FINAL REPORT

U.S. ARMY CORPS OF ENGINEERS

EVALUATION & ASSESSMENT
OF PROPOSED ALTERNATIVES TO
RETROFIT/REPLACE THE EAST SPAN OF THE
SAN FRANCISCO-OAKLAND BAY BRIDGE

October 27, 2000


TABLE OF CONTENTS

EXECUTIVE SUMMARY 3
Introduction 3
Scope of Work 3
Conclusions 3

PROJECT OVERVIEW 5
Background 5
Time Line 5

SCOPE OF WORK 7

DATA ASSESSMENT 8
Significant Data Gaps 9
Moderate Data Gaps 10
Minor Data Gaps 11

KEY QUESTIONS 12
Answers to Questions 13
Question 1 13
Question 1a 14
Question 1b 16
Question 1c 17
Question 2 17
Question 3 18
Question 4 19
Recommendations 23
Supplemental Questions 24

List Acronyms and Abbreviations 25
Proprietary Computer Software Referenced 26

Appendix 1. Chronological Table
Appendix 2. Scope of Work
Appendix 3. Data Catalog
Appendix 4. Retrofit Support Document
Appendix 5. Originally Proposed Replacement Support Document
Appendix 6. Currently Proposed Replacement Support Document
Appendix 7. Review of Cost Estimates and Economic Analyses
Appendix 8. Lifeline Criteria
Appendix 9. Supplemental Questions


EXECUTIVE SUMMARY

Introduction

In response to the 1989 Loma Prieta earthquake, the State of California Department of Transportation (Caltrans) began a program to seismically retrofit all bridges in the state, including the damaged East Span of the San Francisco-Oakland Bay Bridge (SFOBB). As the plan to retrofit the East Span progressed Caltrans decided that it would be more cost effective to replace the structure rather than to retrofit it. Caltrans considered several designs for a replacement span and in 1997 selected a "skyway" design as the best alternative. Subsequently, the Metropolitan Transportation Commission (MTC), representing nine Bay Area counties and acting under authority granted by the California legislature, decided to pay the cost of adding a signature span and "amenities" to the bridge. These amenities included a bicycle / pedestrian path and the signature span, which is a self-anchored suspension (SAS) design that the MTC decided was more distinctive than the skyway design.

Scope of Work

The City and County of San Francisco (the City) and Caltrans have asked the U.S. Army Corps of Engineers (COE) to evaluate key technical decisions made by Caltrans in reaching the conclusion to build a replacement bridge. Specifically, the purpose of the COE's assessment is to examine two broad areas of concern as raised by the City. First, the City believes that, from the standpoint of cost and public safety, it is preferable to retrofit the East Span than to replace it with the currently proposed design. Second, the City believes that the self-anchored suspension design that Caltrans is currently proposing for the replacement span is not seismically safe. The scope of work also includes four key questions regarding retrofit / replacement design, cost and seismic safety that require answers from the COE Team.

The COE Team’s conclusions and responses are based solely on data submitted and documented in the Data Catalog. The COE Team performed no new analyses.

Conclusions

The COE Team can summarize its review and address the aim of this study by stating the following:

Documents provided did not demonstrate that any retrofit alternative met lifeline criteria.

Caltrans' proposed retrofit strategy is not reasonable due to concerns regarding the isolation strategy, incompleteness of design, and definition of performance criteria.

Based on safety considerations, it is the COE Team’s opinion that, at this point in time, a replacement alternative is preferable to a retrofit alternative. A replacement alternative is the path that most quickly resolves the exposure of the public to the seismic vulnerabilities of the existing structure.

Costs for tThe currently proposed replacement alternative bridge areis $565 million higher than for the proposed retrofit. Reliability studies, for comparison, have not been found for either the retrofit or the replacement bridge.

Seismic safety is being addressed as Caltrans’ design team works towards meeting the seismic performance criteria established by design authorities including the Seismic Advisory Board (SAB) and the Engineering and Design Advisory Panel (EDAP).

The replacement bridge does not meet lifeline criteria as defined in the Scope of Work, but is being designed to conform to a unique Design Criteria, including the Safety Evaluation Earthquake (SEE) performance criteria. The design work is not yet complete and and conformance to the SEE criteria cannot be verified. It is the COE Team’s opinion that Caltrans’ design team is highly qualified, using state-of-the-art design methods and is moving along a path to design a bridge that meets the seismic performance criteria.

The performance of the replacement bridge during a Maximum Credible Earthquake (MCE) cannot be determined. The bridge has not been evaluated or designed for a MCE event, which is larger than the SEE event.

