Professor Anil K.

Chopra
Johnson Chair in Engineering
University of California
Department of Civil and Environmental Engineering
Berkeley, CA 94720-1710 USA
[email protected]

Dynamics of structures, earthquake engineering

Although I had known Luis Esteva for many years, the opportunity to work closely with
him came only a few years ago when he become President of the International Association for
Earthquake Engineering (IAEE). I interacted with Luis in my role as Editor of Earthquake
Engineering and Structural Dynamics, the official journal of IAEE. During the period 2002-
2004, the journal underwent major changes, which involved difficult negotiations with other
organizations. I went to him several times asking for advice; in fact, so many times that I am sure
that he grew to dread my emails. Every time I asked for his help, I admired his ability to identify
immediately the essence of the problem, how he was able present his views calmly and
effectively in the midst of disagreement, and was steadfast in searching for solutions that were
fair to all parties.

Anil K. Chopra
August 12, 2005
 ESTIMATING SEISMIC DEMANDS FOR STRUCTURES IN ENGINEERING
PRACTICE: CONSTRAINED BY COMPUTER SOFTWARE, BUILDING CODES, AND
BUILDING EVALUATION GUIDELINES

ANIL K. CHOPRA

ABSTRACT

The main thesis of this paper is that adoption of improved analytical procedures for
estimating seismic demands for structures in engineering practice is constrained by the lack of
computer software that implement these techniques and by prevailing codes of practice. Two
classes of structures, concrete dams and buildings, are chosen to illustrate different aspects of
this contention.

Part A: Arch Dams

Analysis Procedures

The dynamic analysis of arch dams is especially complicated because they must be
treated as three-dimensional systems that recognize the semi-unbounded size of the reservoir and
foundation rock domains, and the following factors should be considered: dam-water interaction,
wave absorption at the reservoir boundary, water compressibility, and dam-foundation rock
interaction.

A series of Ph.D. theses [Chakrabarti and Chopra 1973; Hall and Chopra 1980; Fenves
and Chopra 1984 (a); Fok and Chopra 1985; Zhang and Chopra 1990; Tan and Chopra, 1995] at
the University of California, Berkeley, during 1970 to 1995, culminated in the substructure
method formulated in the frequency domain, and its implementation in computer programs
EAGD-84 for two-dimensional analysis of concrete gravity dams [Fenves and Chopra 1984 (b)]
and EACD-3D-96 for three-dimensional analysis of arch dams [Tan and Chopra 1996] both of
which were distributed pro bono by the National Information Service for Earthquake
Engineering. Extensive parametric studies of the earthquake response of actual dams led to an
understanding of how each of the preceding factors influences the response, and the practical
range of conditions where each factor is significant.

Commonly used finite element analysis techniques for dams, however, ignore most of
these factors. Typically, hydrodynamic effects are represented by an added water mass moving
with the dam, implying water compressibility is neglected. Also ignored in these analyses is the
partial absorption of hydrodynamic pressure waves by the sediments invariably deposited at the
reservoir bottom and sides, or even by rock underlying the reservoir. The foundation rock is
usually assumed to be massless and a portion is included in a finite-element idealization of the
system. This extremely simple idealization of the foundation rock, in which only its flexibility is
considered but inertial and damping effects are ignored, is popular because the foundation
impedance matrix (or frequency-dependent stiffness matrix) is very difficult to determine for
unbounded domains.
 Although all of these factors known to be significant in the response of concrete dams
were included in computer software developed before desktop computer became standard by
graduate students at the University of California, Berkeley, they made no effort to develop input
and output interfaces that were user friendly. Thus, they remained primarily as research
programs with limited application to actual projects. Easier to use, the EAGD-84 program was
chosen as the analytical tool for the seismic safety evaluation of several concrete gravity dams,
but the less convenient EACD-3D-96 program was used only for a few arch dam projects.

