Design Guide for Bridges for Service Life (2013)

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.

49 2.1 introduction Selecting the proper bridge system and incorporating service life design principles into the planning and design process are critical steps in achieving long-term bridge service life. As it is more cost-effective to address service life at the design stage, the design for service life must be approached in a systematic, coherent manner. This chapter provides essential information, steps, and guidelines for selecting and designing optimum bridge systems for both existing and new bridges. More specific details for certain bridge elements, components, subsystems, and materials are pro- vided in other chapters to which the reader is directed. Commonly used bridge systems are examined along with their associated chal- lenges and solutions, with a focus on durability and service life. The discussion covers conventional bridge systems and newer, innovative systems involving accelerated and modular construction. Steel and concrete bridge superstructure types are discussed, but they are not directly compared. Instead, the discussion addresses various service life issues within both steel and concrete superstructures. Section 2.2 provides general information and the advantages and disadvantages of various bridge elements, components, subsystems, and systems currently in use. Section 2.3 summarizes the factors affecting the service life of bridge elements, components, subsystems, and systems using a fault tree analysis approach. (A detailed description of the fault tree is provided in Chapter 1.) Section 2.4 provides strategies that can be used to avoid or mitigate most of the factors affecting service life, along with options for enhancing the service life of those factors. 2 BRiDGE SySTEM SELECTiON

50 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Section 2.5 describes a systematic approach for selecting the most appropriate bridge systems that will accommodate operational requirements and site conditions while achieving the desired target design service life. In addition to primary system selec- tion factors relating to function and initial cost, the necessity of considering service life factors (e.g., importance, potential for obsolescence, element and material durability, element maintenance and possible replacement, and life-cycle cost) is also stressed. Above all, the strategies emphasize that durability and service life of all bridge elements, components, subsystems, and systems must be addressed during the system selection process for new bridges as part of a comprehensive approach to service life design. A similar approach should be implemented for existing bridges as part of a comprehensive plan for extending service life. 2.2 bridge SyStem deScriPtion 2.2.1 Bridge System terminology Guide definitions for various bridge terms—bridge element, component, subsystem, and system—are presented in Chapter 1. The bridge system is the total of all elements, components, and subsystems that make up an entire bridge. As shown in Figure 2.1, a bridge system is initially subdivided into three main components: deck, superstructure, and substructure. These are the primary categories or groupings of subsystems and elements within a bridge that define specific purpose and function. The deck component supports and receives live load and must provide a safe, smooth riding surface for traffic. It transfers live load and deck dead load to other components, which in most cases is to the superstructure. The superstructure compo- nent supports the deck and transmits loads across the span(s) to the bridge supports. The substructure component includes all elements that support the superstructure. It transfers vertical and horizontal loads from the superstructure to the foundation material, such as soil or rock. At abutments, additional vertical and horizontal loads applied from the roadway embankment are also resisted. Often bridge systems are categorized or named by the superstructure type and material. Superstructure components are discussed further in Section 2.2.3. Figure 2.1. Bridge system composition. Bridge System Substructure Component Superstructure ComponentDeck Component

51 Chapter 2. BRiDGE SySTEM SELECTiON 2.2.2 Deck Component 2.2.2.1 Deck Elements Figure 2.2 shows the various elements that make up the deck component, which in- cludes the deck–slab element itself along with other related elements including over- lays and wearing surfaces, expansion joints, drainage elements, railings, and curbs and sidewalks. There are various types of deck–slab elements, including concrete decks [either cast-in-place (CIP) or precast], steel orthotropic decks (including open or con- crete-filled steel grids), and other types including timber and fiber-reinforced polymer. A detailed discussion of bridge decks and related service life issues is included in Chapter 4. Most decks are composite CIP concrete types, but other types composed of precast concrete panels (both partial depth and full depth) and posttensioning have been used, particularly with accelerated construction techniques. Steel deck types, including steel orthotropic decks, are also discussed in Chapter 4. A thorough look at materials used in bridge decks is provided in Chapter 3, and Chapter 9 examines deck expansion devices and joints. 2.2.2.2 Bridge-Deck Drainage The deck drainage subsystem includes inlets or scuppers, pipes and downspouts, and outlets. The main requirement of this subsystem is to remove rainfall-generated runoff from the bridge deck before it collects and spreads excessively in the gutter to encroach on the traveled roadway. The deck drainage subsystem must be designed to deter flow and accompanying corrosive deicing chemicals from contacting vulnerable structural Figure 2.2. Deck component. Deck/Slab Deck Component Steel/Orthotropic Deck Concrete Deck Expansion Joints Open/Sealed Drainage Subsystems Deck/Open Exp Joint RailingsOther: Timber, FRP Overlays Curbs and Sidewalks

52 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE members. Proper maintenance of deck drainage elements is essential to avoid clog- ging and malfunction, and such maintenance requirements must be considered in the design. Open expansion joint drainage includes collection troughs, pipes, and attachments below open expansion joints, such as tooth or sliding plate dams, which collect drain- age, debris, and deicing chemicals that flow through the openings and protect adjacent structural elements. Again, proper maintenance is essential and must be factored into the design. 2.2.2.3 Bridge Railings Materials used in bridge railing designs include various combinations of metal and reinforced concrete. Crash testing requirements have been established by FHWA and AASHTO to provide adequate strength depending on vehicle size and speed. Three general categories of bridge railings are typically considered: traffic railings, pedestrian or bicycle railings, and combination railings. • Traffic railings are designed to contain and safely redirect vehicles. • Pedestrian or bicycle railings are generally located on the outside edge of a bridge sidewalk and are designed to safely contain pedestrians or bicyclists. AASHTO specifications require certain heights and limit the opening sizes between members. • Combination railings are dual-purpose railings designed to contain both vehicles and pedestrians or bicyclists; these railings are generally located at the outside edge of a bridge sidewalk. With this type of railing, there is usually no other barrier between the roadway and sidewalk. Bridge railings are often located in high-splash zones and often subject to harsh environments that effect steel element corrosion, concrete deterioration, and reinforc- ing bar corrosion. Special protection is necessary to ensure long-term service life of these elements. Bridge rails are usually cast, following the deck casting. In these instances, special attention should be paid to the cold joint that will be created between the deck and CIP rail as it provides a natural path for ingress of moisture and causes reinforcement corrosion. 2.2.2.4 Curbs and Sidewalks Curbs and sidewalks are affected similarly to deck slabs. Information relative to these elements is provided in Chapter 4 on bridge decks and Chapter 3 on materials. 2.2.3 Superstructure Component The superstructure component includes the structural subsystem and bearings. A de- tailed discussion of bearing elements is given in Chapter 10. Figure 2.3 shows the vari- ous subsystems and elements that make up the superstructure component.

53 Chapter 2. BRiDGE SySTEM SELECTiON Superstructures are often categorized by • Material type. Steel or concrete is most commonly used. • Structure subsystem type. Girder subsystems are most often used for common span lengths within the 300-ft limit. Longer spans typically use girders, trusses, arches, or cable-supported types, depending on span length. • Superstructure continuity. Many older bridges were simple spans, but more mod- ern bridges are fully continuous or continuous for live load. Continuous spans provide structural continuity that helps distribute traffic loads in case of excessive deterioration of some of the bridge elements. Structural continuity is especially important in instances of bridges with fracture-critical elements. • Jointless systems. Integral abutment construction is gaining popularity in many states. Integral pier construction is used only occasionally. Additional information on integral construction is provided in Chapter 8 on jointless bridges. • Modular construction. Modular systems using prefabricated superstructure ele- ments such as “topped girders” or preconstructed spans are becoming more popu- lar in situations requiring accelerated construction. Durability of connection de- tails is a concern for the long-term service life of these systems. These types of connections are addressed in Chapter 8 on jointless bridges. These categories are often combined in an overall classification of the superstructure, which is frequently used to define the entire bridge system. The most common steel and concrete superstructure types are briefly discussed below. Superstructure Component Girder BearingsStructural SubsystemSteel/Concrete Truss Arch Cable Supported Figure 2.3. Super- structure component.

54 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 2.2.3.1 Steel Superstructures 2.2.3.1.1 Steel Girder Superstructures The most common steel bridge superstructures today are composite multigirder sub- systems that use either rolled beams, plate girders, or tub girders. These systems can be single span or multispan and can be either straight or curved. Either of these can also be skewed. Rolled-beam superstructures using W-shapes are used in shorter spans up to about 100 ft for simple spans and up to about 120 ft for continuous spans. Recently, deeper rolled shapes (44 in.) for bridge applications have become available. When combined with the simple for dead load and continuous for live load (SDCL) concept, these W-shapes can be used for longer spans. Welded plate girders are usually used for spans over 120 ft (NSBA 2008). Figure 2.4 shows typical steel I-girder and tub girder systems. Recently, folded-plate beam sections have been developed for short-span ap- plications. See Section 2.2.3.1.4 for steel modular systems. Until the 1970s, many bridges were designed with systems using two main deck girders combined with transverse floor beams and longitudinal stringers. The perceived notion that two-girder systems are not redundant led to a significant decrease in their use within the United States. However, two-girder systems are very popular in Europe. Multigirder bridges with inherent redundancy are currently preferred by many bridge owners (NSBA 2008). Use of high-performance steels with greater fracture toughness, however, has led to a reevaluation of two-girder systems. Further, a FHWA memo dated June 20, 2012, has paved the way to more use of two-girder systems (FHWA 2012). Figure 2.4. Typical steel girder superstructures. Sources: Courtesy (left) HDR Engineering, Inc., and (right) Palmer Engineering. Deck I-girder system. Tub girder system. Sources: Courtesy (left) HDR Engineering, Inc. and (right) Palmer Engineering. Figure 2.4. Typical steel girder superstructures.

55 Chapter 2. BRiDGE SySTEM SELECTiON A variation to the typical multigirder system is the girder–substringer system, which has been used as an economical concept for longer spans beyond approximately 275 ft. This system uses several heavy girders with wide girder spacing and rolled-beam stringers supported midway between the main girders by truss K-type cross frames. 2.2.3.1.2 Continuity in Steel Systems For many years, bridges were designed as a series of simple spans with expansion joints at each pier because they were easy to design and construct. Leaking joints, however, be- came a leading cause of structural deterioration, and the desire to eliminate joints became prevalent. Multispan steel girder systems were also shown to be much more efficient when designed as continuous systems, making continuous design more commonplace. Multispan systems have typically been fully continuous for both dead load and live load, but new systems, typically with spans up to about 150 ft, have been intro- duced with the SDCL concept. These systems combine the advantage of simple-span construction with the efficiency of live load continuity and the durability of not having joints that can ultimately leak. Recently, extensive research studies have developed practical details for SDCL steel bridge systems (Azizinamini et al. 2003, 2005a, 2005b; Azizinamini 2014; Lampe et al. 2014; Farimani et al. 2014; Yakel and Azizinamini 2014; Javidi et al. 2014). These studies demonstrate that for the SDCL steel bridge system, continuity for live load can be provided by using steel reinforcement placed over the pier, before casting the deck; however, in order to provide continuity, various girder connection details have been used in practice. Figure 2.5 shows two details in use. Splice plates are sometimes used for top flange connections, which are in tension. The SDCL research studies, Figure 2.5. Steel bridge systems using SDCL concept: continuity (a) with top flange splice and (b) without top flange splice. Sources: Courtesy (left) HDR Engineering, Inc., and (right) UNL. (a) (b) Source: Courtesy (left) HDR Engineering, Inc., and (right) UNL. Figure 2.5. Steel bridge systems using SDCL concept: continuity (a) with and (b) without top flange splice.

56 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE however, do not recommend using such detail. Bottom flanges in compression are typi- cally butted with plates and wedges. The disadvantage of the continuity detail with the top flange splice is that the bolts for connecting the top plate have to be tightened after casting the deck. This require- ment creates additional construction sequencing with a separate closure pour over the pier. 2.2.3.1.3 Long-Span Superstructures Figure 2.6 shows examples of long-span girder, truss, arch, and cable-stayed bridge systems. Steel-plate girder systems have been used for spans up to approximately 500 ft. Spans up to 400 ft have been designed economically with parallel flanges. Variable-depth haunched girders have been used in the 350- to 500-ft range. Use of Figure 2.6. Long-span steel bridge superstructures. Source: All photos courtesy HDR Engineering, Inc.; cable-supported bridge photo by Vince Streano. Long-span plate girder. Continuous truss. Tied arch. Cable supported. Source: All photos courtesy HDR Engineering, Inc.; cable-supported bridge photo by Vince Streano. Figure 2.6. Long-span steel bridge superstructures.

57 Chapter 2. BRiDGE SySTEM SELECTiON high-performance steel (HPS 70W) has shown economy for plate girder and tub girder systems in most span ranges over 150 ft, particularly in hybrid combinations. Studies have shown that hybrid configurations using conventional-grade 50W steel in webs and HPS 70W steel in top and bottom flanges in negative moment regions and bottom flanges in positive moment regions are typically the most economical (Horton et al. 2002). Top flanges in positive moment regions are affected by composite action with the deck and cannot realize enough benefit from the use of higher-strength steel to be economical, except for longer spans. Use of HPS 70W steel in long-span negative moment ranges can also permit economical parallel flange design without expensive haunches. Trusses, arches, and cable-stayed and suspension systems have also been used for longer-span applications, typically over 500 ft. For spans up to 300 ft, deck girder systems are the most applicable. Long-span structures can have special needs for addressing long-term service life relating to unique details, inspection, and maintenance. Access for inspection and maintenance can require elaborate systems of inspection walkways and access ladders, particularly for access to fracture-critical members. Older trusses typically require intensive maintenance because of large surface area–to–weight ratios and riveted, built-up members with lacing bars subject to pack-out and other surface corrosion. Truss joint details typically have moisture and debris traps that initiate corrosion, but newer trusses have cleaner surface details that are more easily painted and maintained. Through structures are subject to splash-zone wetting environments for all struc- tural elements near roadway edges. This wetting needs to be considered in a corrosion- protection and maintenance plan. Long-span bridges have large thermal movement requirements that result in large expansion joints. This situation requires additional attention to joint maintenance to prevent deck drainage from spilling through. Heavy loads and large thermal movements also require special bearing designs. Navigation channel crossings are subject to vessel collision and need to be protected. 2.2.3.1.4 Steel Modular Systems New steel systems that provide for accelerated construction include modular construc- tion with a pretopped deck. Modular orthotropic deck systems are also a consideration. These modular systems require special attention to both transverse and longitudinal connection details for achieving long-term durability. Pretopped modular bridge systems are best suited for accelerated bridge construc- tion applications. In these systems, several units consisting of pretopped steel or con- crete girders are placed side by side and joined together using longitudinal closure pours. The service life of these pretopped modular systems is significantly influenced by the service life of the longitudinal closure pour. Pretopped steel modular systems present two major advantages: (1) the use of steel girders significantly reduces the creep and shrinkage deflections, and (2) pretopped steel modular units weigh less. The folded-plate bridge system is a new modular system that offers an economical solution for many short-span bridge applications. The system consists of a series of standard shapes that are built by bending flat plates into inverted tub sections by using a press break, as shown in Figure 2.7.

58 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE The maximum span length for this system is currently about 60 ft, and it is limited by the press break length capacity available in the industry. The process of bending the plate to form a girder can take less than an hour. Geo- metrical variations are obtained simply by changing the bend locations. Span-length requirements are accommodated by varying the depth of the web and the width of the girder top and bottom flanges, thereby providing additional capacity. The ability to provide rapid delivery is one of the major advantages of this system, which uses only 0.375- or 0.5-in.-thick plates. Minimizing plate thicknesses allows for stocking standard plate sizes, which means the girders can be produced and delivered quickly (Azizinamini 2009). Another advantage is that this system can be adapted for accelerated bridge con- struction techniques as well as conventional construction methods. In the case of conventional construction procedures, the deck can be easily formed and constructed using readily available construction equipment, as shown in Figure 2.8 (Azizinamini 2009). In the case of accelerated construction, this system easily allows for rapid con- struction of short-span bridges using prefabricated, pretopped-girder elements. This capability supports the recent trend within the bridge construction industry toward minimizing the interruption of traffic by reducing the amount of construction activity at the bridge site. In the pretopped-girder concept, the tributary width of the concrete deck for each folded-plate girder is cast on the girder before being shipped to the site. In this case, each prefabricated girder unit is a folded-plate girder with a precast deck, as shown in Figure 2.9. The steel girder can be supported at the ends or continuously supported along the length during casting, in which case all dead loads are carried by the com- posite section, thereby reducing deflections (Azizinamini 2009). Figure 2.7. Making of folded-plate girder using break press. Source: Courtesy University of Nebraska– Lincoln (UNL).

