Seismic Design of Non-Conventional Bridges (2019)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

6 A literature review was conducted through the TRID database and Google searches, as well as through general industry contacts and personal references. The literature review produced a number of treatises on seismic analysis and scale testing for cable-stayed bridges. However, most published papers relating specifically to seismic design of non-conventional bridges focused on prediction of non-conventional (predominately cable-stayed) bridge response as opposed to design protocol for the bridge. The major seismic events in California in the late 20th century resulted in a significant amount of research into seismic demand and bridge behavior in seismic events. The seminal document in the literature that presented an evaluation of critical behavior and presented a framework for future design to address extreme seismic demands was the ATC-32 (3). The concept of capacity protection as presented in ATC-32 describes the use of “plastic hinge zones” and “full ductility structures” to provide a basis for directing damage to ductile elements of the bridge while maintaining selected elements as essentially elastic members. The commentary in ATC-32 provided guidelines for when to apply full ductility design, and when it should not be applied, stating “Fully ductile behavior assumes that the designer will take maximum advantage of plastic hinging while ensuring structural safety. This type of action implies considerable dam- age and is reserved for Ordinary Bridges only. Structural action consistent with limited ductility is recommended for Important Bridges and certain critical foundation components.” ATC-32 introduced behavioral criteria for graduated levels for performance-based design (Art 3.21.2.3) that should be applied to “Important Bridges,” which for the present discussion includes non- conventional bridges. These graduated performance levels (minimal damage, repairable damage, and significant damage) and the concept of multilevel hazard specification (ATC-32, Table 1) have remained the reference for design of non-conventional bridges in California and other high seismic regions since the publication of ATC-32 in 1996. These general performance criteria have evolved into more specific strain-based correlations to the qualitative performance levels origi- nally presented in ATC-32, which are now applied for most non-conventional bridges. The availability and capability of nonlinear analysis (4, 5) and the research into both sec- tion and material response behaviors in members subject to dynamic loads in the inelastic range (6, 7, 8) have also evolved since the advent of performance-based design concepts in the ATC-32 document. The ability to model explicit moment-curvature response of steel and rein- forced concrete members within a nonlinear analysis regime allows the structural elements in nonlinear models to predict the range of response for non-conventional bridge structures that address performance criteria at a more detailed level throughout the range of seismic demand, results that are not available to the designer using a bilinear stress-strain definition for rein- forcing steel and the limit-state plastic hinge assumptions on which the Guide Spec and the AASHTO BDS are based. C H A P T E R 2 Literature Review and Survey

Literature Review and Survey 7 Survey Results Fifty state bridge engineers were provided a survey to identify the standards of current practice for non-conventional bridge design in moderate and high seismic regions of the United States. The survey consisted of 13 questions. The question number is provided in the following sum- mary of data. Appendix B contains a copy of the state DOT survey. Question 1: State or Agency? Forty-three of the 50 states responded (86% of respondents). Question 2: Has your Agency directed the design of Seismic Design Category B, C, or D (Guide Spec) or Zone 2, 3, or 4 (AASHTO BDS) bridges in your jurisdiction? Twenty-three of the 43 states that responded to the survey answered “yes” to this question (53% of respondents). Question 3: Has your Agency directed the design of a non-conventional bridge (see definition) for Seismic Design Category B, C, or D (Guide Spec) or Zone 2, 3, or 4 (AASHTO BDS)? In the survey, the scope of “non-conventional bridge” was defined as any cable-supported bridge, long-span arch bridge, delta frame substructure, or any other structure where the owner has elected to adopt special design procedures for a bridge structure or substructure due to that structure not falling within the parameters applicable to the AASHTO BDS or Guide Spec. Twelve of the 43 states that responded to the survey answered “yes” to this question (28% of respondents). Figure 2 displays the findings from Questions 1 through 3. The survey was organized to allow respondents to “opt out” of the details, if they did not operate in a moderate to high seismic zone or had not designed non-conventional bridge struc- tures for seismic requirements. As a result, the size of the operative response pool was reduced to 10 respondents (see Figure 2). The detailed inquiry of experience in design of non-conventional bridges in moderate or high seismic zones included these 10 states: • California • Georgia • Illinois • Indiana • Missouri • New Jersey • New York • Oregon • South Carolina • Washington Questions 4 through 13 only pertain to the 10 states with the experience pertinent to the survey.

8 Seismic Design of Non-Conventional Bridges Question 4: What specification(s) serve as the basis or reference for your design criteria? Options to answer Question 4 were AASHTO LRFD Bridge Design Specification (AASHTO BDS), AASHTO Guide Specification for LRFD Seismic Bridge Design (Guide Spec), ATC-49 (9), agency or project specific only, or other (write-in). The responses to Question 4 are given in Table 1. Five out of 10 respondents use project- or agency-specific criteria, with one using agency criteria exclusively. Those using project-specific criteria are located in the higher seismic regions. Others utilize some combination of criteria founded on either the Guide Spec or the AASHTO BDS. There was only one response that the ATC documents (ATC-32 or ATC-49) were used as references for project-specific criteria (California). Since the ATC-32 and ATC-49 documents are references for the Guide Spec, the responses have been interpreted to reflect that no state is using either ATC standard directly as a criteria document. The question on criteria was not formulated No EQ or NC Exp: 18 NC Exp Only (Q3): 2 EQ Exp Only (Q2): 13 EQ&NC Exp (Q2&Q3): 10 No Response: 7 Figure 2. Composition of states with seismic (EQ), non-conventional (NC) experience, or both. State BDS Guide Spec ATC-49 Agency/Project Specific Other California YES YES YES YES Georgia YES Illinois YES Indiana YES YES Missouri YES New Jersey YES FHWA SEISMIC RETROFIT MANUAL New York YES YES YES Oregon YES YES South Carolina YES Washington YES YES Table 1. Specifications used by states for design of non-conventional bridges.

