Newmark Deformation Analysis

GEO-SLOPE International Ltd. | www.geo-slope.com

1200, 700 - 6th Ave SW, Calgary, AB, Canada T2P 0T8
Main: +1 403 269 2002 | Fax: +1 888 463 2239

Introduction
This example demonstrates how the results of a QUAKE/W analysis can be used in conjunction with
SLOPE/W to estimate the permanent deformations that may occur as a result of the inertial forces
associated with an earthquake.

This type of permanent deformation analysis is referred to as a Newmark analysis. The premise of a
Newmark analysis is that during an earthquake, there will be short moments in time when the inertial
forces (mass times acceleration) plus the initial static forces will exceed the available shear
resistance and during these times the temporary loss of stability will lead to un-recoverable
deformations. The accumulation of the un-recoverable deformations will manifest itself as
permanent deformation after the shaking has stopped.

The details of the implementation for this type of analysis are presented in the SLOPE/W Engineering
Book. Only the highlights of an illustrative example are presented here.

Numerical Simulation
A similar example is presented as part of the SLOPE/W group of examples. A variation of the
SLOPE/W example is presented to enhance the QUAKE/W example list and for convenient reference
when one is examining the QUAKE/W features and capabilities. Figure 1 presents the simple slope
used in this illustrative analysis.

1
Figure 1. Problem configuration.

The first step is to establish the in situ state of stress in the ground that exists before the earthquake
occurs and shakes the ground. In this case, this is done with the QUAKE/W Initial static type of
analysis, which basically involves doing a simple gravity turn-on type of analysis.

The next step is to subject the slope to an earthquake with a time history record shown in Figure 2.
The peak acceleration is 0.4 g at about the 2-sec mark in the 10 seconds of shaking. The earthquake
data for this record has a constant time interval of 0.02 seconds. For 10 seconds, this means there are
500 data points. The output data is saved to file only every 10th time step plus at 20 of the highest
peaks in the record.

0.4

0.3

0.2
Acceleration (g)

0.1

-0.1

-0.2

-0.3

-0.4
0 2 4 6 8 10

Time (sec)

Figure 2. Earthquake time history record.

Only linear-elastic soil properties are used here in this simple illustrative example. The total unit
weight is defined as 18 kN/m3 with a Poisson’s ratio of 0.334 and a Damping Ratio of 0.1. The Gmax is
defined at a constant 20,000 kPa. The left and right boundary conditions are fixed in the x-direction
in the Initial Static analysis, but then fixed only in the y-direction in the Shaking analysis.

2
Slope stability analyses have been developed for both the initial static analysis to determine the
initial factor of safety and again during the shaking analysis to determine the factor of safety versus
time during the earthquake.

Results and Discussion

The results of the Initial Static analysis are the most conveniently inspected by creating contour plots
of the various stress components. Figure 3 shows a typical example.

Figure 3. Maximum static shear stress contours.

Under static conditions, the factor of safety is about 1.13 (Figure 4).

1.129

20
18
16
14
Elevation - m

12
10
8
6
4
2
0
-2
0 10 20 30 40 50 60

Distance - m

Figure 4. Stability of slope under static conditions.

Within the red band in Figure 4, the Factor of Safety varies between 1.129 and 1.229 – about a 10%
variation.

It is intuitively obvious that the earthquake inertial forces will pull the slope in both the up-slope
direction and in the down-slope direction. This is illustrated in Figure 5 and in Figure 6. Figure 7
shows the lateral back and forth displacement of the slope crest.

3
Figure 5. Displacement at the 1.60 second mark (100x exaggeration).

Figure 6. Displacement at the 5.40 second mark (100x exaggeration).

0.1

0.05
Relative X-Displacement (m)

-0.05

-0.1

-0.15

-0.2
0 2 4 6 8 10

Time (sec)

Figure 7. Lateral back and forth movement of the slope crest during the shaking.