59 Chapter 2. BRiDGE SySTEM SELECTiON A typical two-lane rural-road bridge would require only three or four prefabri- cated folded-plate girder units placed side by side and connected longitudinally at the deck, as shown in Figure 2.10. The units can be connected by a variety of methods. Further, construction can use relatively lightweight cranes as a 40-ft-long folded-plate girder with a precast deck segment weighs only about 24,000 lb (Azizinamini 2009). Source: Courtesy UNL. Figure 2.8. Deck forming using conventional approach. Figure 2.8. Deck forming using conventional approach. Source: Courtesy UNL. Source: Courtesy Massachusetts Department of Transportation (MassDOT). Figure 2.9. Precast folded-plate girder unit. Figure 2.9. Precast folded-plate girder unit. Source: Courtesy Massachusetts DOT.

60 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 2.2.3.2 Concrete Bridge Superstructures Several reinforced concrete bridge systems are commonly used in the United States. The type of system implemented at a particular site is generally dictated by economics and the system’s ability to accommodate the required span or geometric requirements such as curvature. The most commonly used concrete bridge superstructures are • CIP concrete slabs; • Precast concrete box beams, including both spread and adjacent box beams; • Precast concrete I-girders, including standard I-girders, bulb-tee girders, and U-beams; • Precast concrete spliced girders, including spliced I-girders, U-beams, and box girders; • CIP posttensioned box girders; • Segmental posttensioned concrete box girders, including both precast and CIP; • Concrete arches; and • Modular pretopped concrete girder units, which are typically used for accelerated bridge construction. 2.2.3.2.1 Cast-in-Place Concrete Slabs Full-depth, CIP concrete slab superstructures consist of a concrete slab that spans substructure units without the aid of supporting stringers, as shown in Figure 2.11. Concrete slab bridges commonly span less than 50 ft and are typically used over minor Source: Courtesy MassDOT. Figure 2.10. Folded-plate girder system. Fig r 2.10. Folded-plate girder system. Source: Courtesy Massachusetts DOT.

61 Chapter 2. BRiDGE SySTEM SELECTiON water crossings. This bridge system was traditionally constructed as a series of simple spans, but in recent years, the use of continuous spans has gained favor, eliminating the joints over the substructure units. This system is commonly reinforced convention- ally, but it can also be posttensioned to increase the span-length range. The haunched posttensioned concrete slab system used can span up to about 100 ft. Many states, especially in the Midwest, own many older concrete slab bridges, mainly constructed in the 1930s, which have a very good performance history. When rated, these older concrete slab bridges usually demand posting. However, research results indicate that older concrete slab bridges possess reserve capacity significantly more than that in- dicated by routine rating calculations (Azizinamini et al. 1995a, 1995b). The main reason for the high capacity of older concrete slab bridges is the higher yield strength of the reinforcement used versus the assumed value in rating calculations. This higher capacity of existing older concrete slab bridges coupled with their good performance record can be advantageous when developing maintenance plans. 2.2.3.2.2 Precast Concrete Box Beams This type of superstructure consists of adjacent precast concrete box beams with a noncomposite deck, adjacent box beams with a composite CIP concrete deck, and spread box beams with a composite CIP concrete deck. Shallower precast solid and voided rectangular slabs also fall into this category. Precast concrete box beams are typically plant-manufactured standard AASHTO-PCI (Precast/Prestressed Concrete Institute) sections that range in depth from 27 to 42 in. and are available in 36- or 48-in. widths. These precast girders are plant produced, which generally results in high-quality products. Figure 2.11. Short-span concrete bridge applications. Source: Courtesy Atkins North America, Inc. CIP concrete flat slab bridge. Transversely posttensioned prestressed slabs. Sourc : Courtesy Atkins North America, Inc. Figure 2.11. Short-span concrete bridge applications.

62 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE NCHRP 2009 provides a synthesis of current practice relating to precast adjacent box beam bridges. This superstructure type is the most prevalent box girder system for short- and medium-span bridges, typically 20 to 127 ft, especially on secondary road- ways. These bridges consist of multiple precast concrete box beams that are butted against each other to form the bridge deck and superstructure. Their advantage is that they eliminate the need for forming when using a composite CIP deck, or they can be used directly with a bituminous overlay in the noncomposite state. Adjacent box beams are generally connected using partial or full-depth grouted shear keys along the sides of each box. Transverse ties are usually used in addition to the grouted shear keys and may vary from a limited number of threaded rods to several posttensioned tendons. In some cases, no topping is applied to the structure, but in other cases a noncomposite topping or a composite structural slab is added. Problems have been encountered with adjacent noncomposite box beam superstructures; these problems are discussed in Section 2.3.3.2.2 (Hanna et al. 2009; NCHRP 2009). 2.2.3.2.3 Precast Concrete Girders Precast concrete I-girders with a composite CIP deck are commonly used concrete bridge superstructures in the 50- to 150-ft-plus span range. These girders are made of high-performance, plant-produced materials and are generally very durable and result in high-quality products. In a bridge system consisting of I-girders with composite CIP slabs, commonly referred to as beam-slab bridges (see Figure 2.12), the longitudinal stringers are often prestressed concrete I-girders using one of six standard AASHTO- PCI sections, Types I through VI, which vary in depth from 28 to 54 in. In addition, Figure 2.12. Prestressed concrete I-girder bridge with composite concrete deck. Source: Courtesy Atkins North America, Inc.

63 Chapter 2. BRiDGE SySTEM SELECTiON newer standard AASHTO-PCI bulb-tee shapes are used in 54-, 63-, and 72-in. depths. These standard I- and bulb-tee shapes accommodate various span requirements up to about 170 ft. Bulb-tee shapes were developed to provide increased efficiency over the original I-shapes. They have wide top flanges, similar to Type V and VI girders, which increase stability for handling and shipping and reduce deck forming. However, bulb-tees also have thinner top flanges, webs, and bottom flanges that reduce weight, and have other flange geometric and proportioning modifications that optimize the sections. A num- ber of states, including Washington, Colorado, Florida, and Nebraska, have developed special bulb-tee shapes by modifying the standard AASHTO-PCI bulb-tee shapes to accommodate local needs and practice. A noteworthy I-girder advancement is the NU I-girder, which was developed by the University of Nebraska–Lincoln (UNL) in cooperation with the Nebraska Depart- ment of Roads and has a series of eight standard shapes with depths ranging from 29.5 to 94.5 in. (Geren and Tadros 1994; Hanna et al. 2010). The NU girders have several section efficiency enhancements, such as wide and thick bottom flanges that enable increased strand capacity for simple spans and provide increased negative moment capacity for continuous spans. The wide bottom flanges also provide increased stabil- ity in shipping and handling. Curved fillets in the top and bottom flanges reduce stress concentration and aid the flow of concrete during fabrication. With the increased section efficiencies, these girders have been used for spans greater than 200 ft (see Figure 2.13). Figure 2.13. A 9-ft- × 3-in.-deep, 213-ft-long, 130-ton NU I-girder being shipped. Source: Courtesy Con-Force Structures, a division of Armtec Limited Partnership, Calgary, Alberta, Canada.

64 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE In many instances, precast concrete I-girders are erected as simple spans and then connected over the piers to form continuous for live load systems that eliminate deck joints. A newer alternative to concrete I-girders is the U-beam, or concrete tub girder, first developed in Texas and now used in other states (including Florida and Washington State), which provides economic and aesthetic spread beam systems. The Texas U54 beam is 54 in. deep, similar to an AASHTO Type IV beam, and can span up to about 140 ft (Ralls et al. 1993). The Florida U-beams have four depths ranging from 48 to 72 in. and can be used in spans ranging from about 100 to 160 ft (Florida DOT 2012). Washington State U-beams are similar and have four depths varying from 54 in. to 72 in. and bottom flanges that are either 4 or 5 ft wide. Similar to the Florida U-beams, these beams can accommodate span lengths up to 160 ft. With a composite concrete deck, U-beams form a trapezoidal box shape, similar to steel tub girders. These beams are typically designed to act as simple spans under both dead load and live load, even when the deck is placed continuous across intermediate supports. As with I-girders, these beams are plant produced and result in high-quality products. 2.2.3.2.4 Precast Concrete Spliced Girders Spliced girders are precast concrete girders fabricated in several segments that are then assembled longitudinally, typically using posttensioning, into a single simple-span or continuous girder for the final bridge structure. They have been used to extend the span lengths of regular short- to medium-span precast concrete girders and are de- signed to utilize the economy and high quality of plant-produced precast girders for longer-span applications. The length and weight of typical precast girders prevents them from being effectively used on spans greater than about 150 ft because of the limitations of transportation equipment and available cranes. However, with spliced girders, precast girder segments with manageable weights and lengths are transported to the construction site and then joined together. This procedure can either be done by splicing girder segments on the ground and erecting them into their final position or by placing girder segments on temporary supports and then splicing them in their final position. Spliced girders have been used for simple spans up to about 220 ft and for continuous spans up to about 320 ft; they have been found to provide an economical concrete superstructure type in span ranges between that of conventional precast gird- ers and segmental box girders. Figure 2.14 shows a typical spliced girder span under construction using temporary supports. Spliced girders are typically used on relatively straight alignments; however, in recent years they have also been used for curved alignments in Nebraska and Colo- rado. Figure 2.15 shows a spliced box girder bridge recently built in Denver, Colorado. Precast spliced girders have some similarity with segmental box girders in that both structure types consist of smaller girder segments that are assembled and con- nected by posttensioning to form a final, longer girder, and both types are erected by staged construction. However, spliced girders and segmental box girders are quite different in the length of segments, types of splices, types of sections, tendon loca- tions, and construction methods. A composite concrete deck is typically cast on spliced

65 Chapter 2. BRiDGE SySTEM SELECTiON Figure 2.14. Spliced I-girder construction. Source: Courtesy HDR Engineering, Inc. Figure 2.15. Spliced concrete curved boxes. Source: Courtesy Summit Engineering Group.

66 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE girders, but the deck slab is typically cast as an integral part of a segmental box girder. Spliced girders use bulb-tee, I-beam, U-beam, or box shapes; segmental box girders are typically box shapes. 2.2.3.2.5 Cast-in-Place Posttensioned Concrete Box Girders on Falsework Posttensioned concrete box girders cast continuously on falsework have become very popular in several states, particularly California, Arizona, and Nevada, and have been used in spans up to about 350 ft. This type of construction lends itself to local con- struction industry practices in which contractors can economically provide the re- quired falsework. Similar to segmental construction, CIP on falsework offers the ad- vantage of longer spans than conventional girders and can easily accommodate curved alignments. CIP construction also allows clean lines and architectural finishes that improve the aesthetics. The use of posttensioning further enhances concrete durabil- ity by providing a superstructure that will remain essentially crack free under service loads. Designing the structures as a frame and using monolithic connections between the superstructure and piers also eliminates bearings, which further eliminates associ- ated future maintenance. A potential disadvantage of this type of construction is diffi- culty in replacing the deck or widening the bridge. Figure 2.16 shows a CIP box girder bridge under construction. 2.2.3.2.6 Segmental Posttensioned Concrete Box Girders (CIP and Precast) Segmental concrete box girder systems have been used when span requirements are greater than what can be achieved with conventional stringer-type girders or spliced girders, in instances of sharp horizontal curvature, or when special aesthetics are re- quired. They have been economical in span ranges from about 250 to 500 ft. This system is further divided into cantilever construction and span-by-span construction and can be either precast or CIP. They can be cast to match the shape of any alignment, Figure 2.16. Cast-in-place box girder bridge on falsework. Source: Courtesy Atkins North America, Inc.

67 Chapter 2. BRiDGE SySTEM SELECTiON making them particularly suited to curvature. Figure 2.17 shows a CIP segmental bridge recently built in Florida by using balanced cantilever construction. Segmental box girder bridges have been observed to improve deck performance as a result of precompression of the deck. Refer to Chapter 3 on materials for additional information on these bridge-deck systems and durability issues relating to details cur- rently in use. 2.2.3.2.7 Concrete Arches Concrete arches have been used for bridges with short spans of about 100 ft to long spans of over 1,000 ft. They are typically considered today only in certain long-span applications because of the relative economy of I-girder and segmental box girders in shorter spans, or when special aesthetics are required. True arches are efficient struc- tural systems because vertical dead load produces axial member compressive forces that are resisted by a thrust at the arch abutments. Concrete has been useful for arches because of its inherent efficiency in compression. Concrete arches have typically been used in deck-type systems in which the arch ribs are below the deck, but they have also been used in some through-type applica- tions in which the arch ribs extend above the deck. Deck arch systems are either closed spandrel types or open spandrel types. Closed spandrel types typically use barrel arches with longitudinal walls along the outside edges of the arches that are either filled or unfilled. Open spandrel types have a series of spandrel columns that transmit deck loads to the arches. Concrete arches in the United States have typically been constructed using either CIP on falsework methods or cable-stayed segmental methods. Figure 2.18 shows the cable-stayed, CIP segmental construction method used for the Hoover Dam bypass concrete arch bridge. Source: Courtesy HDR Engineering, Inc.; photo on right by John Rupe. Figure 2.17. CIP segmental concrete box system. Figure 2.17. CIP segmental concrete box system. Source: Courtesy HDR Engineering, Inc.; photo on right by John Rupe.

68 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 2.2.3.2.8 Modular Pretopped Concrete Girders These types of systems use precast beam elements that are fabricated with a portion of the deck in place as a unit and are erected side by side and connected with a CIP closure joint, posttensioning, composite concrete topping, or a combination of these methodologies. The precast elements commonly consist of conventionally reinforced or prestressed sections that include T-beams, double Ts, and deck bulb-tees. This system is expected to gain popularity with accelerated bridge construction as pressure mounts to expedite construction and to minimize field forming and placing of concrete. Refer to Chapter 4, Bridge Decks, for information concerning CIP closure connections. A recent concept is the NEXT (Northeast EXtreme Tee) beam, which was devel- oped by PCI Northeast along with the departments of transportation for New York, Connecticut, Massachusetts, Vermont, Maine, New Hampshire, and Rhode Island (Culmo and Seraderian 2010). The NEXT beam is a precast, prestressed double-tee section with 8- or 12-ft deck widths and beam depths from 24 to 36 in. that is appli- cable for approximately 40- to 90-ft spans. It is available with a thick top flange that comprises the deck, or with a top flange that creates a form for a composite CIP deck. The NEXT beam is considered as an alternative to traditional adjacent concrete box beams, providing improved durability, lower cost, easier inspection, and accelerated bridge construction. 2.2.4 Substructure Component The substructure component includes all structural elements required to support the superstructure and is typically defined from the underside of the bridge bearings down through the foundation. The function of these elements is to transfer all verti- cal loads from the superstructure to the foundation-supporting strata and to resist horizontal forces acting on the bridge. The transfer of load to the supporting ground can be through spread foundations, piles, or drilled shafts, depending on the strength Figure 2.18. Hoover Dam bypass concrete arch bridge constructed using cable-stayed segmental methods. Source: Courtesy HDR Engineering, Inc.; photo by Keith Philpott.