However, starting in 1996, the U.S. Bureau of Reclamation (USBR) embarked upon a
major program to evaluate the seismic safety of dams and found it necessary to consider water
compressibility, reservoir boundary absorption, and dam-foundation rock interaction effects.
Among the 12 dams investigated were: (1) Deadwood Dam, a 50-m high single curvature dam;
(2) Monticello Dam, a 100-m high single curvature dam; (3) Morrow Point Dam, a 150-m high
double curvature dam; and (4) Hoover Dam, a 221-m high thick arch dam. The results of EACD-
3D analyses of these dams, conducted by L. Nuss and R. Munoz (USBR staff) are presented
next.

Implications of Neglecting Water Compressibility

We compare the earthquake-induced stresses in two of the four dams computed under
two conditions: water compressibility considered or neglected (Figs. 1 and 2). These figures
permit the following observations: the effects of water compressibility, which are generally
significant, vary with the location on the dam surface. By neglecting water compressibility, the
stresses may be significantly underestimated, as in the case of Monticello Dam (Fig. 1), or
significantly overestimated as in the case of Morrow Point Dam (Fig. 2). Thus, water
compressibility should be included in the analysis of arch dams; however, most standard finite
element analysis software neglects water compressibility.

Implications of Neglecting Foundation Rock Inertia and Damping

We compare the earthquake-induced stresses in each of the four dams, computed under
two conditions considering: (1) dam-foundation rock interaction; and (2) foundation rock
flexibility only (Figs. 3 through 6). For these two conditions, the largest arch stress on the
upstream or downstream face of the dam is 476 psi versus 844 psi for Deadwood Dam; 730 psi
versus 1410 psi for Monticello Dam; 665 psi versus 1336 psi for Morrow Point Dam; and 758
psi versus 2204 psi for Hoover Dam. If only foundation rock flexibility is considered (i.e.,
foundation rock is assumed to be massless), the stresses are overestimated by a factor of 2
(approximately) for the first three dams and by a factor of 3 (approximately) for Hoover Dam.

As demonstrated in the above examples, stresses are overestimated as a result of ignoring

the foundation rock material and radiation damping. Because such overestimation of stresses
may lead to overconservative designs of new dams and to the erroneous conclusion that an
existing dam is unsafe, and, hence, requires remediation, it is imperative that dam-foundation
rock interaction effects should be included in earthquake analysis of concrete dams. However,
dynamic analyses for most seismic safety evaluation projects assume massless foundation rock,
thus ignoring dam-foundation rock interaction effects, because most commercial and proprietary
computer software is based, erroneously, on these simplifying assumptions.

Advancement of Engineering Practice

Most of the research developments to include the effects of dam-water interaction, water
compressibility, reservoir boundary absorption, and dam-foundation rock interaction in analysis
of dams were reported over twenty years ago for gravity dams and over ten years ago for arch
dams. However, the profession has resisted using these analytical advances, in part because they
have not been incorporated in widely used commercial software.

The state of earthquake engineering practice for dams is not likely to advance unless
user-friendly software is developed for desktop computers that includes these interaction effects
in linear and nonlinear analysis of dams considering spatial variations in ground motion.

Part B: Buildings

Building Code Development

Most buildings are designed according to procedures specified in building codes, which
define the state of practice. In the United States, for example, the most recent code is the
International Building Code, first developed in 2000 and subsequently revised in 2003. The
structural analysis procedures in this code are based on the NEHRP Recommended Provisions for
Seismic Regulations for New Buildings and Other Structures (1997), which, in turn, are based on
the ATC-3-06 document (1978).

ATC-3-06 represented a major departure from traditional U.S. seismic codes. The new
features relevant to structural analysis included:

• Response modification factors (R factors) to reduce elastic response of the structure due
to realistic ground motions to obtain design forces; the R factors were dependent on
ductility, damping, and past performance of a given structural system.
• Strength-based design instead of allowable stress design.
• Provisions for dynamic (modal) analysis.
• Provisions to consider soil-structure interaction effects.