From the illustrations in the above figures, it is easy to visualize that the factor of safety will go up
and down during the shaking. The objective is to determine how many times and for how long the
factor of safety is below 1.0 when there is a momentary loss of stability.

4
The details of how finite element-computed stresses can be used in SLOPE/W to compute safety
factors are presented in the SLOPE/W Engineering Book. It is sufficient to say here that for each trial
slip surface, SLOPE/W computes the mobilized shear and normal stress at the base of each slice. The
normal stress is then used in conjunction with the specified strength parameters to compute the
available shear resistance at the base of each slice. A ratio of the summation of the available shear
resistance to the summation of the mobilized shear is taken as the safety factor.

For a Newmark-type analysis, the soil strength is deemed to remain constant during the shaking
(undrained behavior). The strength is therefore computed on the basis of the initial static stresses.
Only the mobilized shear is re-computed for each time step that the data is saved to file.

The permanent deformation is computed for 18 trial slip surfaces in this example.

The largest computed deformations occur for Slip Surface 2. The static factor of safety for this slip
surface is 1.197, as shown in Figure 8. Note, it is not the slip with the lowest static factor of safety.

1.197

Figure 8. Slip surface with the largest deformation.

Figure 9 presents the variation in the safety factors during the shaking for this slip surface. Note that
there are moments when the factor of safety is less than unity (1.0), indicating that there are
moments when the slope is temporarily unstable.
Factor of Safety vs. Time
4

3
Factor of Safety

0
0 2 4 6 8 10

Time (sec)

Figure 9. Variation in safety factors during the shaking.

5
SLOPE/W computes the so-called yield acceleration from the data in Figure 10. The yield acceleration
occurs where the factor of safety is 1.0. In this case, the yield acceleration is about 0.10g.
Factor of Safety vs. Average Acceleration
4

3
Factor of Safety

0
-0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4

Average Acceleration (m/sec²)

Figure 10. Factor of safety versus the average acceleration.

Once the yield acceleration is known, SLOPE/W integrates the area under the curve in Figure 11,
where the acceleration is greater than the yield acceleration, which in this case is 0.10 g. The
integrated area gives the velocity as shown in Figure 12.
Average Acceleration vs. Time
0.4

0.3
Average Acceleration (m/sec²)

0.2

0.1

-0.1

-0.2

-0.3

-0.4
0 2 4 6 8 10

Time (sec)

Figure 11. Average acceleration with time.

6
 Velocity vs. Time
0.4

0.3
Velocity (m/sec)

0.2

0.1

0
0 2 4 6 8 10

Time (sec)

Figure 12. Velocity during moments of temporary loss of stability.

Integrating the area under the velocity curve gives the accumulated deformation, as shown in Figure
13.
Deformation vs. Time
0.4

0.3
Deformation (m)

0.2

0.1

0
0 2 4 6 8 10

Time (sec)

Figure 13. Accumulated deformations during moments of temporary loss of stability.

In summary, this analysis indicates that there could be approximately 0.28 m permanent deformation
of this slope if it were to be subjected to the earthquake.

The movement would be along the trial slip surface. In this case, with the slip surface being so steep,
most of the movement would manifest itself as vertical movement, although there could be a slight
amount of rotational movement as well.

Summary and Conclusions

It is vitally important to be mindful of the fact that a Newmark analysis involves only inertial forces
and is applicable only to cases where there will not be a significant loss in shear strength and the
generation of significant excess pore-water pressures.

7
The Newmark analysis is very computationally demanding. For each trial slip surface, SLOPE/W
computes a factor of safety for each QUAKE/W time step for which the data was saved. The
computing time multiplies very quickly if there are a lot of QUAKE/W data files and many trial slip
surfaces.

The recommended procedure is to establish the critical slip surface first, based on the initial static
stresses. Then select a few trial slip surfaces around the static critical slip for a Newmark deformation
analysis.

The QUAKE/W – SLOPE/W integration makes it possible to do a Newmark-type permanent

deformation analysis for any potential slip surface.