MCEER/NCREE Bulletin Articles: Research
A Seismic Retrofitting Manual for Highway Bridges
by Ian G. Buckle, Ian M. Friedland and James D. Cooper
The Seismic Retrofitting Manual for Highway Bridges was developed under the FHWA-sponsored project titled "Seismic Vulnerability of Existing Highway Construction, " FHWA Contract DTFH6]-92-C-00106, by NCEER. The authors acknowledge the assistance of Dr. John Kulicki of Modjeski and Masters Inc., and Dr. Roy A. Imbsen of Imbsen & Associates Inc., in the review phases of the Retrofitting Manual. Questions and comments should be directed to Mr. Ian Friedland, NCEER, (716) 645-3391.
In 1983, the Federal Highway Administration (FHWA) published a set of guidelines for seismic retrofitting of highway bridges. These guidelines presented what was then considered to be the state-of-the-art for screening, evaluating, and retrofitting of seismically deficient bridges. In the 10 years since publication of the FHWA guidelines, there has been significant progress in understanding the response of bridges and in the development of new and improved retrofit technologies. As a consequence, the 1983 guidelines have recently been updated and reissued as a retrofit manual. This article describes the revision, which was performed by NCEER under contract to the FHWA, and discusses a number of significant changes made in the new manual.
Of all the components of a typical highway system, the highway bridge has been the most closely studied for seismic vulnerability. As a consequence, standards have been developed and adopted nationwide for the seismic design of new bridges. By comparison, the seismic retrofitting of existing bridges is a relatively new endeavor. Only a few retrofitting schemes have been used in practice and, given the present state of knowledge, retrofitting is still somewhat of an art requiring considerable engineering judgment.
The San Fernando, California, earthquake of 1971 caused a significant amount of bridge damage and, as a result, initiated a large research effort into seismic bridge design and behavior in the mid- to late 1970s. The majority of this work was jointly sponsored by the California Department of Transportation (Caltrans), the FHWA, and the National Science Foundation. The primary goal of this effort was to minimize the risk of unacceptable damage during a design earthquake. One result of this effort was the publication by the American Association of State Highway and Transportation Officials (AASHTO) of the first comprehensive national highway bridge seismic design guide, the AASHTO Guide Specification for Seismic Design of Highway Bridges, in 1983, which later became Division I-A of the AASHTO Standard Specifications for Highway Bridges.
The San Fernando earthquake also provided the impetus for the FHWA and Caltrans to address the seismic retrofitting of existing bridges to withstand earthquake forces and movements. The culmination of this effort was the publication of the 1983 FHWA report Seismic Retrofitting Guidelines for Highway Bridges (Report No. FHWAIRD-83/007) which was based primarily on research results and information available in the late 1970s. At the time the guidelines were issued, the existing technology for highway bridge retrofitting was limited and many of the proposed techniques had not been field demonstrated. In the 10 or so years since, new and improved technologies for retrofitting bridge columns and footings have been developed and implemented, together with methods to stabilize soils to prevent liquefaction, and to ensure adequate connections between the bridge superstructure and substructure. Many of these advances in the state-of-the-art are the result of an aggressive research program which was begun by Caltrans following the Loma Prieta earthquake which occurred near San Francisco in October 1989.
The Seismic Retrofifting Manual For Highway Bridges
In order to capture these advances in seismic retrofitting and to make the current state-of-the-art available to bridge owners and engineers, the F14WA initiated a project to update the 1983 guidelines. This effort has resulted in a new document titled the Seismic Retrofitting Manual for Highway Bridges which was completed in early 1994 by NCEER under contract to the FHWA.
The new Manual is based primarily on research conducted during the development of the 1983 FHWA guidelines by the Applied Technology Council, current Caltrans Bridge Design Aids, and recent research conducted at the University of California at San Diego and elsewhere.
The Manual offers procedures for evaluating and upgrading the seismic resistance of existing highway bridges. Specifically it contains:
- A preliminary screening process to identify and prioritize bridges that need to be evaluated for seismic retrofitting;
- A methodology for quantitatively evaluating the seismic capacity of an existing bridge and determining the overall effectiveness of alternative seismic retrofitting measures; and
- Retrofit measures and design requirements for increasing the seismic resistance of existing bridges.