Passenger vehicle access and accommodation has been generally addressed in the Design Criteria, that requiring es “Full service almost immediately.” following an earthquake. The Design Criteria does not define “immediate” and the design calculations do not demonstrate how this design requirement is met.


PROJECT OVERVIEW

Background

In response to the 1989 Loma Prieta earthquake, the State of California Department of Transportation (Caltrans) began a program to seismically retrofit all bridges in the state, including the damaged East Span of the San Francisco-Oakland Bay Bridge (SFOBB). As the plan to retrofit the East Span progressed, it became apparent to Caltrans that it would be more cost effective to replace the structure rather than to retrofit it. Caltrans considered several designs for a replacement span and in 1997 selected a "skyway" design as the best alternative. Based on the costs associated with the skyway design, Caltrans formally decided to replace rather than retrofit the east span. Subsequently, the Metropolitan Transportation Commission (MTC), representing nine Bay Area counties and acting under authority granted by the California legislature, decided to pay the cost of adding "amenities" to the replacement span. These amenities included a self-anchored suspension bridge design, which the MTC decided was more distinctive than the skyway design, and a bicycle/pedestrian path.

The City and County of San Francisco (the City) and Caltrans have asked the U.S. Army Corps of Engineers (COE), as a body of independent experts, to evaluate key technical decisions made by Caltrans. Specifically, the purpose of the COE's assessment is to examine two broad areas of concern as raised by the City and its outside consultants. First, the City believes that, from the standpoint of cost and public safety, it is preferable to retrofit the East Span than to replace it with the currently proposed design. Second, the City believes that the self-anchored suspension design that Caltrans is currently proposing for the replacement span is not seismically safe. To expedite the COE's study of these two concerns, the Federal Highway Administration (FHWA), in cooperation with the U.S. Navy, facilitated the COE's communication with appropriate Federal, State, local agencies and their consultants. The U.S. Coast Guard participated as well.

Time Line

As background information and to facilitate an understanding of the decisions made, a summary time line is presented in Figure 1 on the following page. These events are further described in the Chronological Table provided in Appendix 1.

[INSERT FIGURE 1.
TIMELINE]

SCOPE OF WORK

Per the scope of work (Appendix 2), the COE Team conducted its evaluation during the two-phase study.

Phase 1, completed on July 25, 2000, included acquiring and cataloging (see Appendix 3 for the updated and current Data Catalog) all reports, data and analyses provided to the COE Team that address the City’s two broad areas of concern. The COE Team assessed the completeness and quality of that information and whether sufficient data was available to answer the four major questions in the scope of work. Also as part of the data assessment in Phase 1, the COE Team visited the East Span of the San Francisco-Oakland Bay Bridge. The visit included the Oakland Mole, Yerba Buena Island, the cantilever section, the failure span at E9, and the pile cap at E3. In addition, the COE Team viewed the bridge by boat and reviewed half scale test specimens for lattice members from a completed test for the Golden Gate Bridge.

The results of Phase 1 are contained in the U.S. Army Corps of Engineers Interim Letter Report, Evaluation & Assessment of Proposed Alternatives to Retrofit/Replace the East Span of the San Francisco-Oakland Bay Bridge, dated July 25, 2000.

Phase 2 answers the four major questions contained in the scope of work and presents the COE Team’s findings in two letter reports: 1) Interim Final Report USACE Evaluation & Assessment of Proposed Alternatives to Retrofit/Replace the East Span of the San Francisco-Oakland Bay Bridge, completed September 22, 2000 and 2) Final Report USACE Evaluation & Assessment of Proposed Alternatives to Retrofit/Replace the East Span of the San Francisco-Oakland Bay Bridge.

This report is the Final Letter Report and completes the COE Team’s work by addressing Questions 3 and 4 of the scope of work.

DATA ASSESSMENT

The Data Catalog provided in Appendix 3 is current with over 400 documents, some of which contain multiple volumes, and represent over 75,000 pages of material. Most of the documents are loose-leaf three ring binders and cover such areas as:

Historical documents of the as-built structure including plans, news articles, and design and construction articles by the designers.
Test reports covering the performance of steel elements of the existing bridge towers and superstructure.
As-built analysis and retrofit design calculations.
Cost estimates for the retrofit contracts.
Value engineering studies.
Comparisons of the retrofit alternative to the replacement alternative ranging from Caltrans internal memos to the Governors Action Request (GAR) report.
Plans and specifications covering the design of the new replacement alternative.
Project Engineer / Designer notes.
Seismic Safety Peer Review Panel (SSPRP) meeting minutes.
University Research.