The ATC-3-06 document was the result of a massive effort during 1974 through 1976 by
a team of nearly 70 experts from around the country, and it was reviewed by various overview
panels and committees. Although all prominent engineers and academics were involved, its
acceptance into formal code documents was a slow process. Evolution of ATC-3-06 to the 1997
NEHRP provisions took twenty years. Until the analysis procedures in the latter were adopted
into the IBC (2000), they did not become a part of structural engineering practice in the United
States. In contrast, many of the ideas in ATC-3-06 were incorporated very quickly into the 1976
Mexico Federal District Code by Emilio Rosenblueth, who participated in the ATC-3-06 project.
Building Evaluation Guidelines

In contrast to the slow progress in building codes, the guidelines for seismic evaluation of
buildings evolved rapidly after the Loma Prieta (1989) and Northridge (1994) earthquakes in
California. Two major guideline documents appeared: ATC-40 (1996) and FEMA 273 (1997)
[and its successor FEMA 356 (2000)]. According to these guidelines, seismic demands are
computed by nonlinear static analysis of the structure subjected to monotonically increasing
lateral forces with an invariant heightwise distribution until reaching a roof displacement
determined from the deformation of an inelastic SDF system derived from the pushover curve.

According to the ATC-40 guidelines, the deformation of the inelastic SDF system is
estimated by the capacity-spectrum method, an iterative method requiring analysis of a sequence
of equivalent linear system; the method is typically implemented graphically. Unfortunately, the
ATC-40 iterative procedure does not always converge; when it does converge it does not lead to
the exact deformation (Chopra and Goel, 2000). Because convergence traditionally implies
accuracy, the user could be left with the impression that the calculated deformation is accurate,
but the ATC-40 estimate errs considerably. This is demonstrated in Fig. 7a, where the
deformation estimated by the ATC-40 method is compared with the value determined from
inelastic design spectrum theory and three different R y − µ − Tn equations. Both the approximate
and theoretical results are presented for systems covering a wide range of period values and
ductility factors subjected to ground motions characterized by an elastic design spectrum. The
discrepancy in the approximate result presented in Fig. 7b shows that the ATC-40 method
underestimates by 40-50% the deformation over a wide range of periods (Chopra and Goel,
2000).

Prompted by research results, such as those reported above, the ATC-55 (or FEMA-440)
project was launched in 2002 to develop improved procedures to estimate seismic demands. The
two flaws in the ATC-40 capacity spectrum method—lack of convergence in some cases and
large errors in many cases—appear to have been rectified in the FEMA-440 report (Comartin et
al., 2004). The optimal vibration period and damping ratio parameters for the equivalent linear
system are now derived by minimizing the differences between its response and that of the actual
inelastic system. Such an equivalent linear method would obviously give essentially the correct
deformation. However, the benefit in making the equivalent linearization detour is unclear when
the deformation of an inelastic system can be readily determined using available equations for
the inelastic deformation ratio [e.g., Ruiz-Garcia and Miranda, 2003; Chopra and
Chintanapakdee, 2004) or from the inelastic design spectrum (e.g., Chopra, 2001).

The nonlinear static procedure (NSP) in FEMA-356 requires development of a pushover

curve, a plot of base shear versus roof displacement, by nonlinear static analysis of the structure
subjected first to gravity loads, followed by monotonically increasing lateral forces with a
specified invariant height-wise distribution. At least two force distributions must be considered.
The first is to be selected from among the following: first mode distribution, equivalent lateral
force (ELF) distribution, and response spectrum analysis (RSA) distribution. The second
distribution is either the “uniform” distribution or an adaptive distribution; several options are
mentioned for the latter, which varies with change in deflected shape of the structure as it yields.
The other four force distributions mentioned above are defined as follows:
 1. First-mode distribution: s*j = m jφ j1, where m j is the mass and φ j1 is the mode shape
value at the jth floor.
2. ELF distribution: s*j = m j h kj , where h j is the height of the jth floor above the base, and k
varies between 1 and 2, depending on the fundamental vibration period.
3. RSA distribution: s* is defined by the lateral forces back-calculated from the story shears
determined by response spectrum analysis of the structure, assumed to be linearly elastic.
4. “Uniform” distribution: s*j = m j .