69 Chapter 2. BRiDGE SySTEM SELECTiON and stability of the subsurface geotechnical conditions. This component typically in- cludes pier and abutment subsystems, each including several elements, as shown in Figure 2.19. 2.2.4.1 Piers and Bents Piers are intermediate supports for multispan bridges. They can have multiple con- figurations, but typically fall into two major categories, piers and bents, as illustrated in Figure 2.20. A pier subsystem consists of several elements, including a cap beam supporting the main load-carrying elements of the superstructure, which in turn is Substructure Component Abutment SubsystemPier Subsystem Cap, Column(s), Footing(s)/Pile Cap(s), Piles/Drilled Shafts Backwall, Cap Beam, Stem/Breast Wall, Wingwalls, Footing/Pile Cap, Piles/Drilled Shafts, Reinforced Soil Figure 2.19. Substructure component. (a) (b) Source: Courtesy Atkins North America, Inc. Figure 2.20. Pier types: (a) multicolumn supported pier and (b) pile bent. Figure 2.20. Pier types: (a) multicolumn supported pier and (b) pile bent. Source: Courtesy Atkins North America, Inc.

70 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE supported on one or more columns. The columns are supported by foundations that are typically located at or below the finish grade of the adjacent ground. The founda- tion can be a footing bearing directly on rock or soil, or a deep foundation using piles or drilled shafts. T-piers are examples of single-column piers with a cap element. Solid- or wall-type piers are also single-column piers, but are wide enough to support the superstructure without having a separate cap element. A bent consists of a cap beam supporting the main load-carrying elements of the superstructure, which in turn is supported directly on deep foundation elements such as piles or drilled shafts that extend up from the finish grade. In some cases, the term “bent” is also used to describe a multicolumn pier. Common practice is to construct piers with reinforced concrete, although some steel piers with pier caps have been used. When deep foundations are required to sup- port the bent caps, they normally consist of timber; prestressed concrete square, solid round, or hollow cylinder piles; CIP concrete drilled shafts; or high-performance steel or steel pipe sections. Modular, precast concrete pier elements have also been used for accelerated construction. Integral pier cap construction was developed as a way to avoid sharp skews or associated longer spans in interchange ramp bridges and to lower overpass profiles. Integral pier caps also have the advantage of eliminating bearings, which can minimize future maintenance requirements. Figure 2.21 shows a ramp bridge with conventional stacked T-pier construction and a similar ramp bridge with integral pier construction. An integral pier system is advantageous in seismic areas because it integrates the super- structure and substructure and creates frame action. (a) (b) Source: Courtesy HDR Engineering, Inc. Figure 2.21. Ramp bridge with (a) conventional T-piers and (b) integral piers. Figure 2.21. Ramp bridge with (a) conventional T-piers and (b) integral piers. Source: Courtesy HDR Engineering, Inc.

71 Chapter 2. BRiDGE SySTEM SELECTiON 2.2.4.2 Abutments Abutments are provided in multiple configurations, but they can be defined in two major categories, as illustrated in Figure 2.22: stub or spill-through abutments and full abutments. Stub abutments are characterized by sloped embankments under the end span of the bridge. They provide support to the superstructure through a shallow bent cap resting on a pile foundation. Traditionally, full abutments have been characterized by a vertical wall that retains the embankment fill and also transfers the bridge load to the supporting foundation at the base of the wall. Full abutments can also be in the form of a mechanically stabilized earth (MSE) system, which employs a fascia wall connected to a system of reinforcing elements in multiple layers that work with the backfill material to create a composite soil mass. This composite soil mass can then support vertical load and/or act as an earth retention system. There are two types of MSE abutments: true or mixed (Anderson and Brabant 2010). In a true MSE abutment, the bridge superstructure is supported on spread footings bearing directly on the top of the reinforced soil mass. In a mixed MSE abutment, a shallow bent cap with a row of piles is used to support the superstructure behind the MSE fascia wall, and the reinforced soil mass is used to retain the fill behind and adjacent to the end of the bridge. An MSE full abutment is pictured in Figure 2.22. Another recent form of abutment system is the geosynthetic-reinforced soil inte- grated bridge system, which is described in a recent FHWA report (Adams et al. 2011). This is a relatively new abutment system that has been used for accelerated bridge con- struction, typically for short spans up to about 140 ft. The abutment uses alternating (a) (b) Source: Courtesy Atkins North America, Inc. Figure 2.22. (a) Stub or spill-through abutment and (b) MSE full abutment. Fig r 2.22. (a) Stub or spill-t rough abutment and (b) MSE full abutment. Source: Courtesy Atkins North America, Inc.

72 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE thin layers of compacted fill and geosynthetic reinforcement sheets that combine to form a reinforced soil mass foundation that directly supports the bridge superstructure without the need for piles. The geosynthetic reinforcement is connected into layers of precast facing blocks that are placed with the reinforcement and soil backfill. Once completed, the reinforced soil mass is ready to support the bridge. Traditional abutments are typically concrete construction. When deep foundations are required to support the bent caps, they normally consist of timber; prestressed concrete square, solid round, or hollow cylinder piles; CIP concrete drilled shafts; or high-performance steel or pipe pile sections. The types of abutments used also characterize the way the entire bridge system responds to thermally induced longitudinal movements. There are three distinct abut- ment types: 1. Integral abutment; 2. Semi-integral abutment; and 3. Abutment using expansion devices. In integral abutment systems, piles are attached directly to the abutment, and ther- mally induced longitudinal movements are accommodated by the flexibility of the piles. The piles are subject to both axial and flexural moments. In semi-integral abut- ment systems, girders and piles are not directly connected, and the bearings used to support the girders and piles are mainly subject to axial loads. Integral and semi- integral abutment systems are part of different continuous bridge systems. Chapter 8 provides a more in-depth discussion, as well as detailed design provisions, for integral and semi-integral abutment systems. 2.3 FActorS AFFecting Service LiFe All the elements, components, and subsystems that make up the overall bridge system are adversely affected to various degrees by external and internal factors that contrib- ute to reduced service life. External factors typically refer to loads or hazards, which can be both natural and human caused. Internal factors can pertain to such items as structure type (e.g., fracture critical), materials, and design and details. A discussion using a fault tree analysis of critical factors that affect bridge ser- vice life is presented in this section. Figure 2.23 shows the initial fault tree diagram, which identifies factors affecting the service life for a complete bridge system, and Section 2.3.1 discusses these factors. Sections 2.3.2, 2.3.3, and 2.3.4 address specific factors affecting the deck, superstructure, and substructure components, respectively. Section 2.4 addresses options to avoid or mitigate these various factors. 2.3.1 Bridge System fault tree Analysis The fault tree diagram in Figure 2.23 identifies and categorizes causes of reduced ser- vice life or factors affecting service life. (A detailed discussion of fault tree analysis is included in Chapter 1.) In the following sections, these categories are successively sub- categorized at descending levels to identify multiple contributing factors. The fault tree

73 Chapter 2. BRiDGE SySTEM SELECTiON analysis is continued until the basic events or lowest levels of resolution are reached and discussed. The lowest level or basic events require strategies for mitigation. 2.3.1.1 Obsolescence or Deficiency At the highest fault level, reduced service life of a bridge system can be attributed to either obsolescence or deficiency. Obsolescence refers to reduced service life of a bridge as a result of issues related to how it functions, which can be further categorized as • Operational issues related to reduced traffic capacity and safety; • Physical issues related to span layout and clearances; or • Loading issues related to increases in design live load. Many bridges are replaced because of functional issues well before their full poten- tial service life is achieved. Significant increases in corridor traffic demand, caused by such factors as urban planning, land use, and development, can ultimately result in the functional inability of a bridge to provide a required level of service, necessitat- ing bridge widening or replacement. Vertical clearance limitations sometimes prevent Figure 2.23. Factors affecting service life. Reduced Service Life of Bridge System Caused by Obsolescence Due to Natural or Man-made Hazards - Due to Loads Due to Production/ Operation Defects Due to Capacity/ Safety Issues Of Superstructure Component Of Deck Component Due to Span Layout/ Clearance Issues O f Substructure Component Caused by Deficiency Due to Increase in Design Live Load Figure 2.23. Factors affecting service life.

74 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE existing bridges from being widened. Increased corridor traffic can also require replacement of overpass bridges to accommodate widened roadways and increased span requirements below. Major interchanges are sometimes reconstructed because of the need for increased traffic capacity. Often, safety issues relating to inadequate lane and shoulder widths, sharp curves, and inadequate sight distances have a significant effect on service life. Changes in design live load over the life of a bridge can affect service life as it relates to the struc- ture’s ability to safely accommodate increased load. Service life design considerations should evaluate the potential of future opera- tional needs and how those needs might affect the service life of the planned facility. Risks of future obsolescence should be considered, and appropriate choices should be made concerning mitigation or acceptance. Those choices should be incorporated into the design as appropriate considering life-cycle cost analysis. Deficiency refers to reduced service life of a bridge as a result of damage or dete- rioration that can be caused by a variety of primary factors and subfactors, each of which can lead to reduced service life if unmitigated. Deficiency can occur in all three bridge components: deck, superstructure, and substructure. The fault tree in Fig- ure 2.23 continues below the superstructure component, but it applies equally to all three components. Within a bridge system, the interaction between components, deficiencies, or fail- ures within a particular component can have a significant effect on other components. A primary example is deterioration of superstructure and substructure below leaking joints in the deck component (see Section 2.3.1.3.1). Another example is damage to substructure and other superstructure elements caused by failed bearings in the super- structure component (see Section 2.3.4.4). Deficiency can be further attributed to any of three major causes: 1. Load-induced deficiencies; 2. Natural or man-made hazards; or 3. Defects in production and operation. 2.3.1.2 Reduced Service Life Caused by Loads Load-induced deficiencies can be further categorized as those caused by traffic- induced loads or system-dependent loads (see Figure 2.24). The fault tree is continued for each of these load types to identify the basic or lowest levels causing damage or deterioration. 2.3.1.2.1 Traffic-Induced Loads Traffic-induced loads include the effects of truck and other vehicular traffic that are applied to the bridge deck and transmitted throughout the bridge system. Traffic load can ultimately cause damage to bridge elements through fatigue, overload, or wear. Fatigue is structural damage to an element resulting from cyclic loading that results in the initiation and propagation of cracks; it can occur at stress levels considerably below the yield stress. Although fatigue can occur in reinforced concrete and structural

75 Chapter 2. BRiDGE SySTEM SELECTiON steel elements, it is more predominant in steel elements. Chapter 3 on materials dis- cusses fatigue deterioration in reinforced concrete. Early-welded steel structures have a history of cracking at certain types of weld details as a result of load-induced and distortion-induced fatigue. Newer design provi- sions and recommended details have been developed that provide solutions for both load-induced and distortion-induced fatigue that will achieve desired service life. Sec- tion 2.3.3.1.1 provides additional information of fatigue in steel structures, and Chap- ter 7 provides a comprehensive discussion of fatigue and fracture in steel bridges. Overload refers to element overstress or damage resulting from vehicles that exceed maximum gross vehicle weight restrictions or individual axle or tire restric- tions. Overload often results from illegal, nonpermitted vehicles, and it is the third leading cause of bridge failure in the United States behind hydraulic and impact causes (Wardhana and Hadipriono 2003). Overload produces higher stress in members than what was considered in design; it can significantly reduce safety factors against failure, and it can cause cracking in concrete elements. Multiple applications can also affect fatigue behavior and also result in excessive deflection that can affect certain elements, particularly in cases of differential deflection. Figure 2.24. Load-induced deficiencies. Load-Induced System-Dependent Loads Traffic-Induced Loads WearFatigue Time- Dependent Material Properties System- Framing Restraint Thermal Overload Figure 2.24. Load-induced deficiencies.

76 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Because overload occurs on many bridges, the risk of overload should be consid- ered on certain vehicular routes when planning new bridges. It may also be necessary to consider special owner-specified loads to avoid or mitigate this risk. Wear refers to element damage and gradual loss of material caused by friction or rubbing. Decks are susceptible to wear from vehicle tires, especially with the use of studs or chains. Deck wear and abrasion is further discussed in Chapter 4. Wear has been a factor in steel structures, particularly on pins and pin plates in connections, and in bearings, with surface wear in sliding bearings, brass sealing ring wear in pot bear- ings, and pin wear in steel bearings. 2.3.1.2.2 System-Induced Loads System-induced loads include the effects of the bridge system configuration on the behavior of the structure. These effects are accentuated by restraints provided through bridge boundary conditions and can result in significant locked-in stresses. The sys- tem-induced loads can be the result of movements due to time-dependent material properties, thermal movements, or system-framing restraint. Time-dependent material properties refer mainly to shrinkage and creep-related deformations in restrained concrete elements that can result in concrete cracking. This phenomenon is discussed further in Chapter 4 for bridge decks and in Chapter 3 for concrete materials in general. Thermal conditions refer to effects caused by temperature change, which can result in significant stresses in restrained structural members, and in some cases can be as damaging as live load stresses. The effects can be the result of uniform stress across a bridge member or the result of a temperature gradient throughout the depth of a member. System-framing restraint refers to effects caused by boundary condition restraints that prevent normal or intended structural behavior. Improper function or seizing of bearings can result in unintended movement restraint, which can further cause pier cracking and distress. Another example occurs at the ends of skewed integral abut- ments, where lateral movement resulting from the skew can cause cracking and dis- tress in corner details if adequate clearance is not provided to allow for the movement. 2.3.1.3 Reduced Service Life As a Result of Natural or Man-Made Hazards Environmental hazards from both natural and man-made sources can have a signifi- cant influence on bridge service life. These hazards also include effects from areas with adverse thermal climate, coastal climates, and chemical climates, as well as from chemical properties of the materials themselves. Other hazards such as hy- draulic action, collisions, fire or blast, or seismic events can also have considerable effect. These natural and man-made hazards are introduced in the fault tree shown in Figure 2.25.

77 Chapter 2. BRiDGE SySTEM SELECTiON 2.3.1.3.1 Thermal Climate Deicing salts corrosion. Bridge service life is typically severely affected in cold, wet climates because of the heavy use of roadway deicing salt. Salt-contaminated mois- ture penetration directly affects bridge-deck service life by initiating and propagating corrosion in unprotected reinforcing steel and by accelerating concrete deterioration caused by cracking and freeze–thaw damage. All unprotected bridge elements below open or leaking deck joints are subject to salt-contaminated roadway drainage, which causes unprotected structural steel corrosion, concrete reinforcing bar corrosion, and associated concrete cracking and spalling. On overpass bridges, salt spray rising from roadways below affects super structures and the undersides of decks, causing concrete penetration and corrosion of unpro- tected reinforcing and corrosion of unprotected structural steel. The salt spray from vehicles passing underneath the bridge can also affect the service life of weathering steel by keeping the steel continuously wet. Decks, barriers, and deck joints are also susceptible to damage from snow plows used to clear roadways for traffic. Freeze–thaw. Water absorbed into concrete surfaces and contained in cracks can freeze in cold weather conditions. The frozen water tends to expand, causing stresses within the concrete. Cyclic freezing and thawing of the water absorbed in the deck surface fatigues the concrete and results in cracking, scaling, and spalling. Refer to Chapter 3 on materials for additional information on freeze–thaw in concrete. Figure 2.25. Natural or man-made hazards. Natural or Man-made Hazards Thermal Climate Coastal Climate Chemical Climate Fire/Blast Vehicle/ Vessel Collision Reactive Materials Freeze– Thaw Humidity Salt Water/ Spray Corrosion Corrosion- Inducing Chemicals Sulfate Attack ACR ASRDe icing SaltsCorrosion Hydraulic Action Scour Flood/ Storm Surge Seismic Extreme Events i re 2.25. Natural or man-made hazards.