The data provided by Caltrans is voluminous. To locate information and clarify document content, several meetings have been held with Caltrans. These meetings have aided the team’s effort at locating and understanding the data and the data gaps. The data provided by the City consists mainly of letters and reports authored by Professor Astaneh. The City has submitted neither design nor cost information. Follow up meetings have also been held with Professor Astaneh, who represents the City, to allow further clarification of the City’s concerns. The COE Team has spent over 600 man-hours in these meetings with Caltrans and City representatives.

The information in the Data Catalog includes all reports, data, and analyses that have been provided by the City and Caltrans over the course of this study. This information represents the basis for the answers given by the COE Team in response to the questions in the scope of work.

Data gaps were initially identified in the report titled Phase 1 Interim Letter Report, Evaluation & Assessment of Proposed Alternatives to Retrofit/Replace the East Span of the San Francisco-Oakland Bay Bridge, dated July 25, 2000. These Data Gaps have been modified and/or lined out to account for additional data and information that has been submitted since the completion of Phase 1. The revised Data Gap listing is as follows:

Significant Data Gaps

Design criteria summary for the proposed retrofit alternative. Quantify these criteria in terms of allowable stress and strain levels, displacement limits, and other pertinent parameters. Significant to Questions 1, 2,
and 3.

Concise definition of the acceptable level of structural response quantities should be known for seismic performance evaluation of the as-built and retrofitted bridge. Design criteria summary that outlines the pertinent parameters including acceptable stress and strain levels, displacement limits, and other factors that are essential for assessment of the bridge’s seismic performance, was not provided in a single document.
Documents outlining the decision process for data supporting costs in the Governor’s Action Request (GAR). Significant to Questions 1 and 2.

Caltrans’ recommendation for bridge replacement is summarized in the GAR. The GAR provides cost figures from various sources (i.e., Value Analysis Study, Replacement Study For The East Span of SFOBB Seismic Safety Project, and Retrofit vs. New Bridge Economic Analysis study). There is no outline of the criteria used by Caltrans to support their selection of costs that were used in the GAR.

This information is needed to substantiate Caltrans’ cost effectiveness position that a replacement bridge is more cost effective than retrofitting the existing bridge.

Basic Geotechnical and Geology Data. Significant to Questions 1, 2,
and 3.
The information provided for the Oakland Mole contained in Document 319, and in the other documents referred to in Document 319 is comprehensive and appears to be sufficiently complete. However, a similarly comprehensive presentation of geotechnical and geology data for the rest of the bridge alignment is important for foundation analysis of both the retrofit and the new bridge alternatives.
Several key geotechnical design issues have been identified in the various documents. The information received to date does not include a comprehensive presentation of the resolution to the following issues:
Soil/pile interaction loads, particularly for battered piles.

Soil structure interaction model incorporated into the global model of the various sections of the bridge.

Geotechnical information (boring logs, CPT, field tests, laboratory tests, etc.).

Seismology and Selection of Ground Motions. Significant to Questions 1, 2, and 3.
For the retrofit alternative, a comprehensive report on seismology and ground motions has not been made available to us. Only limited ground motion information in various design reports has been provided. However, this information does not provide an overall view of the methodology used to develop multi-support ground motions for the soil structure interaction and the structural analysis of the bridge.
Analysis and design calculation documents for portions of the bridge associated with Contracts 4 and 5 (Foundations E6 – E16) on proposed retrofit. Significant to Questions 1, 2, and 3.
Without this information it may not be possible to determine if sound analysis and appropriate criteria were used for the subject portions of the bridge.
Work in progress / most current cost data on currently proposed replacement 65 percent design review. Significant to Question 3.
Currently Caltrans has submitted for review a 65 percent design along with a 35 percent cost estimate. To ensure accurate, realistic, and complete cost evaluation a 65 percent cost estimate is required. Significant to Question 3.
Moderate Data Gaps
Stage of design to which work each contract had progressed when the decision was made to go to replacement. Significant to Question 1.
With many documents at various stages of design it is difficult to identify which documents are pertinent to the most current design. Without identifying the chronology of events, the decision making process is not clear. The level of design stage of each contract (i.e., conceptual, preliminary, final designs) should be known to accurately evaluate the retrofit alternative. This information is only provided by Caltrans for contract 8, in Document 326.
Meeting minutes, notes and/or letters of meetings for the following groups:
Caltrans Seismic Advisory Board (SAB) from 1990 to present.
Caltrans Seismic Safety Peer Review Panel (SSPRP) from 1997 to 1998 and March 2000 to present.