Literature Review and Survey 9 to isolate current practice to a single source, since most agency design specifications encompass multiple references. Respondents indicated that project-specific criteria were used in some com- bination with the Guide Spec for all non-conventional bridge projects, and of those, all indicated that their criteria included some reference to either the AASHTO BDS or the Guide Spec. The combination of these responses shows that five of 10 work with some form of project-specific criteria, while nine of 10 work with either the AASHTO BDS or the Guide Spec for criteria. Question 5: Do you reference any non-U.S. standards for seismic design criteria? No respondents indicated using any non-U.S. standards. Question 6: What bridge types are included in your non-conventional bridge resume? Options to answer Question 6 were cable-stayed, suspension, arch, extrados, cantilever truss or box girder (over 250-ft main span), or other (write-in). Table 2 documents the bridge types as reported in the survey. Question 7: Approximately how many non-conventional bridges in your jurisdiction have been designed for earthquakes in the last 30 years? Options to answer Question 7 were “less than 2,” “2 to 5,” or “more than 5.” Of the 10 respondents with experience pertinent to the survey, two answered less than 2 bridges, six answered 2 to 5 bridges, and California and New York answered more than 5 bridges (see Table 3). Question 8: Did you use nonlinear time-history analysis for design of any bridge type(s) in Question 6? Seven of the 10 respondents with experience pertinent to the survey responded that non- linear time-history analyses were used for the design of at least one of the bridge designs (see Table 3). State Cable-Stayed Suspension Arch Extrados Cant truss/box girder (250 ft) Other California YES YES YES YES Various* Georgia YES Illinois YES YES YES YES Indiana YES Missouri YES New Jersey YES Moveable Bridges New York YES YES Oregon YES YES YES YES South Carolina YES Washington YES YES YES Tunnel *California listed double deck concrete viaducts, self-anchored suspension, long concrete viaducts, long-span plate girder, and delta girder bridges in the “Other” category. Table 2. Bridge types included in non-conventional bridge resume.

10 Seismic Design of Non-Conventional Bridges Question 9: Did seismic demand control the lateral force or displacement-based design for any of the bridge(s) in Question 7? Eight of the 10 respondents said seismic demands controlled the design. One state answered “no.” For New York, no answer was recorded (see Table 3). Question 10: Have you specified multilevel seismic events (e.g., a 250-year return for operation and 2,500-year return for no-collapse or sim) for your non-conventional bridge design(s)? Six of the 10 states replied that they used multilevel seismic events (see Table 3). Question 11: The reason for specifying a multilevel event was? Options to answers to this question were lifeline status of the facility, financial risk, life safety risk, or other. Three of the 6 respondents using multilevel events did so for the reasons of lifeline status, financial risk, and life safety risk. Two of the respondents were mainly concerned with lifeline risk (see Table 3). Question 12: What are your primary sources for establishing performance criteria? Options to answers this question were AASHTO BDS, Guide Spec, Research, ATC, and Other. Two of the 10 states used both the AASHTO BDS and the Guide Spec as their primary sources for establishing performance criteria. Four other states used the AASHTO BDS or the Guide Spec. California used Caltrans-sponsored research, ATC-32 for projects circa 1996, and listed internal subject matter expertise, external seismic advisory board, project peer review panels, and project technical advisory panels in the “Other” category. Table 4 lists the answers received to this question by state. Question 13: Please indicate whether or not you can provide a plan and elevation drawing for specific structures where you have applied non-conventional bridge seismic design criteria, along with the criteria document that relates to the seismic design basis for the bridge substructure and foundation system(s)? This question was used to track information received from the states. State # Bridges NL TH Analyses EQ Control Multilevel Multilevel Reason California More than 5 Yes Yes Yes Lifeline, Financial, Life Safety Georgia 2 to 5 Yes No No Illinois 2 to 5 No Yes No Indiana Less than 2 No Yes Yes Lifeline, Financial, Life Safety Missouri 2 to 5 Yes Yes No New Jersey 2 to 5 No Yes No New York More than 5 Yes Yes Lifeline, Financial, Life Safety Oregon 2 to 5 Yes Yes Yes Lifeline, Life Safety South Carolina Less than 2 Yes Yes Yes Lifeline Washington 2 to 5 Yes Yes Yes Lifeline Table 3. Summary of responses to Questions 7 through 11.

Literature Review and Survey 11 Survey Summary The survey results show that virtually all non-conventional bridges in high seismic regions of the United States are designed using nonlinear time-history analysis and some degree of performance-based design. The state DOT survey is presented in Appendix B. * California listed ATC-32 for projects circa 1996. **California listed internal subject matter expertise, external seismic advisory board, project peer review panels, and project technical advisory panels in the “Other” category. State AASHTO LRFD BDS AASHTO Guide Spec Research ATC Other California Caltrans Sponsored YES* Various** Georgia YES Illinois YES Indiana YES YES Missouri New Jersey YES New York YES YES Oregon South Carolina In 2000 a compendium of sources Washington YES Project specific Table 4. Primary sources for establishing criteria.