78 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 2.3.1.3.2 Coastal Climate Salt water and spray corrosion. Coastal saltwater environments also have severe effects on bridge service life as a result of wetting and chloride penetration causing corrosion of unprotected reinforcing and unprotected structural steel. Structures in these areas are subjected to a chloride-laden saltwater environment and a combination of wind and wave action that causes these chlorides to become airborne as salt spray. The susceptibility of various bridge components to these envi- ronmental influences depends on their degree of direct contact or their height above the water. Pier columns with direct contact in areas of continual wetting are most susceptible to damage. Wave action hitting substructure units and seawalls or abutments under the bridge tends to cause the salt spray to explode upward, wetting the bottoms of lower-level superstructures and decks. On windy days the salt spray can also land directly on the bridge-deck surface. The salt spray wets the surfaces, leaving a chloride residue that can be absorbed into the concrete, resulting in reinforcement corrosion that in turn can cause cracking, spalling, or delamination. The salt spray can also strike the sides of structural steel members, affecting the service life of coatings and thus directly causing and accelerating steel corrosion. Humidity. High humidity in coastal regions also results in cyclic wetting and dry- ing of bridge surfaces. Concrete materials sensitive to repeated wetting, such as those in which reactive aggregates are used, can have an adverse effect on concrete elements. Continuous wetting and drying also affects coatings on structural steel members and causes steel corrosion. 2.3.1.3.3 Chemical Climate Corrosion-inducing chemicals. Chemical climate influences on bridge service life per- formance can be attributed primarily to airborne corrosion-inducing chemicals from nearby industrial facilities such as chemical plants or oil- and coal-burning facilities. Chapter 3 provides further discussion on chemical influences on concrete, and Chapter 6 discusses the influence of corrosion-inducing chemicals with respect to steel coatings and steel corrosion. Sulfate attack. Exposure to sulfates can cause expansion of concrete material that can cause spalling and cracking and the loss of bond strength between the cement paste and aggregate. Refer to Chapter 3 for additional information on sulfate attack in concrete. 2.3.1.3.4 Reactive Materials Reactive ingredients with the concrete mix can affect concrete service life performance by altering the volumetric stability of the concrete mix design. These influences pri- marily occur naturally. Alkali-silica reactivity (ASR) results in swelling of aggregate particles within con- crete that can lead to spalling, cracking, and general concrete deterioration. Chapter 3 provides additional information on ASR in concrete.

79 Chapter 2. BRiDGE SySTEM SELECTiON Alkali-carbonate reactivity (ACR) results in aggregate expansion within concrete that can lead to spalling, cracking, and general concrete deterioration. Refer to Chap- ter 3 for additional information on ACR in concrete. 2.3.1.3.5 Hydraulic Action Hydraulic action is the leading cause of bridge failure in the United States (Wardhana and Hadipriono 2003). The principal elements of hydraulic action are flood and storm surge and scour. Flood and storm surge. Floods and storm surges can significantly affect bridge service life, including dislodging spans from their bearings and washing them away. Storm surges during major hurricanes are most often the cause of bridge damage. In 2005, for example, Hurricane Katrina devastated the Gulf coastline from Louisiana to the Florida Panhandle and damaged nearly 45 bridges (Padgett et al. 2008). Most of the damaged bridges were adjacent to water, and damage resulted from storm surge– induced loading. Much of the damage was to superstructures, on which typical dam- age included unseating or shifting of decks and failure of bridge parapets. Several bridges suffered damage caused by impacts from loose barges and debris. The most common severe failure was unseating, which often occurred in low-elevation spans. Deck displacements were attributed primarily to a combination of buoyant forces and pounding waves. Superstructure damage largely depended on the connection type between the decks and bents, and the bearings often provided no apparent positive connection between the superstructure and substructure. Figure 2.26 shows the I-10 bridges across Escambia Bay in Florida, which were dislodged during a storm in 2006. Damage caused by superstructure unseating was similar to that experienced by bridges during Katrina. Bridges with low vertical clearance over a waterway can also be vulnerable to damage resulting from debris flow in a flood. Scour. Scour is defined as the erosion or removal of streambed or bank material from bridge foundations as a result of flowing water. Although scour can occur at any time, bridge scour most often occurs during floods when swiftly flowing water has Figure 2.26. Florida I-10 Escambia Bay bridges washed out during a storm in 2006. Source: Courtesy Florida DOT.

80 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE more energy than calm water to lift and carry sediment downriver. A hole is created adjacent to the pier or abutment when material is washed away from the river bottom, exposing or undermining footings, a situation which can compromise the integrity of the structure and lead to failure. Figure 2.27 shows an example of abutment scour. See Section 2.3.4 for factors affecting service life of bridge substructures. 2.3.1.3.6 Extreme Events The following sections describe extreme events that can seriously and abruptly reduce the service life of the bridge system: vehicle or vessel collision, fire, blast, and seismic activity. 2.3.1.3.6a Vehicle or Vessel Collision Vehicle or vessel collision is second to hydraulic effects as the leading cause of bridge failures (Wardhana and Hadipriono 2003). Bridges crossing other roadways with min- imum or low clearance are subject to various types of vehicle collision, particularly involving overheight vehicles. Figure 2.28 shows the effects of a collision in which a truck transporting a hydraulic crane with the boom inadvertently raised struck a con- crete bridge and cut halfway through the entire width of the superstructure. Piers with minimum offset from the edge of the roadway or shoulder are also subject to vehicle collision if not adequately protected by barriers. The risk of vehicle impact should be considered in the design of new bridges, particularly bridges crossing heavy truck routes where a greater probability exists for overheight vehicles, or where there has been a history of impacts from overheight vehicles. Possible mitigation strategies include • Using higher clearances; • Using sacrificial beams to protect load-carrying members; and • Using laser detection systems that set off warning signals if an overheight vehicle is detected. Figure 2.27. Abutment scour. Source: Courtesy U.S. Geological Survey; photo by Bill Colson.

81 Chapter 2. BRiDGE SySTEM SELECTiON Bridges crossing water bodies or waterways are subject to ships colliding with either piers or superstructure. These are rarely occurring extreme events, but they have potentially high consequences. Figure 2.29 shows the aftermath of a ship collision with the original Sunshine Skyway Bridge in Florida. The ship collided with one of the end piers in the main channel three-span unit and took out the pier and subsequently the superstructure unit. Considerations for new bridges should evaluate the span openings required for safe navigation, including horizontal and vertical clearances, and consider appropriate mitigation measures to reduce the risk of collision. Adequate fender systems or other pier protection devices also need to be considered when there is a risk of ship collision. The current LRFD Bridge Design Specifications (LRFD specifications) (AASHTO 2012) provides requirements for new bridge design for both vehicle and ship impacts. 2.3.1.3.6b Fire Fires, as extreme-event hazards for bridges, have a low probability of occurrence, but they can cause significant damage to affected bridge components, including the deck, superstructure, and substructure, and can cause collapse of entire spans. Although fires are considered a low-risk hazard, a recent study by the New York Department of Transportation (DOT) in 2008 showed that nearly three times more bridges have col- lapsed because of fire than earthquakes (Kodur et al. 2010). Fires affecting bridges most typically occur as a result of vehicle accidents either on a bridge or on a roadway or railway crossing below a bridge, but they can also result from fires in adjacent buildings or facilities. Fires can vary in intensity; the most intense Source: Courtesy Kansas Department of Transportation. Figure 2.28. Bridge damaged by truck transporting hydraulic crane. Figure 2.28. Bridge damaged by truck transporting hydraulic crane. Source: Courtesy Kansas DOT.

82 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE are caused by accidents with tanker trucks or railroad tanker cars carrying large quan- tities of highly flammable fuels or chemicals. The temperature of a recent fire below a bridge that was caused by a railroad tanker car collision loaded with 30,000 gallons of methyl alcohol was estimated to be approximately 3,000°F (Stoddard 2002). Recent bridge fires involving tanker trucks carrying diesel fuel and gasoline were reported to have reached temperatures over 2,000°F (Kodur et al. 2010). The extreme high temperatures generated in these types of bridge fires for prolonged periods of time can significantly affect both steel and concrete structures. Figure 2.30 shows an example of a dramatic bridge fire caused by a gasoline tanker truck accident. Steel bridge elements are especially vulnerable to high temperatures because of steel’s high thermal conductivity: the temperature of unprotected steelwork will vary little from that of the fire. These cases can result in loss of strength, significant sagging, and possible collapse. Steel starts to lose strength at about 600°F, and its strength is reduced to about half its yield strength at about 1,100°F (Brandt et al. 2011). At about 1,700°F, the yield strength is only about 10% or less. When fires at steel bridge ele- ments reach these extreme temperatures, significant deformation and sagging usually occurs (if not total collapse), and the affected bridge elements will typically have to be replaced. Figure 2.31 shows extreme sagging in a steel bridge span and heavy concrete pier deterioration after a gasoline tanker truck fire. Figure 2.29. Sunshine Skyway Bridge collapse from ship collision. Source: Courtesy Tampa Bay Times.

83 Chapter 2. BRiDGE SySTEM SELECTiON In cases in which damage to steel bridges sustained during a fire is not obvi- ous (i.e., no clear signs of distress, such as sagging or buckling) the question is often raised as to whether permanent material property effects in heat-affected areas have occurred. It has been reported that steel will begin to encounter phase changes at temperatures greater than 1,300°F, whereby undesirable material properties such as reduced ductility and toughness can result during uncontrolled cooling. The Penn- sylvania Department of Transportation (PennDOT) sponsored a study to examine the effects of fire damage on the structural properties of steel bridge elements (Brandt et al. 2011). The study performed fire tests on painted steel-plate specimens at various tem- peratures up to 1,200°F and exposure times to evaluate changes in surface conditions Figure 2.31. Steel bridge heavily damaged by fire after gasoline tanker truck collision. Source: Courtesy Alabama DOT. Figure 2.30. Intense bridge fire resulting from a tanker truck accident. Source: Courtesy U.S. Fish and Wildlife Service.

84 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE and discoloration and then tested for changes in material properties after cooling. The results showed that up to steel surface temperatures of 1,200°F, the fire-exposed material after cooling still satisfied AASHTO material specifications. The researchers concluded that if excessive distortions or deformations occurred, the steel would likely have been subjected to steel temperatures well in excess of 1,200°F, and the corre- sponding sections would require replacement. Concrete bridge elements are typically able to withstand high temperatures with less distress than unprotected steel elements. Concrete has inherent fire-resistant prop- erties because of its relatively low thermal conductivity, which insulates interior por- tions of the member, including reinforcement and prestressed steel, from high surface temperatures. However, concrete does experience a reduction in strength and modulus of elasticity with high temperature. Strength reduction is largely a function of type and size of aggregate. Concrete with siliceous aggregate (materials consisting of silica and including granite and sandstone) begins to lose strength at about 800°F, and it is reduced to about 55% at 1,200°F. Concrete containing lightweight aggregates (manu- factured by heating shale, slate, or clay) and carbonate aggregates (including limestone and dolomite) retains most of its compressive strength up to about 1,200°F (Bilow and Kamara 2008). The following list compares concrete temperature with typically encountered signs of distress and concrete color (Shutt 2006): • Up to 200°F—Little or no concrete damage; • 500°F—Surface crazing, localized cracks, iron-bearing aggregates begin to acquire pink or red color; • 700°F—Cracks appear around aggregate, numerous microcracks present in cement paste; • 900°F—Purple or gray color appears if enough iron and lime are present; • 1,000°F—Serious cracking of paste and aggregates occurs because of expansion. Purple or gray color may become more pronounced; • 1,500°F—Cement paste is completely dehydrated with severe shrinkage cracking and honeycombing. Concrete may begin to become friable and porous; • 2,200°F—Some components of concrete begin to fail; and • 2,500°F—Concrete fails completely. In all cases, fire-damaged structures should be evaluated as quickly as possible once the fire is extinguished to determine the extent and severity of damage. The limits of concrete damage can often be tested with an impact–rebound hammer. Concrete core samples can be taken for petrographic examination, which will determine the extent of damage within the overall concrete matrix. Steel coupons can be taken to evaluate changes in material properties.

85 Chapter 2. BRiDGE SySTEM SELECTiON 2.3.1.3.6c Blast The possibility of terrorism against our nation’s bridges is an ever-increasing threat. The risk of blast attack is typically considered very low for most bridges, but major bridges or bridges along major corridors that have high economic or sociopolitical impact can have greater risk. By their nature, bridges are very accessible to vehicles carrying explosive devices traveling either on the bridge or below on crossed road- ways. They are also susceptible to ships or boats carrying explosive devices below. Because bridges vary in type and size, the assessment of blast vulnerability can be very complicated. Until recently, there has been little work done, and methods of analysis and information available concerning the effects of blasts on highway bridges have been scarce. Extensive research is being undertaken by FHWA (Duwadi and Munley 2011) to further understand the behavior and effects of blast loadings on bridge elements. Part of the work of these studies is to develop methods for evaluating risk and risk-mitiga- tion strategies. The most significant research in the area of blast-resistant design guide- lines for highway bridges is being conducted under National Cooperative Highway Research Program (NCHRP) Project 12-72, which has been recently documented in NCHRP Report 645 (Williamson et al. 2010). The response evaluation of reinforced concrete bridge columns was a key part of this investigation. Other recent research by Agrawal and Yi (2008) dealing with blast-load effects on highway bridges developed computer models and showed through simulation analysis that seismic capacities and blast-load effects are strongly correlated. Kiger et al. (2010) focused on the response of posttensioned box girder bridges under blast loads in their report on bridge vulner- ability assessment and mitigation against explosions. Blast loads are considered one of the extreme hazards affecting bridges; even a small amount of explosive can produce severe localized damage to a bridge element. In some cases, this localized damage can potentially progress to global collapse of the structure (Kiger et al. 2010). Various factors affect the potential damage to a bridge from a blast, including • Size and type of explosive charge. Small explosive devices can have varied effects depending on their placement and the size of bridge element, but large truck bombs can be disastrous. The Oklahoma City Federal Building bombing in 1995 is an example of the devastating effect of a large truck bomb. • Proximity to blast (standoff distance). The distance from the blast to a bridge ele- ment is a critical parameter in determining the blast effect. For a given size blast, the effect will reduce significantly with relatively small increases in distance from the blast. • Depending on the size and standoff distance, three blast categories exist: contact, close in, and plane wave (Williamson et al. 2010). Contact blasts are very close and create high-intensity, nonuniform loads in which breaching, or complete loss of material at a section in a bridge element, can occur. In this case, there can be extensive local destruction. A close-in blast is farther away but still results in a

86 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE localized spherical shock wave striking the structure to produce a nonuniform load and impulse-dominated response. A plane wave blast is far enough away to produce essentially planar shock waves and a uniform load on the structure. In this case, the structure will be loaded in a manner that leads to global deformation and will be resisted by the entire structure or a number of combined elements. • Location of blast. Blasts can occur above or below the deck. Above-deck blasts can affect the deck itself and any structural elements above the deck, such as in a through truss, arch, or cable-supported bridge. Blasts below the deck would typi- cally have more effect on pier columns, but a sufficiently large blast can affect the superstructure. Below-deck blasts can also have greater intensity because of the enclosed effect created by the overhead structure; above-deck blasts have more freedom to dissipate without shock wave reflection. • Type and size of bridge element. Members with greater mass, hardness, and flex- ibility have greater blast resistance. • Structural redundancy. Having multiple load paths is a key factor in resisting over- all structural collapse with any type of individual member failure. Multicolumn piers or multigirder superstructures are typically able to redistribute internal forces and provide greater resistance to overall structure collapse. A risk-management approach can be taken for bridges with greater potential of fire or blast damage. These bridges can be identified by reviewing major corridors that would experience the greatest economical and sociopolitical impact if damaged by these extreme-event hazards. Potential mitigation including local protective measures, alternative routes, or accelerated reconstruction strategies can be evaluated for these higher-risk bridges. 2.3.1.3.6d Seismic Events Earthquakes, including those of moderate intensity, are extreme-hazard events that can cause significant damage to bridges, particularly to existing bridges that were de- signed under older codes and have not been retrofitted. The 1971 San Fernando earthquake in California, which resulted in numerous bridge collapses, has often been cited as a watershed event in bridge engineering because it demonstrated the inadequacy of the seismic bridge design practices of the time (Buckle et al. 2006). FHWA became a major sponsor of bridge seismic research shortly afterward, including research on retrofitting existing bridges. Later major California seismic events such as the 1989 Loma Prieta and 1994 Northridge earthquakes, and the 1995 Kobe, Japan, earthquake caused significant bridge damage and collapse, which also led to further research and understanding of bridge seismic behavior (Azizinamini and Ghosh 1997). The observed damage and knowledge gained from these previous events, along with extensive research undertaken since 1971, have led to significantly improved seis- mic bridge design specifications, including the Guide Specifications for LRFD Seismic Bridge Design (AASHTO 2011) and LRFD Bridge Design Specifications (AASHTO