Caltrans retrofit strategy meetings from 1990 to present, including design engineer’s preparation for the meetings.
Significant to Questions 1, 2, 3, and 4.
This information is necessary in providing an outline of the review process and identifying changes in project direction recommended by these advisory groups. Significant to Questions 1, 2, 3, and 4.
Analysis strategy using various computer models, including the relationship between the various global and local models for the retrofit alternative. Significant to Question 1.
This information is necessary to show the relationship of local and global models. Although specific sections of the bridge can be analyzed separately, ultimately the bridge must function as a whole. With the information provided it is difficult to determine whether or not the results of the local and global models are compatible. In addition, the information does not show consistency between bridge components.
Minor Data Gaps
Material test reports and/or summaries for the condition of the existing foundations, including concrete, steel, and timber. Significant to Questions 1, 2, and 3.
Lack of this information limits the ability to assess studies of existing foundation information and proposed retrofit designs. Evaluation of soil structure interaction depends on an understanding of existing material properties.

KEY QUESTIONS

In preparing a response to the scope of work’s key questions, the COE Team’s approach is to state the question and present a summary conclusion. Detailed analysis for each question are referenced and contained in the appendixes.
In responding to the key questions, the COE Team has based its conclusions on the data submitted and documented in the Data Catalog. During the course of the study, discussions were held with Caltrans and City representatives to help the team gain a better understanding of the project and to assist in locating relevant information within the documents provided. Only written, verified documents have been used in development of the conclusions. The COE Team performed no new analyses. Where data can be corroborated and supported by a document in the Data Catalog the document’s number is referenced in brackets.
This study combines a short time schedule with the daunting task of reviewing a massive amount of documentation on both the retrofit and replacement projects (studies, university research, plans, specifications, etc.). The documents span nearly a decade since the Loma Prieta earthquake. It has been a challenge for the COE Team to separate and review all the pertinent project data. The goal has been to piece together the relevant data needed to give unqualified answers to the key questions. Both Caltrans and the City have gone to great lengths to provide the needed data to the COE Team. However, for the data provided, the level of completeness is only sufficient for the COE Team to give qualified answers.
The questions are in and of themselves complex and difficult to answer in a straightforward manner. The COE Team provided the most complete answer possible, using the information provided. The answers are based on a holistic (global) perspective, encompassing the total length of the bridge.

The following section makes reference to stages of project planning and design using acronyms; the following provides a key to these design stages as defined by Caltrans:

Advanced Planning (AP) 0 – 35% Complete
General Plans (GP) 35 – 75% Complete
Plans, Specifications, & Estimates (PS&E) 75 – 95% Complete



Answers to Questions

Question 1: Was Caltrans' selection of the proposed retrofit alternative reasonable -- i.e., was it based on appropriate criteria and sound analysis, including consideration of realistic, accurate and complete cost figures?

Conclusion: Caltrans had separated the retrofit design into eleven design projects. These designs were in various stages of completion from AP to PS&E. It is the COE Team’s conclusion that Caltrans initially used a structured approach to evaluate alternate retrofit strategies, but as explained below, the selected retrofit strategy does not appear to be reasonable due to concerns regarding the isolation strategy, incompleteness of design, and definition of performance criteria.

Isolation Strategy

Caltrans' proposed retrofit alternative is seismic isolation of truss systems with the exception of the cantilever section. The suspended portion of the cantilever section is cut off and isolated using two new pier supports, while the rest of the cantilever section remains fixed to its piers but is strengthened by edge trusses. Ordinarily, an isolation system is considered for relatively rigid structures to elongate their period of vibration in order to reduce seismic force demands and to provide additional damping through friction or other means. Most spans of SFOBB are long-period structures with fundamental periods of vibration in the range of several seconds.

The seismic force demands for such long-period spans in their existing conditions are approximately at the same level of the proposed isolated spans. On this basis, the use of an isolation system appears unreasonable. Documents submitted for review do not demonstrate why a flexible structure with low seismic force demands should be stiffened by concrete encasement and then softened back to its original condition using isolation bearings. Computer analyses of the isolated bridge are based on unrealistic modeling and input assumptions and they provide limited results. The validity and effectiveness of the isolation retrofit strategy has not been demonstrated.

The following statement, from the Seismic Advisory Board’s meeting minutes (January 3 and 4, 1991), supports the concerns stated above: “Because of the sensitivity of base-isolated structures to the longer periods of free-field ground motion, base isolation should be avoided at very soft sites such as those on San Francisco Bay fill” [Document 372]. Two letters to Director James van Loben Sels (December 4 and 5, 1995) also document the concern regarding unprecedented use of an isolation system.

Incomplete Design

The proposed retrofit strategy design of the entire bridge is incomplete. None of the 11 design projects that comprise the retrofit have a finalized, verified retrofit solution, particularly the cantilever truss spans and their foundations. Retrofit designs for the cantilever portion of the bridge including the superstructure and foundations, have not been completed and only preliminary concepts have been derived. No analyses have been provided to demonstrate that they are reasonable and workable.

