical slope/w models of an existing earthen embankment dam with different factors of safety. Both models consider a failure of the upstream slope of the embankment. The only difference between the two models is how the approximately same phreatic surface is input.

"Model 1" has a phreatic surface defined by a parent seep/w analysis and "Model 2" is a clone of "Model 1" that has a piezometric surface manually defined by a coordinate-traced-copy of the phreatic surface developed by the "Model 1" seep/w parent analysis. "Model 2" does have a slightly lowered phreatic surface near the midslope of the downstream slope (to represent a proposed drain), but I don't believe this should affect the pore water conditions in the slip surfaces since the failure is of the upstream slope where the phreatic surfaces for both models are essentially equivalent.

I've printed the model report for both models and can confirm that all other inputs are the same (side function, slip surface optimization, entry and exit points, convergence settings, materials, etc.). "Model 1" has a factor of safety that is three tenths (0.3) higher than "Model 2". Does anyone have any insight as to what could be causing this discrepancy?

Thanks

Answers

  • TrevorKent
    TrevorKent Posts: 10

    Hi Ryan, thanks for reaching out. The discrepancy comes from the fact that Model 1 gets its water conditions from SEEP/W and Model 2 gets its water conditions from a piezometric line. SLOPE/W tends to overestimate the pore-water pressure when the phreatic surface is defined with a piezometric line that is downward sloping (to represent downward flow), which is usually the case in an embankment. This is because it uses the vertical distance from the base of a slice to the phreatic surface to calculate the pore-water pressure.

    When SLOPE/W uses the water conditions calculated with SEEP/W, the pore-water pressure values at the base of the slices are more accurate. In the case of a downward sloping phreatic surface, the pore-water pressure used in SLOPE/W will be lower due to the equipotential lines being angled. I've set up an example where you can see that even on the upstream slope of the dam, the equipotential lines are different. The first picture is with the drawn piezometric surface and the second picture is with the SEEP/W pore-water pressure conditions.

    image.png
    image.png

  • RyanGoldsmith
    RyanGoldsmith Posts: 2

    Hey Trevor, I see the different equipotential lines in each of my two models — makes sense. Thanks for the response and detailed explanation!

  • TrevorKent
    TrevorKent Posts: 10

    Awesome! I'm glad I was able to clear that up for you.

Categories