87 Chapter 2. BRiDGE SySTEM SELECTiON 2012); advanced concepts for seismic retrofit, such as Seismic Retrofitting Manual for Highway Structures: Part 1—Bridges (Buckle et al. 2006); and guidance for seismic design of foundations, such as LRFD Seismic Analysis and Design of Transportation Geotechnical Features and Structural Foundations (Kavazanjian et al. 2011). The current approach adopted in the LRFD specifications (AASHTO 2012) is to design new conventional (or ordinary) bridges for a design earthquake, or level of ground motion, that represents the largest motion that can be reasonably expected during the life of the bridge. It implies that ground motions larger than the design earthquake could occur during the life of the bridge, but their likelihood of happening is small. This likelihood is usually expressed as the probability of exceedance, and it may also be described by a return period in years. The specifications call for a design earthquake that has a 7% probability of exceedance in 75 years (a return period of approximately 1,000 years). Bridges designed and detailed under these provisions may suffer damage, but they should have a low probability of collapse. Key principles used for the development of these specifications are that small to moderate earthquakes should be resisted within the elastic range of the structural components without sig- nificant damage, and large earthquakes should not cause collapse of all or part of the bridge, although they may cause significant damage requiring replacement. One of the key considerations in seismic design is repairability of the damage to bridges during moderate seismic events. Oftentimes the so-called minor damage may require complete replacement of the bridge. During the 1995 Hyogoken-Nanbu earth- quake in Kobe, Japan (Bruneau et al. 1996; Chung 1996; Shinozuka et al. 1995; Azizinamini and Ghosh 1997), steel bridges suffered damage to superstructure ele- ments, including inadequate cross-frame detailing leading to lateral bending of the girder webs near the girder ends. The damage resulted in major retrofit activities and the closing of major highways, such as the Hanshine Expressway, which was closed for more than a year. The Kobe experience demonstrated that even minor damage to steel bridges during seismic events can result in damage that could be very difficult to repair. Among the lessons learned is that critical elements of the bridge that are difficult to inspect and repair must be protected from any level of damage and remain elastic dur- ing the entire seismic excitation. Service life design philosophy needs to be considered when following seismic design principles by examining the effects of repair on traffic interruption after small to moderate earthquakes. In particular, the areas with potential to form plastic hinges, as described in this section, must be detailed so that the repair can proceed with little or no interruption to traffic. The major areas of concern are substructure elements, in which most plastic hinges are anticipated. The superstructure elements of the bridge are mainly kept elastic during the entire seismic event. Seismic load behavior is largely unknown; consequently, the design philosophy for buildings and bridges is to work on the behavior of the structure under known condi- tions. Specifically, the design objective is to predefine the damage locations and design them accordingly by providing adequate levels of ductility. In the case of bridges, the preferred damage locations are at the ends of pier columns (formation of plastic hinges). In the direction of traffic, it is preferred to put columns in double curvature,

88 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE as shown in Figure 2.32. This arrangement allows larger portions of the pier column (two plastic hinges versus one for single curvature) to participate in energy dissipation. In the transverse direction, pier columns are usually designed to act in single cur- vature, as shown in Figure 2.33. Under longitudinal excitation, plastic hinges are located near the top and bot- tom of the columns; under transverse excitation, the plastic hinge is located near the bottom of the pier column. The main design feature in the seismic design of bridges is to keep the superstruc- ture elements completely elastic during an entire seismic event. Repairing any elements of the superstructure, even “minor” damage, could be very time consuming and result in a major interruption to traffic. In a capacity design approach, which is routinely used for designing bridges in seismic regions, the elements that should remain elastic are referred to as protected elements. Inelasticity is then forced to take place at pre- defined locations within the substructure. The predefined damage locations are the weak links, or fuses, that control the level of forces to be transmitted to the super- structure elements. Figure 2.32. Deflected shape of a three-span bridge under a longitudinal (along traffic) direction. (a) (b) Figure 2.33. Deflected shape of pier column in (a) longitudinal and (b) transverse directions. Figure 2.32. Deflected shape of a three-span bridge under a longitudinal (along traffic) direction. (a) (b)

89 Chapter 2. BRiDGE SySTEM SELECTiON In the capacity design approach, protected elements are designed for the largest possible force effects they might experience; the design considers the overstrength that may exist because of higher actual material strength than that specified in design. Areas with Seismic Risk Although earthquakes are sometimes considered primarily a California or West Coast problem in the continental United States, data produced by the U.S. Geological Survey (USGS) National Seismic Hazard Mapping indicates that at least 40% of the United States is subject to damaging, ground-shaking levels of seismic risk (Kavazanjian et al. 2011). Since 1996, USGS has developed and updated maps that have depicted areas in the United States with various levels of seismic risk. These maps display earthquake ground motions for various risk levels, including a 2%, 5%, and 10% probability of being exceeded in 50 years. Figure 2.34 shows the USGS seismic hazard map depicting peak ground acceleration levels with a 2% probability of being exceeded in 50 years, or a return period of approximately 2,500 years. Along with areas along the West Coast, these maps show areas of high seismic risk in the central and eastern United States. Performance During Earthquakes Moehle and Eberhard (2000) discuss various causes and types of damage that bridges experience during earthquakes. Key factors that affect the type and severity of bridge damage include the following. Figure 2.34. Seismic hazard map of peak ground acceleration levels with a 2% probability of being exceeded in 50 years. Source: Courtesy U.S. Geological Survey.Source: Courtesy U.S. Geological Survey. Figure 2.34. Seismic hazard map of peak ground acceleration levels with a 2% probability of being exceeded in 50 years.

90 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE • Close proximity to fault rupture. Such proximity results in ground motions having high horizontal and vertical ground accelerations and large-velocity pulses. • Soil conditions. Soft soil sites can significantly amplify ground motion. Soil lique- faction and lateral spreading results in permanent substructure deformation and loss of superstructure support. • Structural configuration. Bridges are most vulnerable that have excessive defor- mation demands in rigid, nonductile elements; complex or nonuniform structural configuration; curved or skewed configuration; or nonredundancy. Major types of damage include • Unseating at joints. Superstructure expansion joints introduce a structural irregu- larity that can have catastrophic consequences. These joints can occur within a span or at substructure supports. Irregular ground shaking can induce superstruc- ture movements that can cause a span to unseat. Unrestrained superstructures can be toppled from their supporting substructures as a result of shaking or differential support movement associated with ground motion. Bridges with short seats are especially vulnerable. Use of restrainers has been effective in minimizing this risk. Figure 2.35a shows a span unseating failure on the Oakland Bay Bridge in San Francisco during the Loma Prieta earthquake in 1989. • Superstructure damage. Superstructures typically have sufficient strength to remain elastic during earthquakes and are unlikely to be the primary cause of collapse of a span. However, certain types of superstructure damage have been observed, including bearing damage, pounding of adjacent units at expansion joints, and transverse bracing or diaphragm damage. • Substructure damage. Substructures typically sustain the most damage, which can be categorized by column failure and abutment damage: – Column failure. The lateral load capacity of a pier is limited by the shear or flexural strength of its columns. For nonductile reinforced concrete columns, shear failure is often the primary mode of failure when the column is subject to large inelastic demands during strong earthquakes. Column failure is often the primary cause of bridge collapse during earthquakes (Moehle and Eberhard 2000). Most damage to columns can be attributed to inadequate detailing, which limits the ability of the column to deform inelastically. In concrete columns, de- tailing inadequacies can produce flexural, shear, splice, or anchorage failures. In steel columns, local buckling has been observed to lead progressively to collapse (Moehle and Eberhard 2000). Figure 2.35b shows a nonductile column shear failure on the Cyprus Street Viaduct in San Francisco that occurred during the Loma Prieta earthquake in 1989. – Abutment damage. Damage to shear keys and wingwalls is often prevalent.

91 Chapter 2. BRiDGE SySTEM SELECTiON 2.3.1.4 Reduced Service Life Resulting from Production or Operation Defects Decisions made for the production of a bridge and activities during its operation can have a significant influence on overall bridge service life. These production and opera- tion influences (shown in the fault tree in Figure 2.36) include decisions made dur- ing the design and detailing of the bridge, quality of fabrication and manufacturing, quality of construction, the level of inspection performed during operations, and the level and quality of maintenance. Each of these categories can be further developed to identify the lowest or basic levels causing deficiency, but these lower levels can vary significantly for each bridge system, component, or element type. The discussion in this section addresses general issues that are common to all. Figure 2.35. 1989 Loma Prieta earthquake damage near San Francisco, California: (a) Oakland Bay Bridge upper roadway span unseating and collapse and (b) Cyprus Street viaduct support column collapse. Source: Courtesy U.S. Geological Survey. (a) (b)

92 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 2.3.1.4.1 Design and Detailing Decisions made during the system selection, design, and detailing phase of a bridge project can significantly affect the service life of the bridge. It is incumbent on design- ers to understand the implications of these decisions in order to help in making ratio- nal choices that will improve service life. Examples of design and detail issues causing reduced service life include • Using bridge systems with deck joints that can ultimately leak and cause service life issues below. See Chapter 8 on jointless bridges; • Providing inadequate drainage systems that allow moisture to remain on bridge decks, leading to deterioration; or improper layout, capacity, or slopes on drainage elements that ultimately lead to clogging and malfunction; • Dealing with moisture trap details on steel bridges that hold water and debris, resulting in coating damage and steel corrosion. See Chapter 6 on corrosion pro- tection of steel bridges; • Using fatigue-prone details; and • Not considering design errors. Effective quality assurance and quality control in design process is necessary to avoid errors. 2.3.1.4.2 Fabrication and Manufacturing Defects in fabrication or manufacturing can lead to reduced service life in steel or concrete bridge elements. Undetected fabrication defects can lead to fatigue damage in steel structures. Production/Operation Defects Design/ Detailing Fabrication/ Manufacturing Construction Inspection Maintenance Figure 2.36. Production or operation defects. re 2.36. Producti n or operation defects.

93 Chapter 2. BRiDGE SySTEM SELECTiON 2.3.1.4.3 Construction Defects or damage in construction can reduce service life in steel or concrete bridge elements. Poor concrete placement and curing practices can have significant effects. See Chapter 3 on materials and Chapter 4 on decks for further discussion of concrete placement. The transportation and erection of both steel and concrete girders can become an issue if not handled properly. As high-performance materials (high-performance steel and high-performance concrete) are increasingly used for bridge construction, girders tend to become longer and their webs slimmer. Transportation of these girders can cause higher stresses or out-of-plane bending, which can result in cracking. Girder stability during erection, particularly in curved steel girders, needs to be carefully addressed. 2.3.1.4.4 Inspection Proper inspection during bridge operation is essential to identify defects and issues early, before more serious conditions develop. 2.3.1.4.5 Maintenance Lack of maintenance or inadequate maintenance can allow deterioration to initiate throughout the bridge system and develop into serious conditions for which the only alternative is costly component or element replacement. Applying the appropriate bridge preservation treatments and activities at the appropriate time can extend bridge service life at a lower lifetime cost (FHWA 2011b). The Bridge Preservation Guide (FHWA 2011b) provides general guidance on the importance and benefits of preventive maintenance as part of an overall bridge preser- vation program. Examples of various cost-effective preventive maintenance activities that can be applied to decks, superstructure, and substructure components are further discussed in Section 2.4.1.3. 2.3.2 factors Affecting Service Life of Deck Component The deck component includes several elements as described in Section 2.2.2. Bridge-deck service life and the factors affecting service life are described in detail in Chapter 4. Concrete bridge decks are particularly affected by thermal and coastal environments in which chloride penetration can cause reinforcing steel corrosion leading to concrete cracking and spalling. This cracking can cause the concrete sur- rounding the steel reinforcement to reach the corrosion threshold limit in environ- ments in which the top of the bridge deck is exposed to chlorides, such as deicing salts. Other concrete deck issues include wear and freeze–thaw damage. Bridge expansion devices, commonly referred to as bridge joints, are discussed in Chapter 9. Drainage systems are most affected by lack of maintenance that causes clogging and malfunction.

94 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Bridge railings are affected by wet chloride environments that cause corrosion of reinforcing steel and concrete cracking and spalling in concrete railings. This condi- tion is exacerbated at cold joints between the concrete barrier and the top of the slab, where salt moisture can easily penetrate and cause reinforcing corrosion. The same environments cause corrosion of steel railings. 2.3.3 factors Affecting Service Life of Superstructure Component 2.3.3.1 Steel Superstructures The principal causes of steel element deterioration in steel bridge systems are fatigue and fracture (addressed in additional detail in Chapter 7) and corrosion (discussed in Chapter 6). 2.3.3.1.1 Load-Induced Deficiency: Fatigue and Fracture Early-welded steel structures have a history of cracking at certain types of weld details as a result of load-induced and distortion-induced fatigue. Cracking at I-beam cover plate terminations or at other longitudinal weld terminations in tension zones has been particularly evident. Cracking in girder webs as a result of out-of-plane bending within stiffener web gap regions next to cross-frame attachments also became a common problem. Subsequently, extensive research and laboratory testing have provided an understanding of fatigue behavior, and different weld detail types were found to have varying levels of fatigue susceptibility. Newer design provisions and recommended details were developed that provide solutions for both load-induced and distortion- induced fatigue that will achieve desired service life. Steel bridges can fail by fracture, which is the rapid, unstable propagation of a larger flaw, most likely the result of fatigue. Fatigue crack initiation is independent of steel type and strength, but possible brittle fracture is influenced by steel toughness, among other variables. Early steels were more susceptible to brittle fracture, but in recent years, new high-performance steels—HPS 50W, 70W, and 100W—have been developed with very high toughness characteristics. Although somewhat more costly than conventional-grade steels, high-performance steels are now encouraged when applicable, particularly in nonredundant or fracture-critical applications. Use of high- performance steel allows time for any fatigue cracks that may have developed to be found during regular bridge safety inspections, before fracture can occur. Fatigue should not be an issue in new steel bridges designed in accordance with the latest LRFD specifications. Extensive research in recent years has identified causes and solutions for fatigue- and fracture-related problems. When using proper details and fabrication methods, both load-induced and distortion-induced fatigue problems should not be an issue in achieving desired service life.