Criteria

A general statement for seismic design criteria has not been defined. Criteria is inconsistently applied and continually modified. Criteria appear to change as the efforts on the cantilever portion of the bridge progressed.

As can be expected for a bridge of this complexity, Caltrans appeared to struggle with design and cost issues to meet lifeline criteria. Further, they did not have any reasonable degree of confidence that a retrofit alternative could be designed to meet lifeline performance criteria. At the time that the decision was made to proceed with the replacement alternative, Caltrans’ documents indicate that the retrofit design did not meet lifeline criteria and was being designed to meet the lesser criterion of no-collapse.

COE Team’s detailed review of the data supporting this question can be found in Appendix 4, Retrofit Support Document.

Question 1a: Did Caltrans adequately consider/evaluate other retrofit alternatives, including a West Span-type retrofit and other steel retrofits, and did this evaluation include consideration of realistic, accurate and complete cost figures?

For informational Purposes: The West Span retrofit scheme is to directly strengthen the steel tower members with additional steel components (as verbally provided by FHWA). The West Span refers to the suspension bridge west of Yerba Buena Island.

Conclusion: The COE Team has found that Caltrans considered numerous other retrofit alternatives as reflected in Table 1. The alternatives considered apply to all aspects of the retrofit including foundations, towers and superstructure. The alternatives were not usually evaluated to a level of being able to produce realistic, accurate and complete cost figures. However, to make prudent decisions for retrofit, this is not always necessary. Even though the COE Team questions the reasonableness of Caltrans' selected retrofit alternative, it does not disagree with the decision process that led to that selection.

The pursuit of a valid retrofit scheme should not be compared to the preliminary design stages of a new bridge structure as this question suggests. Choosing several global schemes for the retrofit and taking them to a 30 percent design level in order to find the “best” solution is not the normal process for retrofit. This is more practical for a new structure because the type is not restricted by existing conditions. Retrofit alternatives, however, are limited by the existing bridge.

An accepted process in developing a valid retrofit scheme is to consider (brainstorm) possible options and, based on discussions of technical feasibility, aesthetics, and preliminary costs, bring forward the most promising overall scheme or strategy. Caltrans conducted an initial analysis to identify the seismically vulnerable items for the existing “as-built” bridge. Given these items, various retrofit schemes for all the components of the bridge were brainstormed and discussed. Schemes considered are shown in Table 1.

Two general retrofit strategies typically evaluated are: 1) to strengthen elements; or 2) to divert (reduce) forces away from elements that lack capacity for the design load. Forces can be diverted by adding members or by using seismic isolation. These two retrofit strategies are generally considered on two levels: 1) global (entire bridge, or at least by superstructure, frame-by-frame, type), or 2) local (individual components or elements). The selected retrofit includes both of these strategies.

On the local element-by-element level, valid “alternatives” should have been thoroughly evaluated. The COE review team could not verify in the documents provided that an adequate evaluation took place at this level to support several key decisions regarding the strategy path taken. The critical step of abandoning a more typical “strengthening” scheme for the selected retrofit scheme (using seismic isolation bearings) was not adequately documented.

Considering the West Span type retrofit, Caltrans had evaluated this retrofit in a comparison with the selected concrete-encasement type. The evaluation was done by the Contract 2 design team for the towers on Yerba Buena Island. The foundations for these towers are supported by rock. Therefore, the effect of the additional concrete dead load is not as detrimental to the tower foundation as it is on the foundations for the caisson and pile supported piers. This comparison, which led to the conclusion that the concrete encasement was the better solution (steel versus concrete), considered only the cost of the tower retrofit and not the impact to the pile and caisson foundations.

As stated in the conclusion for Question 1, a final valid scheme for the selected retrofit alternative had not been achieved. However, the decision process that Caltrans had followed for this project was adequate. After consideration of various alternatives, what Caltrans considered to be the most promising retrofit scheme, was brought forward into the analysis and design phase. This approach makes use of the vast experience and knowledge available at Caltrans, by quickly considering and eliminating less plausible solutions, and saving the expense of investigating non-viable alternatives.

COE Team’s detailed review of the data supporting this question can be found in Appendix 4, Retrofit Support Document.

Question 1b: Did Caltrans adequately consider/evaluate the ability of other retrofit alternatives, including a West span-type retrofit and other steel retrofit, to meet lifeline criteria? Which (if any) retrofit alternatives meet lifeline criteria?

Conclusion: The documents provided did not demonstrate that any retrofit alternative met lifeline criteria. Consequently, Caltrans did not evaluate in detail the ability of other retrofit alternatives to meet lifeline criteria.