The Manual does not prescribe requirements dictating when and how bridges are to be retrofitted. The decision to retrofit a bridge depends on a number of factors, several of which are outside the scope of the Manual. These include, but are not limited to, the availability of funding, as well as political, social, and economic considerations. The primary focus of the Manual is directed towards the engineering factors.
Before seismic retrofitting can be undertaken for a group of bridges, they must first be classified according to their Seismic Performance Category (SPC). The SPC is determined by a combination of seismic hazard and structure importance.
Seismic hazard is reflected in the Acceleration Coefficient, A, which is assigned to all locations covered by Division I-A of the AASHTO Standard Specifications for Highway Bridges. When multiplied by the acceleration due to gravity, g, the product, Ag, represents the likely peak horizontal ground acceleration that will occur due to an earthquake sometime within a 475 year period. More rigorously, this acceleration has a 10 percent probability of being exceeded within a 50 year time frame.
Bridge importance is not so readily quantified. Two Importance Classifications (1) are specified in the Manual: essential and standard. "Essential" bridges are those which must continue to function after an earthquake or which cross routes that must continue to operate immediately following an earthquake. All other bridges are classified as "standard." The determination of the Importance Classification of a bridge is necessarily subjective and consideration should be given to societal/survival and security/defense requirements.
The societal/survival evaluation addresses a number of socioeconomic needs and includes, for example, the need for access for emergency relief and recovery operations immediately following an earthquake.
Security/defense requirements may be evaluated using the 1973 Federal-Aid Highway Act, which required that a plan for defense highways be developed by each state. The defense highway network provides connecting routes to military installations, industries, and resources not covered by the Federal-Aid primary routes.
An "essential" bridge is therefore one that satisfies one or more of the following conditions:
- A bridge that is required to provide secondary life safety; e.g., a bridge that provides access to local emergency services such as hospitals. This category also includes those bridges that cross routes which provide secondary life safety, and bridges that carry lifelines such as electric power and water supply pipelines;
- A bridge whose loss would create a major economic impact; e.g., a bridge that serves as a major link in a transportation system;
- A bridge that is formally defined by a local emergency plan as critical; e.g., a bridge that enables civil defense, fire departments, and public health agencies to respond immediately to disaster situations. This category also includes those bridges that cross routes which are defined as critical in a local emergency response plan and those that are located on identified evacuation routes; or
- A bridge that serves as a critical link in the security/defense roadway network.
Based on these considerations for seismic hazard and importance, four Seismic Performance Categories (SPC) are defined in the Manual as shown in table 1.
Table 1. Seismic Performance Category
A £ 0.09
0.09 < A £ 0.19
0.19 < A £ 0.29
0.29 < A
These SPC's are assigned differently from those in the AASHTO specifications for new design, where no allowance for structure importance is made in seismic zones with acceleration coefficients less than 0.29. In view of the high cost of retrofitting, it is important to be able to distinguish between "essential" and "standard" structures and especially so in low to moderate seismic zones. Such a distinction also enables a more rational allowance to be made for the nature of the seismic hazard in the central and eastern U.S. where the maximum credible earthquake is expected to be significantly larger than the "design" earthquake (475 year-event). This implies that if an essential bridge in the east is to remain fully operational following a large earthquake, it will need to be retrofitted to a standard higher than that required by the current specification for new construction. This observation is reflected in the assignment of SPC's for essential bridges in table 1.
The Retrofitting Process
Seismic retrofitting is one solution for minimizing the hazard of existing bridges that are vulnerable to serious damage during an earthquake. Because not all bridges in the highway system can be retrofitted simultaneously, the most critical bridges should be retrofitted first. The selection of bridges for retrofitting requires an appreciation for the economic, social, administrative, and practical aspects of the problem, as well as the engineering aspects. Seismic retrofitting is only one of several possible courses of action; others include bridge closure, bridge replacement, or acceptance of the risk of seismic damage. Bridge closure or replacement is usually not justified by seismic deficiency alone and will generally only be considered when other deficiencies exist. Therefore, for all practical purposes, a choice must be made between retrofitting or accepting the seismic risk. This choice will depend on the importance of the bridge and on the cost and effectiveness of retrofitting.
The process of retrofitting bridges involves an assessment of a multitude of variables and requires the use of considerable judgment. It is therefore helpful to divide the process into three major stages. These are:
- Preliminary screening;
- Detailed evaluation; and
- Design of retrofit measures.