95 Chapter 2. BRiDGE SySTEM SELECTiON 2.3.3.1.2 Deficiency Caused by Natural or Man-Made Hazards: Corrosion Corrosion, the result of exposure to oxygen and moisture, is a fundamental limitation of steel as a construction material. The process is greatly accelerated in the presence of chloride ions from roadway deicing salt or salt spray in a marine environment. Deck drainage with deicing salt leaking through open deck joints is a leading cause of steel element corrosion in bridges. Corrosion control should be designed into the overall steel bridge system. Use of systems that eliminate or minimize deck joints will have a significant effect in reducing corrosion. Details that serve to protect and keep the steel dry should be included in the design. Among these are bridge system solutions that eliminate deck joints, thus preventing salt-contaminated drainage from reaching steel elements below. Salt marine environments or locations subject to deicing salt spray from below also create harsh environmental conditions conducive to corrosion; thorough clean- ing and zinc-rich primer coating systems can provide long-term protection. However, requirements for related long-term coating maintenance cannot be overlooked and must also be designed into an overall corrosion protection plan. To achieve long-term bridge durability, a corrosion-resistance plan must be a design requirement for every new or rehabilitated steel structure. This plan should include the use of best painting practices and a maintenance plan that addresses paint- ing priorities and timetables. Best painting practices now include paint systems that contain metallic zinc as the corrosion-resistant pigment. Zinc coatings provide galvanic protection to the steel because zinc (the more noble metal) will oxidize (corrode) in preference to the steel. To protect the zinc-coating layer from oxidation, additional coating layers are applied over the zinc-rich primer. Many studies have demonstrated the value of zinc coatings as a steel protection system. In addition to zinc-rich paint, these zinc coatings also include galvanizing and metalizing. Their use should be considered carefully as part of the best plan for achiev- ing extended service life. Weathering steel has also found widespread use in steel bridges in both unpainted and painted applications. Weathering steel is corrosion resistant in some circumstances, but it is adversely affected by continual drainage and roadway salts, particularly below joints. Typically, special coatings are applied in these locations when using weathering steel. In addition to weathering steel, a new structural stainless steel for bridges, ASTM A1010, has been developed for use in severe corrosive environments. 2.3.3.2 Concrete Superstructures 2.3.3.2.1 General Deficiencies There are numerous causes of deterioration in concrete superstructure elements. Chap- ter 3 on materials discusses the various factors influencing concrete durability. Typi- cally, the deficiencies are caused by three main factors that were described in the fault tree analysis in Section 2.3.1:

96 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE • Load-induced influences – Traffic causing vibration, impact, or wear; and – Restrained thermal movement causing internal stress and cracking. • Natural or man-made hazards – Environmental influences, including effects from moisture and freezing and thawing, and reinforcing corrosion caused by chloride exposure; and – Chemical influences, including exposure sulfates, carbon dioxide, alkalis, and various acids. • Defects in production or operation, primarily defective placement, curing, and maintenance. The degree and severity of concrete deterioration depends on the level of load and environmental influences to which the bridge is subjected. The durability of concrete exposed to these influences is highly dependent on design practice, materials, and their proportioning and workmanship during construc- tion. Although all of these influences are important, the principal deterioration of concrete elements is the corrosion of steel reinforcement, which results in severe crack- ing, spalling, and delamination of the surrounding concrete. Cracking of the concrete often compromises the protection provided by the depassivated zone around the steel reinforcement. Concrete superstructures in thermal or marine environments are not exposed to the same concentration of chlorides as the top of a bridge deck that may be directly in contact with deicing salts. They are, however, susceptible to the same type of corro- sion, only at a slower rate through the same mechanism of failure. 2.3.3.2.2 Adjacent Box Beam Deterioration Adjacent box beam bridges have been used for many years and have generally per- formed well. However, a commonly reported problem has been with cracking in the longitudinal grouted joints between adjacent beams, which results in reflective cracks in the top wearing surface (Hanna et al. 2009). This has been reported in both compos- ite and noncomposite bridges (Hanna et al. 2009; NCHRP 2009; Russell 2011). The longitudinal cracking can significantly affect the overall durability and structural be- havior of these types of bridges. Open surface cracking allows penetration of roadway drainage, often with de-icing chemicals, which can penetrate the sides and bottoms of beams and cause corrosion of both nonprestressed and prestressed reinforcement and concrete freeze–thaw drainage (see Figure 2.37). In addition, shear key cracking can adversely affect the load distribution among the beams, resulting in loaded beams carrying a greater proportion of load than what was assumed in the design (Hanna et al. 2009). In some cases, these adverse conditions have led to significant deterioration and actual beam failure, requiring replacement of the entire bridge. In December of 2005, a fascia beam on the Lakeview Drive Bridge over I-70 in Washington, Pennsylvania,

97 Chapter 2. BRiDGE SySTEM SELECTiON failed near midspan and fell to the highway below, as shown in Figure 2.38. Inspection immediately afterwards revealed heavy concrete spalling and corrosion of strands on the failed beam bottom flange. The inspection also found strand corrosion on other box beams, and the bridge was permanently closed and ultimately replaced (Hanna et al. 2009; Naito et al. 2006). (a) (b) Source: Courtesy HDR Engineering, Inc. Figure 2.37. Longitudinal cracking in adjacent box beam bridge: (a) underside of beam cracking and (b) top of roadway reflective cracking . Figure 2.37. Longitudinal cracking in adjacent box beam bridge: (a) underside of beam cracking and (b) top of roadway reflective cracking . Source: Courtesy HDR Engineering, I c. Figure 2.38. Failure of adjacent box beam bridge in Pennsylvania. Source: Courtesy Pennsylvania DOT.

98 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 2.3.4 factors Affecting Service Life of Substructure Component The numerous causes of substructure deterioration can be categorized in three areas: • Improper detailing and improper consideration of appropriate forces resulting from applied mean recurrence-level event forces, such as scour, vessel collision, and earthquake; • Deterioration caused by corrosion and section loss, primarily from chloride intru- sion; and • Seized bearings and unintended movement restraint. 2.3.4.1 Mean Recurrence-Level Event Forces As bridge service life increases, bridges are subjected to environmental conditions for a longer period of time. Many of these conditions are accounted for in design by the use of traditional recurrence values of extreme environmental conditions such as those for hydraulic stages, wind loads, and seismic events. Design for vessel impact is treated similarly. The increased service life from a 50-year (pre-LRFD) to a 75-year (LRFD) to a 100-year service life increases the statistical probability of exceeding the design recurrence-level event. The importance of considering these events has been exemplified by numerous bridge incidents, including, but not limited to • The 1987 collapse of the I-90 Bridge over Schoharie Creek in New York State, which emphasized the importance of providing adequate hydraulic openings and flow characteristics under bridges. Undermining of embankments, such as shown in Figure 2.39, and undermining of pier foundations as a result of scour remain primary causes of bridge failures around the nation. Source: Courtesy Atkins North America, Inc. Figure 2.39. Scour undermining of bridge abutment. Figure 2.39. Scour undermining of bridge abutment. Source: Courtesy Atkins North America, Inc.

99 Chapter 2. BRiDGE SySTEM SELECTiON • Multiple catastrophic vessel and bridge accidents around the world from the 1960s to the mid 1980s (Knott and Larsen 1990), including the 1964 and 1974 collapses of the Pontchartrain Bridge in Louisiana and the 1980 collapse of the Sunshine Skyway Bridge in Florida. • Impact forces from motor vehicles, similar to vessel collision, and subsequent fires that have resulted in structural damage to substructure units. Susceptibility to damage from these aberrant vehicle impacts underscores the importance of pier protection. • The 1971 San Fernando earthquake, and subsequent major earthquakes identified in Section 2.3.1.3.6d, which caused catastrophic damage to numerous bridges and stimulated research relating to bridge performance during seismic events, such as the Standard Specifications for Seismic Design of Highway Bridges (AASHTO 1983). 2.3.4.2 Substructure Deterioration Caused by Material Deterioration, Corrosion, and Section Loss Corrosion deterioration in substructure elements stems from numerous causes, including • Chloride intrusion from leakage of expansion joints and bridge drainage where deicing salts are used to remove snow and ice from bridge decks; • Chloride intrusion from direct salt splash from traffic traveling on roadways below the bridge where deicing salts are used to remove snow and ice from the pavement; • Chloride intrusion found in marine and brackish water environments affecting exposed elements (such as those shown in Figure 2.40); and • Corrosion from concrete cracking induced by ASR and other concrete quality issues. Many of the issues that affect the durability of the substructure are similar to the issues affecting the bridge in general. Leakage of expansion joints and bridge drainage create a major problem for the superstructure and substructure below the leak. Strate- gies to address the causes and possible relief of these leaks are addressed in Chapter 9 on joints. Strategies to address concrete quality issues, such as ASR and others, are addressed in Chapter 3. Issues related to the substructure in a marine environment or in grade separations in which deicing salts can be splashed on supporting members are addressed in this chapter. Degradation of concrete and steel structures in aggressive corrosive environments, such as the splash zone in a marine environment, has historically led to a reduction in service life of numerous structures. The areas particularly susceptible to chloride intru- sion are the splash zone, areas of poor concrete consolidation, spalls, and pile splices. The degradation of piles and other deep foundation elements in marine environ- ments has spawned an enormous concrete and steel protection and repair industry that has developed numerous products dealing with the preservation of these deteriorating structural elements (Heffron 2007). Many of these products are relatively new, and

100 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE their long-term effectiveness and expected life are not verifiable with historic data. Some of the more promising techniques include • Cathodic protection with embedded sacrificial anodes; • Pile jacketing, as shown in Figure 2.41; • Metalized coatings; • Crystalline admixtures for crack sealing; • Repassivation through the removal of chloride ions; and • Various combinations of these techniques. 2.3.4.3 Potential Effect of Climate Change and Service Life of Bridges in Coastal Areas The greatest potential impact of climate change on the service life of bridges in coastal areas is the potential rise in sea level and the associated increased risk of flooding and erosion. In general, limate-related impacts are already being observed in the United (a) (b) Source: Courtesy Atkins North America, Inc. Figure 2.40. Marine pile degradation: (a) jacket failure on concrete pile and (b) corroded steel pile. Figure 2.40. Marine pile degradation: (a) jacket failure on concrete pile and (b) corroded steel pile. Source: Courtesy Atkins North America, Inc.

101 Chapter 2. BRiDGE SySTEM SELECTiON States and its coastal waters, and empirical projections suggest that many of these im- pacts will grow in severity in the future (U.S. Global Change Research Program 2009). Evidence shows that sea level has increased along most of the U.S. coast over the past 50 years. In some areas along the Atlantic and Gulf coasts, these increases have been greater than 8 in. (America’s Climate Choices: Panel on Adapting to the Impacts of Climate Change 2010). Predicting the potential future sea level rise, however, has a great deal of uncer- tainty. Vermeer and Rahmstorf (2009) developed a model to determine the projected sea level rise from assumed levels of greenhouse gas emissions and subsequent tempera- ture increases, and they evaluated three scenarios of assumed greenhouse gas emissions as compiled by the Intergovernmental Panel on Climate Change (IPCC). Figure 2.42 shows the projected sea level rise through 2100 that was developed for each of these scenarios (marked as B1, A2, and A1F1 on the right side of the graph). The shaded areas represent the potential variation in each projection. The outer light-gray area represents additional uncertainty in the projections resulting from uncertainty in the fit between temperature rise and sea level rise. From this graph, the overall range in poten- tial sea level rise (considering uncertainties) is from about 2 ft to about 6 ft by 2100. All these projections are considerably larger than earlier sea level rise estimates for 2100 provided in a previous IPCC report (AR4), which is shown by the vertical bars on the right side of the graph. This earlier report, based on data from or before 2005, did not account for potential changes in ice sheet dynamics and resulted in predictions ranging from about 0.6 ft to 2 ft by 2100. Also the first part of the graph shows the observed annual global sea level rise over the past half-century relative to 1990 (Vermeer and Rahmstorf 2009; America’s Climate Choices: Panel on Advancing the Science of Cli- mate Change 2010). Figure 2.41. Pile restoration with pile encapsulation and epoxy grout fill. Source: Courtesy Atkins North America, Inc. Source: Courtesy Atkins North America, Inc. Figure 2.41. Pile restoration with pile encapsulation and epoxy grout fill.

102 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Although there is still debate on the level of severity that will be experienced over the next decades because of climate change, it is important to consider this factor when planning for new bridges or retrofitting existing bridges located in areas that could be affected. Currently there are no organized, formal plans for considering the possible effects of climate changes on existing and new bridges located in coastal areas. However, the major impact could be on the substructure and splash zone of the columns located in the water. 2.3.4.4 Seized Bearings and Unintended Movement Restraint The structural design of substructures is in part based on a distribution of longitudinal and transverse forces associated with the allowable movement of the superstructure. Fixed bearings provide an anchor that is intended to restrict “walking” of the super- structure that can result from shrinkage and cycling of expansion and contraction. The bearings are usually located longitudinally near the point of zero movement of a sup- ported multispan superstructure unit. Care must be exercised in locating the so-called point of zero movement, especially in the case of curved girder bridges. The existence of such a point could be viewed as an assumption more than a reality. Chapter 8 on jointless bridges provides a detailed discussion of the point of zero movement for curved girder bridges. The point of zero movement is more meaningful in instances of straight bridges with zero skew. Existence of skew or curvature complicates the deter- mination of the point of zero movement. These fixed bearings, used in combination with bearings designed to allow the superstructure either to move or slide over the top of the substructure, reduce restraining forces that would otherwise be required to resist the movement. Improper function or seizing of the bearings results in unintended movement restraint that can raise the force resisted by the substructure well above Figure 2.42. Projection of sea level rise from 1990 to 2100. Sources: Vermeer and Rahmstorf 2009 and America’s Climate Choices: Panel on Advancing the Science of Climate Change 2010.

103 Chapter 2. BRiDGE SySTEM SELECTiON the intended design. This unintended restraint can cause unanticipated cracking with greater potential for corrosion. Proper bearing performance, which is essential to sub- structure durability, is addressed in Chapter 10. 2.4 oPtionS For enhAncing Service LiFe This section describes options for enhancing service life and addresses the factors iden- tified in Section 2.3. 2.4.1 System-Related options 2.4.1.1 Functional Options Sometimes bridges are replaced because of functional issues well before their full potential service life is achieved. The following considerations should be incorporated into a bridge system when there is a probability that future bridge widening or crossed- roadway widening may be necessary: • Use bridge system types that can be widened, particularly superstructures. Multi- girder superstructures typically lend themselves to widening, but CIP concrete structures provide additional challenges for this modification. • Consider longer spans when crossing roadways that have the potential for widening. • Consider additional vertical clearance when setting limits for bridges that have the potential for future widening. Bridge widening along a deck cross slope can infringe on minimum vertical clearances if additional clearance is not provided at the beginning. 2.4.1.2 System Configuration Options How the bridge system is configured will significantly affect the service life. Leaking deck joints have been identified as one of the leading causes of system deterioration. The considerations discussed in this section should be incorporated into the system selection to avoid or mitigate this risk (see Chapter 8 on jointless bridges). 2.4.1.2.1 Integral Abutments Consider integral abutments to eliminate joints at abutments. Fully integral abutments eliminate deck joints and bearings, and semi-integral abutments eliminate deck joints. In addition to their many other advantages with respect to enhanced service life, joint- less integral abutment bridges provide lower initial cost. This type of bridge is increas- ing in popularity, and its use is encouraged when appropriate. Chapter 8 on jointless bridges provides step-by-step design provisions for jointless integral abutment bridges. Unlike the current practice, there is no need to impose arbitrary limits on maximum length of bridges.

104 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 2.4.1.2.2 Maximum Length Limits for Continuity Using the procedure specified in Chapter 8, establish the maximum lengths for con- tinuity to minimize the number of joints in long, multispan viaducts. Length in the following types of structures should be considered: • Superstructure with integral abutments and no joints. Use design provisions stated in Chapter 8 on jointless bridges to establish the maximum length for integral abutments. • Long continuous superstructure with joints only at abutments. Consider the maxi- mum length for the structure type that can accommodate joints only at abutments without any intermediate joints. • Multiple continuous units with interior joints (viaduct construction). Consider the maximum length for unit layout between interior joints to minimize the number of joints. • Structures requiring deck joints. Consider joint systems that are more leak resis- tant. See Chapter 9 on expansion joints. 2.4.1.2.3 Continuity over Piers Various systems for providing continuity over piers that eliminate deck joints should be evaluated: • Fully continuous. These systems are continuous for dead and live loads and are suitable for all span lengths, but they are typically more economical for longer spans, on which the benefit of dead load continuity is better realized. • Simple for dead load, continuous for live load. These systems are becoming very popular in the 150-ft span range because of the ease and speed of their construc- tion, but girders must carry all dead load in positive bending. • Link slab. Link slabs are economical and popular when the deck is made continu- ous over intermediate supports while the beams remain simple span without any continuity. This concept is further discussed in Chapter 8. • Integral pier caps. The use of integral pier caps eliminates joints and bearings while lowering the roadway profile, which can add further economic benefit. However, it requires special details for the integral connection, and the system interaction between superstructure and substructure must be considered. 2.4.1.2.4 Fixed-Pier and Expansion Pier Layout Proper layout of fixed- and expansion-pier locations can help balance loads to piers while minimizing superstructure thermal movements. Considerations include • Traditional layout (single fixed pier, others expansion). Providing a single fixed pier near the center of the bridge focuses longitudinal loads to one location, which

105 Chapter 2. BRiDGE SySTEM SELECTiON is usually acceptable for minimum height bridges and balances thermal move- ments at adjacent piers and abutments as much as practicable. • Multiple pier fixity. This method offers a benefit in taller pier situations, in which longitudinal loads can be distributed to additional piers. In addition, tall pier flex- ibility minimizes temperature loads that develop. The relative stiffness of multiple fixed piers must be considered in distributing longitudinal loads and in determin- ing temperature forces. • Integral piers. The use of integral piers creates a fixed-pier condition and has the benefit of eliminating joints and bearings and lowering the roadway profile. De- pending on the type of detail, it can also provide longitudinal frame action in resisting longitudinal loads. • Orientation of expansion bearings on curved and skewed alignments. 2.4.1.3 Maintenance Considerations The Bridge Preservation Guide (FHWA 2011b) provides general guidance on the im- portance and benefits of preventive maintenance as part of an overall bridge preser- vation program. Examples of various cost-effective preventive maintenance activities that can be applied to decks, superstructure, and substructure components are pre- sented, including • Sealing or replacing leaking deck joints before deterioration can begin on elements below; • General bridge cleaning, including decks, joints, drainage systems, bearings, tops of piers, and all elements below-deck joints; • Placing deck overlays on aging decks; • Installing cathodic protection or electromechanical chloride extraction; • Applying concrete sealants or coatings; • Spot painting or zone painting steel elements; • Retrofitting fatigue-prone details; • Lubricating bearings; • Jacketing concrete piles in marine environments; • Installing scour countermeasures; and • Removing large debris from stream channels. 2.4.1.4 Access Considerations Proper accessibility to all components and elements below deck for inspection and future maintenance is essential for achieving long-term service life. Accessibility and maintainability considerations must be included as part of the overall bridge system configuration.