Each of these force distributions pushes the building in the same direction over the height of the
building (Fig. 8).

The limitations of the FEMA-356 force distributions are demonstrated in Figs. 9 and 10,
where the resulting estimates of the median story drift and plastic hinge rotation demands
imposed on the SAC−Los Angeles buildings by the ensemble of 20 SAC ground motions are
compared with the “exact” median value determined by nonlinear response history analysis
(RHA). The FEMA-356 lateral force distributions provide a good estimate of story drifts for the
3-story building. However, the first-mode force distribution grossly underestimates the story
drifts in the upper stories of the 9- and 20-story buildings, showing that higher-mode
contributions are especially significant in the seismic demands for upper stories. Although the
ELF and RSA force distributions are intended to account for higher-mode responses, they do not
provide satisfactory estimates of seismic demands for buildings that remain essentially elastic
(SAC Boston buildings) or buildings that are deformed far into the inelastic range (SAC Los
Angeles buildings) (Goel and Chopra, 2004). The “uniform” force distribution seems
unnecessary because it grossly underestimates drifts in upper stories and grossly overestimates
them in lower stories. Because FEMA-356 requires that seismic demands be estimated as the
larger of results from at least two lateral force distributions, it is useful to examine the upper
bound of results from the four force distributions considered. This upper bound also significantly
underestimates drifts in upper stories of taller buildings but overestimates them in lower stories
(Fig. 9). The FEMA-356 lateral force distributions provide a good estimate of plastic hinge
rotations for the 3-story building, but either fail to identify, or significantly underestimate, plastic
hinge rotations in beams at the upper floors of 9- and 20-story buildings (Fig. 10). Many
discussions of the potential and limitations pushover analysis are available in the literature (e.g.,
Naeim and Lobo, 1998; Krawinkler and Seneviratna, 1998; Elnashai, 2001; Fafar, 2002).

Despite the limitations of the ATC-40 procedure for estimating the target roof
displacement and the FEMA 273/356 nonlinear static procedure for estimating seismic demands,
these procedures have had profound influence on engineering practice not only in the United
States, where these procedures were developed, but in several other countries, and they continue
to be used. Eventually, they will be replaced by improved procedures that are emerging, but their
acceptance into codes and guidelines is likely to be a slow process. Because it is so difficult to
make major modifications to such documents, it is imperative that in the future any analytical
procedure should be thoroughly researched and tested before it is included.
Improved Nonlinear Static Procedures

It is clear from the preceding discussion that the seismic demand estimated by NSP using
the first-mode force distribution (or others in FEMA-356) should be improved. One approach to
reducing the discrepancy in this approximate procedure relative to nonlinear RHA is to include
the contributions of higher modes of vibration to seismic demands. It is well known that when
higher-mode responses are included in the RSA procedure, improved results are obtained for
linearly elastic systems.

Although modal analysis theory is strictly not valid for inelastic systems, the fact that
elastic modes are coupled only weakly in the response of inelastic systems (Chopra and Goel
2002, 2004) permitted development of the modal pushover analysis procedure (MPA). In this
MPA procedure, the peak “modal” demand rn (n denotes mode number) is determined by a
nonlinear static or pushover analysis using the modal force distribution s*n = mφ n (m is the mass
matrix and φ n is the nth vibration mode) up to a target displacement determined from the
deformation of the nth-“mode” inelastic SDF system; and the peak modal demands rn are then
combined by an appropriate modal combination rule.

Figures 11 and 12 show the median values of story drift and beam plastic rotation
demands, respectively, for SAC—Boston, Seattle, and Los Angeles buildings—including a
variable number of “modes” in MPA superimposed with the “exact” result from nonlinear RHA.
The first “mode” alone is inadequate in estimating story drifts, but with a few “modes” included,
story drifts estimated by MPA are generally similar to the upper floors of all buildings and also
in the lower floors of the Seattle 20-story building. Including higher-“mode” contributions also
improves significantly the estimate of plastic hinge rotations. In particular, plastic hinging in
upper stories is now identified, and the MPA estimate of plastic rotation is much closer—
compared to the first-“mode” result—to the “exact” results of nonlinear RHA.