Each of these stages is outlined below and described in further detail in the Manual. Figure 1 illustrates the retrofitting process for each SPC.
Preliminary screening of an inventory of bridges is recommended to identify those bridges which are seismically deficient and those in the greatest need of retrofitting. This is particularly useful when a comprehensive retrofitting program is to be implemented.
The Manual describes a method for developing a Seismic Rating System which may be used to prioritize bridges on a highway system according to their need for seismic hazard reduction. Factors considered in the seismic rating process include structural vulnerabilities, seismic and geotechnical hazards, and bridge importance or criticality. In this way, the most hazardous bridges are identified. Bridges high on the list should be investigated further to determine the benefits of retrofitting. Because the decision to retrofit depends on political, social, and economic factors as well as engineering issues, high priority bridges may not necessarily be retrofitted. On the other hand, bridges with a lower priority may need to be retrofitted immediately.
One very important consideration that is not adequately reflected in the Seismic Rating System is the relationship of the bridge to other bridges on the system that may also be damaged during an earthquake. These types of considerations should be made prior to making a detailed evaluation of the seismic capacity of the bridge.
A further consideration when deciding if retrofitting is warranted is the age and condition of the bridge. It would not be rational to spend a large amount to retrofit a bridge with only five years of service life remaining. An unusually high seismic vulnerability may, however, be a justification to accelerate closure or replacement of such a bridge.
A bridge in poor physical condition that is scheduled for nonseismic rehabilitation should be given a higher priority for seismic retrofitting, since construction savings can be realized by performing both the nonseismic and seismic work simultaneously.
The above examples do not represent all possible cases, but they do illustrate some of the principles involved in a retrofitting decision. In most cases, the Seismic Rating System is used as a guide to making retrofitting decisions, but not as the final word. Common sense and engineering judgment will be necessary in weighing the actual costs and benefits of retrofitting, against the risks of doing nothing. Also, the effect on the entire highway system must be kept in mind.
The preliminary screening process recommended in the new Manual is demonstrated in figure 2, where the terms are defined as follows: A is the acceleration coefficient for the bridge site; I is the bridge importance; V is the vulnerability rating of the bridge; S is the soil site coefficient; and E is the seismic hazard rating, which is based on the acceleration and site coefficients.
Two alternative methods for the detailed evaluation of existing bridges are currently available. One is based on a quantitative assessment of the "capacities" and "demands" of individual components of a bridge structure. The other evaluates the lateral strength of the bridge as a complete structure.
Capacity/Demand Method: The first method was proposed in the 1983 FHWA Retrofit Guide and has been used in a modified form by Caltrans since the early 1980's. In this method, the results from an elastic spectral analysis are used to calculate the force and displacement "demands" which are then compared with the "capacities" of each of the components to resist these forces and displacements. In the case of reinforced concrete columns, ultimate capacities are modified to reflect the ability of the column to resist postelastic deformations. Capacity/Demand (C/D) ratios are intended to represent the decimal fraction of the design earthquake at which a local failure of the components is likely to occur. Therefore, a C/D ratio less than 1.0 indicates that component failure may occur during the design earthquake and retrofitting may be appropriate.
An overall assessment of the consequences of local component failure is necessary to determine the need for retrofitting. Retrofitting should be considered when an assessment indicates that local component failure will result in unacceptable overall performance. The effect of potential retrofitting may be assessed by performing a detailed reevaluation of the retrofitted bridge.
The determination of what constitutes a serious consequence of component failure will depend on the importance of the bridge. Collapse of the structure is serious in almost all cases since there is always a potential for loss of life in such an occurrence. In other cases, severe distortions or critical loss of strength will impair the ability of the bridge to carry emergency traffic which is unacceptable for certain important bridges. Repairability of seismic damage is also a consideration. If repairs can be made quickly without serious delays to traffic, damage may be acceptable. This is another area in which engineering judgment is required.
Once it has been determined to consider retrofitting, acceptable methods may be selected from among a number of those suggested in the Manual. If the seismic response of the structure is affected, then a reanalysis should be performed and a new set of component C/D ratios calculated. The new C/D ratios will reflect a change in the size of the earthquake that will cause serious damage. A decision to use any retrofitting method will be based on a relative benefit-to-cost analysis. Hypothetically, this benefit-to-cost analysis may be objective and rigorous, but it is more likely that it will be subjective and based, in large part, on judgment.