106 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 2.4.2 Deck Component options For options related to deck components, see Chapter 4. For options related to joint elements, see Chapter 9 on expansion joints. 2.4.3 Superstructure Component options 2.4.3.1 Steel Superstructures For options related to controlling fatigue, see Chapter 7 on fatigue and fracture. For options related to corrosion resistance, including paint systems, galvanizing, metal- izing, and use of corrosion-resistant steels, see Chapter 6 on corrosion protection of steel bridges. 2.4.3.2 Concrete Superstructures 2.4.3.2.1 General Strategies Several strategies have been developed to address the durability of concrete systems, subsystems, and components. These strategies, which are fully described in Chapter 3 on materials, include the following: • Proportioning concrete to provide low permeability and low cracking potential; • Using noncorrosive materials, such as stainless steel, for reinforcement, and pro- tective coatings, such as epoxy coating; • Prestressing or posttensioning elements to eliminate cracking; • Applying other solutions, such as cathodic protection and electrochemical chloride extraction; and • Using various combinations of these strategies. The use and application of these strategies is highly dependent on the environment to which the concrete systems will be exposed. A single strategy that fits all conditions within the United States does not exist. These strategies must be reviewed for applica- bility by each governing agency. 2.4.3.2.2 Solutions for Adjacent Concrete Box Beams Adjacent box beam bridges have experienced service life issues in recent years that have resulted in failures, as illustrated in Section 2.3.3.2.2. Hanna et al. (2009) reported on a survey conducted by a Precast/Prestressed Concrete Institute sub committee that inves- tigated current practices in the design and construction of adjacent box girder bridges in the United States and Canada (PCI 2008). NCHRP (2009) reported on a synthesis of highway practice of using these types of bridges and presented more results of this survey. Russell (2011) summarized the NCHRP report in a PCI State of the Prac- tice Report. The survey found that 29 states and three provinces are currently using adjacent box girder bridges. Most of these transportation agencies have experienced

107 Chapter 2. BRiDGE SySTEM SELECTiON premature reflective cracks in the wearing surface on the bridges built in the late 1980s and early 1990s. These agencies have emphasized the importance of eliminating these cracks because they allow the penetration of water and deicing chemicals, which leads to the corrosion of reinforcing steel in the sides and bottoms of concrete boxes. The following are examples of the preventive actions that the states and provinces have recommended on the basis of lessons learned in the last two decades, as reported by Hanna et al. (2009) and repeated by NCHRP (2009): • Use of CIP deck on top of the adjacent boxes to prevent water leakage and to uni- formly distribute the loads on adjacent boxes; • Use of nonshrink grout or appropriate sealant instead of the conventional sand– cement mortar in the shear keys, in addition to blast cleaning of key surfaces be- fore grouting. A few states have also recommended the use of full-depth shear keys because of their superior performance over traditional top flange keys; • Use of transverse posttensioning to improve load distribution and minimize dif- ferential deflections among adjacent boxes. Adequate posttensioning force should be applied after grouting the shear keys to minimize the tensile stresses that cause longitudinal cracking at these joints; • Use of end diaphragms to ensure proper seating of adjacent boxes and intermedi- ate diaphragms to provide the necessary stiffness in the transverse direction; • Use of wide bearing pads under the middle of the box (centered) to eliminate the rocking of the box while grouting the shear keys. NCHRP also recommended con- sideration of a three-point bearing system (2009). Using sloped bearing seats that match the surface cross slope is also recommended; • Use of adequate concrete cover and corrosion inhibitor admixtures in the concrete mix to resist the chloride-induced corrosion of reinforcing steel; and • Eliminating the use of welded connections between adjacent boxes and avoiding dimensional tolerances that result in inadequate sealing of the shear keys. In addition to the major recommendations listed above from Hanna et al. (2009), NCHRP (2009) listed a number of other design, fabrication, and construction recom- mendations that would provide increased overall performance based on the PCI survey and review of other literature. Hanna et al. (2011) reported on a study that developed a new transverse connec- tion system for adjacent box girder bridges that would perform better than methods currently used. The study evaluated the potential of eliminating internal box girder diaphragms combined with the use of non-posttensioned transverse connection sys- tems that are designed to transfer shear and moment between adjacent box girders. The study discussed that although the use of transverse posttensioned diaphragms to connect adjacent box girders is an effective and practical solution in many cases, it has some disadvantages. Posttensioning of skewed bridges is difficult and may have to be staggered and done in stages. Staged construction leads to a significant increase in construction cost and duration because of the variation in diaphragm location, the

108 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE large number of posttensioning operations, and the excessive traffic control required for replacement projects. Moreover, posttensioned diaphragms depend on the shear keys to achieve the desired continuity. Shear keys need to be properly cleaned, sand- blasted, sealed, and grouted, which adds complexity to the system and makes it sus- ceptible to cracking and leakage. Hanna et al. developed a revised approach that eliminates diaphragms and uses top and bottom transverse ties. Two systems were developed and referred to as the wide-joint system and the narrow joint system. The wide-joint system uses full-depth and full-length shear keys, 5 in. wide, between boxes that are then filled with self- consolidated concrete. The narrow joint system also uses full-depth and full-length shear keys, but the keys are narrower and filled with nonshrink grout. Both sys- tems use transverse high-strength rods through the top and bottom flanges to pro- vide transverse connection. A more detailed description of the narrow joint system is included below. Figures 2.43 and 2.44 illustrate the narrow joint system concept. It uses Grade 75 threaded rods with coupling nuts for connection over the joints between boxes, and the rods are end-anchored by washers and nuts. The rods are spaced every 8 ft and pro- vide continuous connection that transfers shear and moment between adjacent boxes more efficiently than do middepth transverse ties at discrete diaphragm locations. A slight modification is made to the standard box cross section by developing full-length horizontal and full-depth vertical shear keys, as shown in Figure 2.43. The boxes are fabricated with a plastic duct at the top and bottom flanges to create openings for the threaded rods, as shown in Figure 2.44. The bottom duct is inserted between the two layers of prestressing strands, and the top plastic duct is located 3 in. from the top surface to provide adequate concrete cover. Vertical vents are provided at one side from each box to allow the air to escape while grouting the ducts (Hanna et al. 2011). Figure 2.43. Narrow joint system box dimensions. Source: Hanna et al. 2011.

109 Chapter 2. BRiDGE SySTEM SELECTiON Hansen et al. (2012) developed a modified version of the narrow joint system described above, which allows for posttensioning of the transverse rods while still eliminating diaphragms. The rods are placed in ducts located inside the box girder voids, adjacent to the inside surfaces of the top and bottom flanges, and are not through the center of the flanges as shown for the narrow joint system above. It was determined that posttensioning increased the capacity and efficiency of the section because joints are placed under compression and are less likely to experience reflective cracking and leakage. 2.4.3.3 Bearing Options See Chapter 10 for a complete discussion of bearing options. Use of steel-reinforced elastomeric bearings is considered the best option for achieving long-term service life. 2.4.4 Substructure Component options Substructure deficiencies are primarily the result of operational and natural or man- made hazards, including the following. h Plastic du ct 4″ × 4″ × ½″ Washer 0.5″ Vent 0.5″ Vent Coupling nut 5″ long Nut for 0.75″ diameter rod 0.75″ diameter × 3′ × 10″ Threaded Rods Figure 2.44. Narrow joint system connection details. Source: Hanna et al. 2011.

110 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE • Element material deterioration due to thermal, coastal, or chemical climate and reactive materials. Strategies for mitigating these effects are discussed in – Chapter 3 on materials, – Chapter 5 on corrosion protection of reinforced concrete, and – Chapter 6 on corrosion protection of steel bridges. • Hydraulic action, which includes flood and storm surge and scour. • Vessel collision. Operational issues include frozen or locked expansion bearings, which are discussed in Chapter 10 on bridge bearings. 2.4.4.1 Hydraulic Action 2.4.4.1.1 Flood and Storm Surge Following hurricanes Ivan in 2004 and Rita in 2005, which damaged numerous bridges along the Gulf Coast, FHWA and 10 states sponsored a study that culminated in the Guide Specifications for Bridges Vulnerable to Coastal Storms (AASHTO 2008). This guide specification recommends that when practical, a vertical clearance of at least 1 ft above the 100-year design wave crest elevation, which includes the design storm water elevation, should be provided. In response to the large uncertainty in the basic wave and surge data needed to determine the wave crest elevation, the AASHTO study further recommends additional freeboard. If this vertical clearance is not possible, the bridge should be analyzed and designed to resist storm wave forces, and other wave force mitigation measures should be implemented, such as venting to reduce buoyancy forces. The Florida DOT issued Temporary Design Bulletin C09-08 (Florida DOT 2009), which required the implementation of the AASHTO Guide Specifications for Bridges Vulnerable to Coastal Storms. In the Florida DOT bulletin, the importance or critical- ity of bridges, as described in Table 2.1, is a factor in evaluating the risk of damage and potential consequences. The Florida DOT further recommends that for all bridges subject to coastal storms, simple and inexpensive measures that enhance a structure’s capacity to resist storm forces should be implemented. For example, placing vents in all diaphragms and tABLE 2.1. bridge imPortAnce LeveL Importance Level of Design Extremely critical Strength limit state for little or no damage Critical Extreme-event level state for repairable damage Noncritical Evaluation of wave forces not required

111 Chapter 2. BRiDGE SySTEM SELECTiON venting all bays will reduce the effects of buoyancy forces on a structure. Anchoring the superstructure to the substructure to reduce or prevent damage from storm surges should also be considered. 2.4.4.1.2 Scour The LRFD specifications require that scour at bridge foundations be designed for 100-year flood events or from an overtopping flood of a lesser recurrence interval. In addition, the bridge foundations are to be checked for stability for a 500-year flood event or from an overtopping flood of lesser recurrence level. FHWA’s Hydraulic Engineering Circular No. 18: Evaluating Scour at Bridges (Arneson et al. 2012) provides guidelines for designing bridges to resist scour and improving the estimation of scour at bridges. Riprap remains the countermeasure most commonly used to prevent scour at bridge abutments. A number of physical additions to the abutments of bridges can help prevent scour, such as the installation of gabions and stone pitching upstream from the foundation. The addition of sheet piles or interlocking prefabricated concrete blocks can also offer protection. These countermeasures do not change the scouring flow, and they are temporary as the components are known to move or to be washed away in certain flood events. FHWA recommends design criteria and countermeasures in Arneson et al. (2012) and in Bridge Scour and Stream Instability Countermeasures: Experience, Selection, and Design Guidance (Lagasse et al. 2009), such as avoiding unfavorable flow pat- terns; streamlining the abutments; designing pier foundations resistant to scour with- out depending on the use of riprap; and other countermeasures that may be available. To reduce the potential for scour, the bottom of spread footings should be placed below the scour design depth, and piles or drilled shafts should be designed by assum- ing all material above the maximum scour depth is unavailable for load resistance. Floods also place extreme lateral loads on piers and bents, which should be consid- ered in design for bridges in locations with a high risk of flooding. In these cases, the presence of soil and any corresponding load or resistance should only be considered below the minimum scour elevation. 2.5 StrAtegy SeLection This section outlines an approach for selecting the most appropriate bridge systems to accommodate operational requirements and site conditions, while also achieving the desired target design service life. The process combines the requirements for selecting bridge systems on the basis of operational needs and initial construction cost with requirements for service life and life-cycle cost. The approach presented in this section must be developed in conjunction with strategies presented in subsequent chapters, which address materials and specific components and elements, such as bridge deck, joints, or bearings, in additional detail.

112 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Providing bridge systems with enhanced service life requires a complete under- standing of the potential deterioration mechanisms, or factors affecting service life. These mechanisms, described in Section 2.3, are associated with load-induced con- ditions, local environmental hazards, production-created deficiencies, and lack of effective operational procedures. The avoidance or mitigation of these deterioration mechanisms through the appropriate selection of enhancement techniques is described in Section 2.4. The overall system selection process involves a detailed evaluation of these mechanisms as they would affect each major bridge component, subsystem, and element and identification of a group of individual strategies that together define an optimum bridge system configuration. This integrated approach of combining opera- tional and service life requirements will result in the optimum bridge system with the greatest potential for enhanced service life. 2.5.1 Service Life Design methodology Chapter 1 provides information concerning design methodologies for service life. 2.5.2 System Selection Process outline A process for selecting the optimum bridge system is shown in flowcharts in Figures 2.45 through 2.47. The outlined process involves four major steps, which are de- scribed in the various numbered flowchart blocks: 1. Block 1: Identifying demand, which includes requirements that the bridge must satisfy; 2. Block 2: Identifying feasible bridge system alternatives that satisfy requirements; 3. Blocks 3 through 12: Evaluating alternatives for service life; and 4. Block 13: Comparing and selecting the optimum alternative. Each of these steps and related flowchart blocks are described in Section 2.5.3. For ease of reading, the final flowchart step in Figures 2.45 and 2.46 is repeated at the beginning of Figures 2.46 and 2.47, respectively. Further examples corresponding to the blocks are provided in Table 2.2. 2.5.3 System Selection Process Description The following is a discussion of the steps in the selection process relating to the flow- chart blocks in Figures 2.45 through 2.47. Block 1. The first step is to identify demand parameters that the bridge must sat- isfy, including a. Operational requirements. These relate to issues such as the type of corridor, traffic and truck percentages, vehicle loads, and mixed-use requirements. This information establishes requirements for capacity, number of lanes, and other operational issues.

113 Chapter 2. BRiDGE SySTEM SELECTiON b. Site requirements. These typically relate to issues that affect bridge layout, in- cluding features crossed, geometrics involving curvature or skew, geotechnical data, and other constraints. c. Service life requirements. Based on the type of corridor and traffic demand, a judgment is made, usually by the bridge owner, as to the importance of the bridge and the target service life for which it should be designed. d. Future considerations. Based on the type of corridor, an evaluation is made regarding the potential for future needs. This might include the probability of future traffic demand that would require bridge widening or the potential of having to widen any crossed roadways that might affect span layout. Overall, these requirements relate to how the bridge must function and how long it should last. They may also include limitations on how a bridge might be constructed, how much it should cost, or how it might look in a given setting. Further examples of these items are listed in Table 2.2. 2. Identify feasible system alternatives that satisfy design provisions of LRFD specifications and satisfy operationaland site requirements. 1. Identify local operational and site requirements that affect bridge layout and service life requirements. 3a. Follow each branch of fault tree. 3b. Identify individual factors affecting service life. Bridge System Selection Process 3. For system alternative being considered, evaluate all components, subsystems, and elements against factors affecting service life by conducting fault tree analysis. Go to the next alternative. From Block 12 Figure 2.45. Blocks 1 to 3b of integrated system selection and design process.