COE Team’s detailed review of the data supporting this question can be found in Appendix 4, Retrofit Support Document.

Question 1c: Did Caltrans adequately consider/evaluate the costs of retrofitting the span to meet lifeline criteria?

Conclusion: The data reviewed clearly shows that Caltrans did not have a reliable retrofit solution. Therefore, a retrofit solution that could be classified as meeting lifeline performance criteria did not exist. The cost data reviewed by the COE Team were found to be adequate and supportable to the level of design completed. In this case, that level, as stated by Caltrans, was to no-collapse and not lifeline conditions [Document 267]. Analysis to substantiate either performance level is not evident. The COE Team found that Caltrans used sound judgement and estimating procedures, including the use of appropriate cost items, which were consistent and accurate to the level of design under consideration. A cost was not specifically developed for an alternative that would meet lifeline criteria.

COE Team’s detailed review of the data supporting this question can be found in Appendix 4, Retrofit Support Document.

Question 2: Was Caltrans' cost-benefit analysis comparing the originally proposed replacement alternative vs. the proposed retrofit alternative reasonable -- i.e., was it based on appropriate criteria and sound analysis, including consideration of realistic, accurate and complete cost figures?

Conclusion: The COE Team found that the procedures used by Caltrans to form the cost-benefit analyses were reasonable, and Caltrans used sound judgment and estimating procedures, including the use of appropriate cost items. The items are consistent and accurate for the level of design under consideration.

Caltrans’ cost figures for the retrofit strategy include the appropriate elements needed to produce a reasonable budgetary tool commensurate with the level of design. The cost presented represent a broad range of numbers and values that were based on engineering and cost assumptions. The lifecycle costs used in the economic analyses could not be substantiated by the data submitted and reviewed, but did represent a reasonable range of costs for this type of analysis.

Caltrans’ cost figures for the originally proposed replacement alternative are based on appropriate criteria and sound analysis. Support documentation is provided in Appendixes 5 and 7.

Cost items considered in the cost-benefit or lifecycle analysis include traffic delays, hazardous work areas, lane closures, work conducted in traffic, lead base paint abatement, worker and public safety, and maintenance as well as costs associated with working with steel and concrete construction over water.

Document 250 is the primary lifecycle/economic analysis report while Documents 23 and 249 also address the lifecycle costs of the retrofit alternative and the originally proposed replacement alternative. Document 250 is supported by Documents 23 and 249. These documents make the same conclusion, i.e., that the replacement approach is preferable based on lifecycle costs.

Even though backup data is limited, the economic or lifecycle analyses sufficiently addresses the significant issues and costs (limited data includes cost items and probabilistic methods to estimate seismic damage, etc.). The lifecycle analyses are reasonable. The lifecycle costs as presented by Caltrans indicate that the decision to select a replacement alternative would be justified given a retrofit with substantially less construction costs.

Question 3: How does the currently proposed replacement alternative, including as well any work in progress, compare to various retrofit alternatives in terms of a) cost and b) seismic reliability (including ability to meet lifeline criteria)?

Part a) Cost Conclusion: Table 2 summarizes and compares the construction and design cost estimates for the proposed retrofit alternative, the originally proposed replacement alternative (Skyway), and the currently proposed replacement alternative (SAS). For comparison purposes, cost estimates for each alternative have been adjusted with a 3 percent yearly escalation factor to bring costs to a year 2000 level.

Table 2 Cost Comparisons in Millions ($) See Notes in Margin

ITEM Proposed Retrofit
1996
(Document 253) Originally Proposed Skyway 30%
1998
(Document 263) Currently Proposed
SAS 65% w/ Amenities
2000
(Document 370) Mainspan NA 149.10 390.79 Skyway NA 526.60 565.59 YB Trans NA 50.50 88.58 OTD NA 29.00 101.43 YB Detours NA 49.00 46.45 Demo NA 54.10 54.10 Struc. Total 733.50 858.30 1,246.94 Roadway Cost 62.00 89.50 83.56 Support Cost 126.70 155.40 244.25 TOTAL 922.20 1,103.20 1,574.75 Escalated to 2000 $s 1,038 1,170 1,574.75 The cost of the SAS alternative is approximately $565,000,000 higher than the cost of the proposed retrofit. The cost of the SAS alternative is approximately $405,000,000 higher than the cost of the Skyway alternative. This increased cost is due primarily to the addition of the signature span and amenities – bikeway/pedestrian path and lighting. Support costs have also increased significantly.

Part b) Seismic Reliability Conclusion: In reviewing the available documents, reliability studies have not been found for the retrofit alternative(s) or the currently proposed replacement alternative. For the currently proposed replacement, there are discussions that relate to reliability. The discussions can be found in the Ventry Value Analysis Reports [Documents 169 & and 170]. However, these discussions do not specifically address the currently proposed replacement bridge, nor do they present quantitative reliability reports.