Lateral Strength Method: The Manual also provides an alternative analysis approach which examines the lateral strength of the bridge as a system, or at least individual segments of the bridge as a system, and determines, through an incremental collapse analysis, the load deformation characteristics of the bridge up to collapse. The fraction of the design earthquake that can be resisted without collapse is then an indicator of the need for retrofitting and the extent of strengthening required. This procedure therefore determines the strength and ductility of the critical collapse mechanism but it can also be used to identify the onset of damage when serviceability criteria may be important. The method emphasizes deformation capacity rather than strength since, although strength is important, it is less important than the ability to sustain substantial deformations without collapse. It is believed that fewer bridges assessed under this procedure will be found in need of retrofit than by the C/D ratio method. When retrofit is required, it should be less extensive. The increased level of effort required of the designer will then be offset by reduced retrofit costs in the field.
There are two alternative strategies that a designer may adopt when faced with retrofitting a bridge. One is based on conventional strengthening techniques which increase the capacity of the structure to meet the likely demand. This is the most common approach used in the United States at this time. The second strategy is based on reducing the demand on the structure such that its existing capacity is sufficient to withstand the given earthquake. This latter approach involved the use of an earthquake protective system, such as seismic isolation or the addition of a mechanical energy dissipation device. Both strategies are described and detailed in the Manual.
Conventional Retrofit Measures: The new Manual describes conventional retrofit measures for bearings, seats, and expansion joints, including joint and bearing restrainers, bearing seat extensions, and overall bearing replacement. The Manual also discusses techniques for strengthening reinforced concrete substructures through column jacketing and wrapping, cap/column and column/footing joint strengthening, and cap beam retrofitting. Retrofit measures for foundations include strengthening footings for flexural and shear strength, ensuring adequate reinforcement anchorage, and sufficient strength in the pile/footing connection to resist overturning or uplift, along with problems related to abutments and approach slabs. Finally, the Manual also discusses problems associated with hazardous sites, including sites with liquefiable soils, bridges on or near unstable slopes, and bridges crossing or near active faults. These techniques represent the current state-of-the-art; however, the art is changing rapidly at this time.
Earthquake Protective System: The term "earthquake protective system" includes passive and active devices which can be installed in a bridge to minimize the seismic demand on the members of the structure. Active systems are considered outside the scope of the Manual but passive devices are being used in several states as a cost-effective retrofit measure for many bridge types. Passive systems discussed in the Manual include mechanical systems, which simply dissipate energy and thus reduce response, and seismic isolation systems, which change the natural period of a bridge so that earthquake loads are significantly reduced. The Manual provides a discussion on seismic isolation concepts and on some of the options available at this time for design and implementation.
The new Seismic Retrofitting Manual for Highway Bridges contains detailed information about each of the major steps in the bridge seismic retrofit process. It describes procedures for preliminary screening of bridges along with two alternate procedures for the detailed evaluation of them. These evaluation methods include a quantitative evaluation of the C/D ratios for individual bridge components, and an alternative method based on assessment of a structure's lateral strength.
The procedures for evaluating bridges for retrofitting also include the identification and assessment of retrofit measures. A number of potential retrofitting measures and retrofit design requirements are discussed in the Manual. Specifically, retrofit measures for the types of bridge components which have performed poorly during past earthquakes are discussed in detail. Retrofitting by these or other equivalent methods should be considered when components are identified by a detailed evaluation as being deficient.
Detailed design of retrofit measures should be performed using the guidelines contained in the Manual in conjunction with the current AASHTO bridge design specifications. If possible, components which are selected for retrofitting should generally be strengthened to conform to the specifications for new construction, even though the structure may otherwise be seismically deficient.
The decision to use a retrofitting scheme will be based on an assessment of its effectiveness in preventing unacceptable overall performance, the cost of retrofitting, and the remaining service life of the bridge. The Manual also includes several worked example problems intended to help illustrate the process of planning the retrofitting of a typical highway bridge.
The Manual was completed in the spring of 1994 and submitted to FHWA. It is expected that it will be published and distributed by the FHWA in the fall of 1994.
NCEER Bulletin, April 1994, Vol. 8, No. 2