Based on structural dynamics theory, the MPA procedure retains the conceptual
simplicity and computational attractiveness of the standard pushover procedures with invariant
lateral force distribution. Because higher-mode pushover analyses are similar to the first-mode
analysis, MPA is conceptually no more difficult than procedures now standard in structural
engineering practice. Because pushover analyses for the first two or three modal force
distributions are typically sufficient in MPA, it requires computational effort that is comparable
to the FEMA-356 procedure, which requires pushover analysis for at least two force
distributions. Without additional conceptual complexity or computational effort, MPA estimates
seismic demands much more accurately than FEMA-356 procedures (Goel and Chopra 2004).

Advancement of Engineering Practice

Building codes should explicitly state the underlying basis for each provision, so that it
can be improved as we develop a better understanding of structural dynamics and earthquake
performance of structures. Such a transparent format should be combined with a mechanism to
incorporate improvements more quickly than has been feasible in the past.
 Design codes and evaluation guidelines for buildings should be consistent with research
results and actual performance of structures during earthquakes. They should be tested
extensively before promulgation. The ATC-40 and FEMA-273 documents do not seem to have
satisfied this requirement.

Presently, developers of commercial software are followers, in the sense that they
implement analysis procedures specified in building design codes and evaluation guidelines. To
advance the state of engineering practice, they should become leaders in incorporating research
results—for example, improved nonlinear static procedures—into their software so that the
profession can use these procedures during the many years they are being considered by
committees for possible adoption in building evaluation guidelines.

Closing Comments

This paper has attempted to demonstrate that adoption of improved analytical techniques
for estimating seismic demands for structures in engineering practice is constrained by the lack
of computer software that implement these techniques and by prevailing codes of practice. The
structural types and analytical procedures chosen to illustrate this contention were obviously
based on my research experience. However, the overall issue is of broader concern and other
researchers may be able to identify supporting examples from their own experience.

References

Applied Technology Council (1978). Tentative provisions for the development of seismic
regulations for buildings, Report ATC-3-06, NBS Special Publication 510, NSF Special
Publication 78-08.

Applied Technology Council (1996). Seismic evaluation and retrofit of concrete buildings,
Report ATC-40, ATC, Redwood City, Calif.

Building Seismic Safety Council (1997). NEHRP guidelines for the seismic rehabilitation of
buildings, FEMA-273, Federal Emergency Management Agency, Washington, D.C.

Chakrabarti, P., and A. K. Chopra (1972). Earthquake Response of Gravity Dams Including
Reservoir Interaction Effects, Report No. EERC 72-6, Earthquake Engineering Research Center,
University of California, Berkeley, 161 pp.

Chopra, A. K. (2001). Dynamics of Structures: Theory and Applications to Earthquake

Engineering, 2nd ed., Prentice Hall, Upper Saddle River, N.J., 844 pp.

Chopra, A. K., and C. Chintanapakdee (2004). Inelastic deformation ratios for design and
evaluation of structures: single-degree-of-freedom bilinear systems, ASCE, J. Struct. Engrg., 130
(9), 1309−1319.
Chopra, A. K, and R. K. Goel (2000). Evaluation of NSP to estimate seismic deformation: SDF
systems, ASCE, J. Struc. Engrg., 26, 482−490.

Chopra, A. K., and R. K. Goel (2002). A modal pushover analysis procedure for estimating
seismic demands for buildings, Earthq. Engrg. Struc. Dyn., 31, 561−582.

Chopra, A. K., and R. K. Goel (2004). A modal pushover analysis procedure to estimate seismic
demands for unsymmetric-plan buildings, Earthq. Engrg. Struc. Dyn., 33 (8), 903−927.

Comartin, C. D., et al. (2004). A summary of FEMA-440: improvement of nonlinear static

seismic analysis procedures, Proceedings, 13th World Conf. Earthq. Engrg., Paper 1476,
Vancouver, British Columbia, Canada.