In keeping with the Scope of Work, the COE tTeam is not producing any new data or analyses, and cannot answer this question directly without performing a reliability analysis.

Question 4: Is the currently proposed replacement alternative seismically safe? How will the currently proposed replacement alternative perform in a maximum credible earthquake? Specifically, does the currently proposed replacement alternative meet lifeline criteria? To what extent and how quickly could it accommodate passenger vehicles?

This question is answered in 4 parts:

Part 1. Is the currently proposed replacement alternative seismically safe?

Conclusion: It is the COE Team’s opinion that Caltrans’ design team is moving along a path to design a bridge that meets the seismic performance criteria established by the SAB and EDAP. The COE Team’s response is based on the following.

Seismic safety depends on the actual performance of the proposed replacement bridge during a seismic event. The expected performance is determined by the predictability of the engineering criteria that is used for design and construction. The criteria are agreed upon by authorities in various fields and are updated as new events provide additional information and experience. Such criteria become the applicable standard of practice. The degree to which a design conforms to these criteria determines if the bridge will perform in an expected manner based on past experience.

The proposed replacement bridge is a unique structure and a unique set of seismic performance criteria have been developed to guide design and construction. The criteria have been developed by Caltrans based on industry guidelines (AASHTO, BDS, ATC, AISC, etc.) and input from industry experts (i.e., Seismic Peer Review Panel). The SAB and EDAP have approved the seismic performance criteria needed to produce a seismically safe bridge.

The design for the SAS is in various stages of progress and is not complete. As such the COE Team cannot verify conformance to the design criteria. However, a review of the design work completed to date shows intent to meet the seismic design and performance criteria.

Part 2. How will the currently proposed replacement alternative perform in a Maximum Credible Earthquake (MCE)?

Conclusion: The SAS replacement alternative is not being designed for MCE ground motions, rather it is being designed based on the 1500-year SEE ground motions (further explained in Appendix 6).

The SAS has not been evaluated or designed for the larger MCE event. Caltrans and other authorities have decided to use the SEE and not the MCE.

The Response Spectra graph above compares the various seismic events used to design the SAS. A comparison is made between the SEE and the MCE events for the replacement bridge. The dark, solid line depicts the SEE event, while the dotted line depicts the MCE. For this replacement bridge with its inherent period, the MCE is a greater force than the SEE.

Additionally, the graph shows the seismic event that has been used for the retrofit. The graph shows that the SAS is to be designed with a higher force than the retrofit and is therefore assumed to be more reliable.

Part 3. Specifically, does the currently proposed replacement alternative meet lifeline criteria?

Conclusion: According to the definition of Lifeline criteria contained in the Scope of Work, the current replacement design does not meet lifeline criteria. The Scope of Work defines Lifeline criteria in terms of a MCE event. Caltrans is not designing the SAS to a MCE event; rather Caltrans is designing the SAS to a SEE event. Given this conflict, the COE Team has evaluated the bridge against its given design performance criteria for the SEE event. The SEE event criteria include parameters, in general terms, for design, performance, damage, and repair for both daily and emergency operations.

Design work on the SAS is progressing and is not yet complete. The documents reviewed by the COE Team lack the content necessary to verify conformance with the SEE performance criteria, but do show ability to conform. For instance, it has been noted that the design engineer is using additional analysis to develop his judgment and understanding as a means to predict the performance of the bridge.

However, from the documents presented, it is difficult to verify that correct loads are being used in both analysis and design for such bridge components as the suspension-span tower and deck, as well as the skyway pile caps. For example, member sizes appear assumed or are unidentified in models. Consistent with the assumed intent of the SEE criteria, the ability to replace deck joints in a timely manner remains to be shown. Appendix 8 provides assessment results concerning lifeline criteria, and Minimum and Important Bridges. Appendix 6 provides an assessment of MCE vs. SEE ground motions and additional details and examples of unverified analysis.

Given the ongoing work and the qualifications of the engineers, it is reasonable to believe that conformance issues will be resolved as design progresses. Present review notes a lack of verification and reference in the work presented by the engineer of record. This noted lack of verification can either portend a source of error or it can become an impetus to demonstrate conformance of the bridge to the Design Criteria and its SEE criteria.

Part 4. To what extent and how quickly could it accommodate passenger vehicles?

Conclusion: The COE Team has found no information to indicate how quickly passenger vehicles can be accommodated. According to Document 367 (Volume 1 of 41), the design goal for the bridge is to return to full service almost immediately after an earthquake. The term “full service almost immediately” after an earthquake is not defined in Document 367, Design Criteria.