114 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Does factor apply? Yes No 4. Identify consequence and determine appropriate strategy for avoidance or mitigation. 5. Collect all strategies and incorporate into bridge system being considered to develop draft configuration. Are all factors considered? Yes NoGo to the next factor. 6. Determine if the draft configuration satisfies requirements and has compatibility between all components, subsystems, and elements. Is configuration okay? Yes No 7. Make appropriate modifications in components, subsystems, or elements for compatibility. 8. Develop final configuration of system alternative being considered. 3b. Identify individual factors affecting service life. Figure 2.46. Blocks 3b to 8 of integrated system selection and design process.

115 Chapter 2. BRiDGE SySTEM SELECTiON 12. Are all alternatives considered? No Yes 13. Compare all final alternatives. Evaluate advantages and disadvantages and select most cost-effective alternative. Go to the next alternative Block 3 10b. Identify maintenance. Is service life greater than or equal to the system service life? Yes No 10a. Identify rehabilitation or replacement. 11. Incorporate requirements for rehabilitation, replacement, and maintenance into final configuration of system being considered, and compute life-cycle cost. 9. Predict service life of various components, subsystems, and elements. 8. Develop final configuration of system alternative being considered Figure 2.47. Blocks 8 to 13 of integrated system selection and design process.

116 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE tABLE 2.2. SyStem SeLection ProceSS For oPerAtionAL And Service LiFe reQuirementS Stage and Major Steps Process Preliminary Planning or Type, Size, and Location Stage 1. Identify demand requirements Operational Demand (Functionality) Requirements and Corridor-Related Items Identify corridor-, function-, and traffic-related demand requirements: • Corridor type such as interstate, urban arterial rural, or other; • Traffic volumes and required capacity; • Truck volumes; • Special vehicle uses such as oversize vehicles or tanker trucks; • Traffic maintenance requirements; • Mixed-use requirements; and • Vehicle loads and special vehicle load requirements. Service Life Requirements (Durability and Long-Term Performance) and Corridor- Related Items Identify service life requirements: • Bridge importance; • Target design service life; • Potential future needs according to corridor type, including – Potential for future bridge widening, – Potential for future widening of crossed roadways, and – Vertical clearance requirements related to future bridge widening or widening of crossed roadways. Local Site-Related Requirements Identify local site-related issues that the bridge must accommodate: • Geometrics, curvature, and skew; • Features crossed; • Horizontal and vertical clearances; • Hydraulic or waterway requirements; • Navigation requirements; • Utility issues, either carried or crossed; • Other physical boundary conditions; • Geotechnical issues; • Environmental issues; • Drainage requirements and special criteria; • Access for construction; and • Aesthetics and sustainability. (continued)

117 Chapter 2. BRiDGE SySTEM SELECTiON Stage and Major Steps Process 1. Identify demand requirements (continued) Local Items Identify local relevant environmental or man-made hazards that affect service life (follow fault tree): • Climate type – Thermal, – Coastal, and – Chemical; • Potential for hydraulic action hazard – Flood, and – Scour; • Potential for wind hazard; • Potential for other extreme-event hazards – Vehicle or vessel collision, – Fire or blast, and – Seismic event potential. Construction Constraints Identify construction requirements that can affect service life: • Construction phasing requirements; • Construction schedule requirements (such as accelerated bridge construction); and • Special local construction preferences. 2. Identify alternative solutions Identify feasible alternative bridge systems including span layouts, structure types, and materials that accommodate operational and site requirements. Alternative solutions should • Accommodate span requirements and constraints; • Accommodate curvature, profile, and skew requirements; • Provide optimal span balance; • Provide optimal superstructure and substructure cost balance; • Identify superstructure options; • Identify alternative bridge-deck options; • Identify superstructure–substructure connection options; • Identify expansion joint location and type options; • Identify bearing type options; • Identify substructure options; and • Identify foundation options. 3. Evaluate and compare alternatives Identify system service life improvement strategies that avoid future potential obsolescence by considering • Bridge types that can be widened; • Span lengths that accommodate widening of crossed roadways; and • Additional vertical clearance for bridges that may need to be widened. Evaluate factors affecting service life and identify strategies to avoid or mitigate all hazards. For each alternative bridge system, estimate potential service life of components, subsystems, and elements and compare with target design service life of bridge system. Determine rehabilitation, replacement, and maintenance requirements over target design service life. Determine estimated life-cycle costs over target design service life. tABLE 2.2. SyStem SeLection ProceSS For oPerAtionAL And Service LiFe reQuirementS (continued) (continued)

118 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Stage and Major Steps Process 4. Select optimal alternative Compare operational advantages and disadvantages of bridge system alternatives: • Identify local preferences for structure types and construction; • Compare estimated life-cycle costs; and • Select optimal cost-effective system considering operational and service life requirements and cost–benefit analysis. Final Design Stage 1. Design Design in accordance with strength and serviceability provisions of LRFD specifications. Include specific design and details to address service life issues identified in preliminary stage: • Design provision and details that allow for potential future deck replacement; • Plan and access for inspection and future maintenance and rehabilitation; • Drainage plan; • Bridge-deck protection plan; • Reinforcing bar or prestressing steel corrosion protection plan for all reinforced or prestressed concrete elements; • Concrete element protection plan; • Fatigue- and fracture-resistance plan for steel superstructure; • Corrosion protection plan for steel superstructure; • Design provisions and details for potential bearing replacement; and • Mitigation plan for applicable extreme-event hazards. 2. Maintenance issues Identify recommended future maintenance requirements for achieving design service life related to deck, superstructure, and substructure components: • Deck maintenance plan; • Drainage system maintenance plan; • Steel superstructure coating maintenance plan; • Expansion joint maintenance plan; • Bearing maintenance plan; and • Substructure maintenance plan. tABLE 2.2. SyStem SeLection ProceSS For oPerAtionAL And Service LiFe reQuirementS (continued) Block 2. The next step is to identify feasible system alternatives that satisfy opera- tional and site requirements, while also satisfying various provisions of the LRFD specifications. This step would typically include a. Preparing alternative bridge span arrangements that accommodate features crossed, horizontal and vertical clearances, geometric requirements, and other construction constraints. Geotechnical requirements affect foundation costs; when possible, span layouts need to consider the relative costs between super- structure and substructure in achieving optimum span lengths. Optimum span lengths will also vary for different superstructure types, and the interaction between optimal span and superstructure type needs to be considered in set- ting span layouts. b. Identifying feasible deck alternatives, such as CIP concrete, precast concrete, or other. These choices are further discussed in Chapter 4.

119 Chapter 2. BRiDGE SySTEM SELECTiON c. Identifying feasible superstructure alternatives that accommodate geometric and span-length requirements. Various bridge systems are described in Section 2.2. Feasible superstructure alternatives might include steel or concrete girders or concrete segmental box girders, among others. d. Identifying feasible substructure and foundation types that are compatible with superstructure and geotechnical requirements. e. Verifying preliminary member or element sizes at this stage through prelimi- nary design. Block 3. After feasible alternative bridge systems are identified, the next step is to evaluate each alternative against factors that affect service life by using the fault tree analysis described in Section 2.3. Within each system, each component, sub system, and element should be considered. Procedures in other chapters also need to be followed in evaluating certain specific components or elements such as bridge deck, joints, and bearings or in evaluating effects on materials such as ASR or freeze–thaw. A system- level evaluation of these components and elements is necessary first to assure compat- ibility among all components within the system. Block 3a. In evaluating the various components, subsystems, and elements within a system alternative, each branch of the fault tree illustrated in Figure 2.23 should be followed. The initial categories to be considered include obsolescence, mostly pertain- ing to inadequate capacity, or deficiency, pertaining to damage or deterioration. Defi- ciency is further subdivided into deficiency caused by loads, natural or human-caused hazards, or production or operation defects, which are shown in Figures 2.24, 2.25, and 2.36, respectively. The fault tree branches end with basic events or the lowest levels of resolution, which are the individual factors to be considered. The individual factors within fault trees for which the specific mitigation strategies need to be devel- oped are placed in circle symbols. Block 3b. Each factor is systematically examined and evaluated in regard to its application to the various bridge system components, and a decision is made as to whether that factor applies. For example, in the case of natural or human-caused haz- ards, is the bridge in a thermal climate, which is a cold, wet climate with heavy use of roadway deicing salt? If the bridge is not in this type of climate, this factor would not apply, and the evaluation continues to the next factor. If the bridge is in this type of climate, the evaluation proceeds to the next step. Block 4. If the individual factor applies, the potential consequence of that fac- tor should be evaluated and an appropriate strategy or strategies should be identified to either avoid the factor or mitigate its influence. For example, if the bridge is in a thermal climate, strategies to avoid or mitigate the potential of corrosion caused by deicing salts would need to be identified. Special strategies such as stainless steel deck reinforcement or metalizing the ends of steel girders under the deck expansion joints need to be included. Strategies to mitigate various factors affecting service life are given in Section 2.4. After determining the appropriate mitigation strategy, the evalu- ation proceeds to consider the next factor.

120 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Block 5. After all factors are considered, the identified strategies are summarized and integrated into the bridge system to develop a draft final configuration. Block 6. This draft configuration then needs to be checked to be certain all require- ments are satisfied and whether all identified strategies to improve service life are con- sistent with one another and are compatible between various components and elements. Block 7. If inconsistencies or incompatibilities are found, they need to be resolved by making appropriate modifications to the strategies or to the affected components or elements. For example, in evaluating a steel superstructure in a thermal climate, a strategy might be to metalize the girder ends below the deck expansion joints at abut- ments to mitigate the possibility of corrosion. A substructure evaluation, however, might identify a strategy to eliminate expansion joints and use integral abutments. These two strategies are inconsistent in that if the integral abutment strategy is imple- mented, the strategy for metalizing steel girder ends could be unnecessary. A resolution is made as to which strategy to implement. Block 8. After all inconsistencies are identified and resolved and the final con- figuration satisfies all requirements, the final configuration of the system alternative is developed. It is feasible to develop more than one configuration capable of meeting the service life requirements. See Chapter 1 for an example. Block 9. The next step in the process is to predict the service life of the various bridge components, subsystems, and elements within the final bridge system alterna- tive under consideration. Deterioration models to quantitatively predict service life are limited or nonexistent, so often the prediction is made on the basis of experience or expert opinion. The predicted service lives of the various components, subsystems, and elements are then checked to see if they will be equal to or greater than the owner- specified service life of the bridge system. Block 10a. If the predicted service lives of the various components, subsystems, and elements are not equal to or greater than the owner-specified service life of the bridge system, future rehabilitation and/or replacement of these components, sub- systems, or elements will have to be anticipated at certain intervals during the system service life. For example, if the bridge deck is predicted to have a service life less than the system service life, the rehabilitation plan might include milling and overlaying at an anticipated interval. A replacement plan would include the anticipated extent of replacement and interval. Block 10b. Future maintenance requirements will have to be identified whether or not the component, subsystem, or element service life meets the system service life. In the case of bridge decks, maintenance might include washing the deck surface to remove salt or cleaning gutters and drains to permit proper drainage. Block 11. After all requirements for replacement and/or rehabilitation and main- tenance are determined, these requirements are incorporated into the final alternative system configuration, which now includes the system layout with a compatible and consistent set of service life strategies and a replacement, rehabilitation, and main- tenance plan. The corresponding initial construction cost and life-cycle cost for this complete alternative system configuration is then computed. Block 12. When the evaluation is completed for this alternative, the engineer should return to Block 3 and repeat the evaluation steps for the next identified bridge

121 Chapter 2. BRiDGE SySTEM SELECTiON system alternative. When all identified system alternatives have been evaluated, the designer should proceed to the last step. Block 13. The last step is to compare the final alternative bridge system configura- tions and select the optimum alternative. Often this is done in a matrix-type evaluation in which various key performance categories, determined specifically for the bridge, are weighted and evaluated for each alternative. Example performance categories might include service life, traffic impact, construction duration, construction complex- ity, site suitability, local preference, or aesthetics. In this process, the advantages and disadvantages of the various alternatives are compared. Initial construction costs and life-cycle costs are also compared. With this type of evaluation, the optimum bridge system can be identified as part of a complete cost–benefit selection process. 2.5.4 System Process tables The system selection process is further expanded in Table 2.2, which illustrates the process design phases of the preliminary planning stage (also called the type, size, and location stage) and the final design stage for a new bridge. Table 2.2 supplements the information and examples provided in Sections 2.5.2 and 2.5.3. 2.5.5 Existing Bridges Many of the service life considerations for new design are also applicable to existing bridges. An inherent difference for existing bridges, however, is that the bridge system has already been selected, designed, and constructed, and may have been in service for a considerable time. An existing bridge has been subjected to factors affecting service life and may have already experienced some level of deterioration. The level of deterio- ration will dictate the type of rehabilitation required for restoration. The process for restoring and extending the service life of an existing bridge involves a detailed system evaluation rather than a system selection. The process for existing bridges will follow a similar flowchart to that shown in Section 2.5.2, except that it will start at Block 3 and will entail an evaluation of the existing system components, subsystems, and elements against the various factors in the fault tree branches. Strategies for mitigating the factors are more limited and are more focused on protecting the existing elements and materials. However, they could also consider various forms of retrofit, such as converting simple spans with deck joints to continuous spans without deck joints. The final evaluation is to determine the optimal protection strategy that will extend the existing bridge service life. 2.6 bridge mAnAgement A major component of the Guide and this chapter addresses the type of data that should be maintained for a bridge from design through fabrication, construction, and opera- tion. The framework for this documentation is the introduction of the Owner’s Manual described in Chapter 1. A bridge Owner’s Manual should be provided for unique bridges or when requested by a bridge owner. The information to be included in the Owner’s Manual is essential for proper future inspection and maintenance of the bridge in order to achieve the bridge’s target design service life. The Owner’s Manual should be provided

122 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE to the bridge owner just before opening the bridge to traffic. The bridge Owner’s Manual is similar to the design calculation document that is usually provided to the bridge owner. It is recommended that the Owner’s Manual be reviewed by an independent engineer. Chapter 1 provides a more detailed description of the bridge Owner’s Manual. Engineering judgment must be exercised in identifying the type of information to be included in the bridge Owner’s Manual. A partial list of information for the bridge Owner’s Manual includes the following: • Target design service life of the bridge as determined by the owner; • Overall process and philosophies used to address the service life design; • List of all assumptions and special data used in the service life design process; • All factors affecting service life that were identified in the initial and final service life design, with adequate justification; • All strategies that were designed into the bridge to avoid or mitigate factors affect- ing service life; • Procedure used to estimate the service lives of the bridge elements, components, and subsystems; • Description of any special steps or requirements that must be followed during construction; • Specific maintenance needs for various bridge elements, components, and sub- systems in order to achieve their expected service lives; • All considerations that were incorporated into the overall bridge system design to accommodate rehabilitation or replacement of those items, including the expected schedule; • Identification of all “hot spot” areas of the bridge that would require special in- spection or data to be collected during inspection that could be coordinated with the FHWA-sponsored Long-Term Bridge Performance Program; and • Health monitoring of unique bridges to develop a comprehensive bridge manage- ment system that might be needed and should be described in detail. The bridge Owner’s Manual should also describe how the bridge was designed, constructed, and intended to function from an operational perspective, including • Design loads, particularly any special vehicle types; • Expected superstructure deflections; • Expected movements at expansion joints and bearings; • Relevant as-built data; and • Construction methods and procedures. In summary, the bridge Owner’s Manual should provide a clear picture of the pro- cedure used to address service life design and what will be needed to keep the bridge operational for the intended service life.