Fenves, G., and A. K. Chopra (a) (1984). Earthquake analysis and response of concrete gravity
dams, Report No. UCB/EERC-84/10, Earthquake Engineering Research Center, University of
California, Berkeley, 213 pp.

Fenves, G., and A. K. Chopra (b) (1984). EAGD-84: A computer program for earthquake
analysis of concrete gravity dams, Report No. UCB/EERC-84/11, Earthquake Engineering
Research Center, University of California, Berkeley, 78 pp.

Elnashai, A. S. (2001). Advanced inelastic static (pushover) analysis for earthquake applications,
Struc. Engrg. Mech., 12 (1), 51−69.

Fajfar, P. (2002). Structural analysis in earthquake engineering—a breakthrough of simplified

nonlinear methods, Proceedings, 12th Eur. Conf. Earthq. Engrg., London, Paper Ref. 843.

Fok, K.-L., and A. K. Chopra (1985). Earthquake analysis and response of concrete arch dams,
Report No. UCB/EERC-85/07, Earthquake Engineering Research Center, University of
California, Berkeley, 207 pp.

Goel, R. K., and A. K. Chopra (2004). Evaluation of modal and FEMA pushover analyses: SAC
buildings, Earthq. Spectra, 20, (1), 225−254.

Hall, J. F., and A. K. Chopra (1980). Dynamic response of embankment, concrete-gravity and
arch dams including hydrodynamic interaction, Report No. UCB/EERC-80/39, Earthquake
Engineering Research Center, University of California, Berkeley, 220 pp.

Krawinkler, H., and G. D. P. K. Seneviratna (1998). Pros and cons of a pushover analysis of
seismic performance evaluation, Engrg. Struc., 20 (4-6), 452−464.

Naeim, F., and R. M. Lobo (1998). Common pitfalls in pushover analysis, Proceedings, SEAOC
Annu. Conv., Reno, Nev.

Ruiz-Garcia, J., and E. Miranda (2003). Inelastic displacement ratios for evaluation of existing
structures, Earthq. Engrg, Struc. Dyn., 32 (8),1237−1250.
Tan, H. C., and A. K. Chopra (1995). Earthquake analysis and response of concrete arch dams,
Report No. UCB/EERC-95/07, Earthquake Engineering Research Center, University of
California, Berkeley,168 pp.

Tan, H. C., and A. K. Chopra (1996). EACD-3D-96: A computer program for three-dimensional
earthquake analysis of concrete dams, Report No. UCB/SEMM-96/06, University of California,
Berkeley, 131 pp

Zhang, L., and A. K. Chopra (1991). Computation of spatially-varying ground motion and
foundation-rock impedance matrices for seismic analysis of arch dams, Report No. UCB/EERC-
91/06, Earthquake Engineering Research Center, University of California, 121 pp.
 (a)

(b)

Figure 1. Peak values of earthquake-induced stresses in Monticello Dam computed under two
conditions: (a) water compressibility considered, and (b) water compressibility
neglected.
 (a)

(b)

Figure 2. Peak values of earthquake-induced stresses in Morrow Point Dam computed under
two conditions: (a) water compressibility considered, and (b) water compressibility
neglected.
 (a)

(b)

Figure 3. Peak values of earthquake-induced stresses in Deadwood Dam, considering (a) dam-
foundation interaction, and (b) foundation rock flexibility only.
 (a)

(b)

Figure 4. Peak values of earthquake-induced stresses in Monticello Dam, considering (a) dam-
foundation interaction, and (b) foundation rock flexibility only.
 (a)

(b)

Figure 5. Peak values of earthquake-induced stresses in Morrow Point Dam, considering (a)
dam-foundation interaction, and (b) foundation rock flexibility only.
 (a)

(b)

Figure 6. Peak values of earthquake-induced stresses in Hoover Dam, considering (a) dam-
foundation interaction, and (b) foundation rock flexibility only.
 (a) (b)
500 100
Acceleration Velocity Displacement Acceleration Velocity Displacement
200 Sensitive Sensitive Sensitive Sensitive Sensitive Sensitive
100
50 50 NH
KN