Information in Documents 344 shows an expected time-scenario for a post-seismic event. This is not a design document, but is the basis used to develop Design Criteria found in Document 367, Volume 1. The post-earthquake scenario calls for steel plates to be placed at failed deck joints within hours to allow for traffic at reduced speeds. Construction activities to replace deck joints would begin within 3 months. [Document 344]

The development of the design is based on the elements of the structure remaining essentially elastic during the SEE. Displacement damage is assumed to be limited to replacement of the deck joints. It is assumed by Caltrans that any other damage can be addressed without impacting traffic. [Document 367]

In summary, no design information demonstrates restoration of traffic for any time frame other than Document 367, Volume 1, which requires “…almost immediate” service.

Recommendations

As indicated in the Scope of Work, actions needed to answer the Questions should be identified. In response, the following actions should be considered to further answer, or refine answers for Questions 3 and 4:

Design Calculations should be completed for a comprehensive document. This document should be complete with references, narratives, discussions, and conclusions. The intent is to provide a ready reference for the bridge owner. Future engineers will be able to rapidly determine the designer’s intent to facilitate the work for repairs, modifications, etc.

An independent check of the design should be completed.

The bridge should be evaluated for a design that addresses the San Andreas MCE ground motions. These ground motions appears to be more forceful than the SEE ground motions in the period range significant to the bridge.

The possible effects of permanent ground movements on the bridge response should be addressed. These movements are associated with accumulation of seismically induced strains in the soils surrounding and/or beneath the pile foundations.

The stability of the rock slope at Pier 1 should be reviewed to confirm that it is seismically stable and consistent with the Fugro-Earth Mechanics, Inc., recommendations.

A feasibility evaluation should be performed comparing the performance of vertical and battered piles in order to justify the installation costs and complexities of battered piles.

The currently estimated permanent pile settlements during an earthquake should be checked during the iterative design process.

Consideration should be given to performing a cyclic pile load test to check the assumed soil degradation rates.

Movement at joints should be evaluated and prototype joints should be laboratory tested with loadings that would simulate the MCE displacement demands.

A constructibility review should be performed for the bridge. In particular, the COE Team has identified the pile cap/ pile connection as a prime focus. The bridge design should be reviewed for constructibility to ensure reliable conformance to the SEE performance criteria.


SUPPLEMENTAL QUESTIONS

During the course of this study many questions, some of which were not specifically contained in the agreed to scope of work, were asked of the COE Team. In an effort to help all parties reach agreement and make informed decisions, the COE Team, within the context of its mandated scope of work, has addressed these questions in Appendix 9.

List Acronyms and Abbreviations

A/E Architecture/Engineering AASHTO American Association of State Highway and Traffic Transportation Officials ADT Average Daily Traffic AISC American Institute of Steel Construction AISC American Institute of Steel Construction AP Advance Planning API American Petroleum Institute ARS Accelerated Response Spectra ASA Assistance Secretary of the Army, Civil Works ATC Applied Technology Council BDS Bridge Design Specifications (Caltrans’) Caltrans State of California Department of Transportation CISS Cast-In-Steel-Shell City (the City) City and County of San Francisco COE US Army Corps of Engineers CPT Cone Penetration Test CPT Cone Penetrometer Test D/C Demand Capacity Ratio DEIS Draft Environmental Impact Statement EDAP Engineering and Design Advisory Panel – MTC Task Force Established Early 1997 EIS Environmental Impact Statement EQ Earthquake FEE Functional Evaluation Earthquake FEM Finite Element Model F-EMI Fugro-Earth Mechanics Inc. FHWA Federal Highway Administration GAR Governor’s Action Request GP General Plans LRFD Load Resistance and Factored Design M&O Maintenance and Operations MCE Maximum Credible Earthquake MTC Metropolitan Transportation Commission – Regional transportation planning agency for the Bay Area PS&E Plans, Specifications, and Estimates RSA Response Spectra Analysis SA San Andreas SAB Caltrans Seismic Advisory Board – Established Summer 1990 by Governor’s Executive Order D-86-90 SAS Self-Anchored Suspension SEE Safety Evaluation Earthquake SFOBB San Francisco Oakland Bay Bridge SSI Soil-Structure Interaction SSPRP Seismic Safety Peer Review Panel – Established Spring 1997 Task Force MTC Bay Bridge Design Task Force – Established early 1997 TBPRP Caltrans Toll Bridge Peer Review Panel – Established 1994 to review and guide retrofit strategies for State owned toll bridges THA Time-History Analysis UCB University of California, Berkeley YBI Yerba Buena Island