Discrepancy (%)
20 VFF
10 NH
D, cm

0
5 KN
VFF
2
ATC−40
1 −50
0.5

0.2 µ=4
µ=4
0.1 −100
0.05 0.1 0.2 0.5 1 2 5 10 20 50 0.05 0.1 0.2 0.5 1 2 5 10 20 50
T ,s T ,s
n n

Figure 7. Deformations computed by ATC-40 and from inelastic design spectrum using three
different R y − µ − Tn equations (identified by NH, KN, and VFF; part (a) compares
deformations and part (b) shows discrepancy in ATC-40 method. [Adapted from
Chopra and Goel, 2000.]

(a) (b) (c) (d)

0.195 0.281 0.367 0.119
0.167 0.21 0.177 0.11
0.149 0.165 0.0654 0.11
0.131 0.126 0.042 0.11
0.112 0.0913 0.0446 0.11
0.0923 0.062 0.0466 0.11
0.0714 0.0381 0.0702 0.11
0.051 0.0197 0.0981 0.11
0.0311 0.00719 0.0896 0.112

Figure 8. FEMA-356 force distributions for SAC-Los Angeles 9-story building: (a) first mode,
(b) ELF, (c) RSA, and (d) “uniform.
 (a) (b) (c)
3 9 20
Nonlinear RHA
FEMA
16
1st Mode
2 ELF 6
12
RSA
Floor

Uniform 8
1 3
4

G G G
0 1 2 3 4 5 6 0 2 4 6 8 0 1 2 3 4 5 6
Story drift ∆ or ∆ ,%
NL−RHA FEMA

Figure 9. Median story drifts for SAC-Los Angeles buildings: (a) 3-story; (b) 9-story; (c) 20-
story. Determined by nonlinear RHA and four FEMA-356 force distributions: first
Mode, ELF, RSA, and “uniform” (Goel and Chopra 2004).

(a) (b) (c)

3 9 20
Nonlinear RHA
FEMA
16
1st Mode
2 6 ELF
12
RSA
Floor

8 Uniform
1 3
4

G G G
0 0.02 0.04 0.06 0 0.02 0.04 0.06 0 0.02 0.04 0.06
Beam plastic rotation, rad

Figure 10. Median plastic rotations in interior beams of SAC−Los Angeles buildings: (a) 3-
story; (b) 9-story; (c) 20-story. Determined by nonlinear RHA and four FEMA-356
force distributions: first Mode, ELF, RSA, and “uniform” (Goel and Chopra 2004).
 Boston Seattle Los Angeles
9

6 NL−RHA
MPA
Floor 1 "Mode"
3 2 "Modes"
3 "Modes"

9−Story
G
0 0.5 1 1.5 2 0 1 2 3 4 5 0 1 2 3 4 5

20

16 NL−RHA
MPA
12 1 "Mode"
Floor

3 "Modes"
8 5 "Modes"

4
20−Story
G
0 0.5 1 1.5 2 0 1 2 3 4 5 0 1 2 3 4 5
Story drift, ∆MPA or ∆NL−RHA (%)

Figure 11. Median story drifts determined by nonlinear RHA and MPA with variable number of
“modes”; P-∆ effects due to gravity loads are included (Goel and Chopra 2004).

Seattle Los Angeles

9

6 NL−RHA
MPA
Floor

1 "Mode"
3 2 "Modes"
3 "Modes"

9−Story
G
0 0.02 0.04 0.06 0 0.02 0.04 0.06

20

16 NL−RHA
MPA
12 1 "Mode"
Floor

3 "Modes"
8 5 "Modes"

4
20−Story
G
0 0.02 0.04 0.06 0 0.02 0.04 0.06
Beam Plastic Rotation (rad) Beam Plastic Rotation (rad)

Figure 12. Median plastic rotations in interior beams determined by nonlinear RHA and MPA
with variable number of “modes”; P-∆ effects due to gravity loads are included (Goel
and Chopra 2004).