| nisee |
National Information Service for Earthquake Engineering
University of California, Berkeley |
Structural Engineering, Mechanics and Materials Department, University of California Berkeley
Note: The text of this article first appeared in the American Concrete Institute publication SP-127, Earthquake-Resistant Concrete Structures - Inelastic Response and Design, (S.K.Ghosh, editor)in 1991. It is re-published here for educational purposes by kind permission of the authors and the ACI.
Observations on the performance of structures during strong earthquakes have served as an age- old means of educating builders on proper and improper construction of earthquake load resisting systems. In regions that have long been inhabited, and that are subjected to relatively frequent strong ground shaking, design procedures have evolved that result in relatively good performance of engineered structures. Although such design procedures are not universally applicable because of regional differences in construction styles, structural engineers can learn much by studying such procedures. The structural engineer can also draw upon a broad data base of engineering observations that have been reported systematically following recent earthquakes. It is the objective of this paper to highlight some of the observations as they apply to design of reinforced concrete structures.
The damaging potential of strong earthquakes is well known and is accepted as an underlying premise by most design codes. For example, the Seismology Committee of the Structural Engineers Association of California (SEAOC) adopts the philosophy that structural damage is acceptable during the rare earthquake but that collapse is not acceptable in any event (1). Thus, an engineer should not be surprised if buildings designed according to the SEAOC recommended lateral force provisions exhibit structural damage following a severe earthquake. However, an observation of performance that suggests the possibility of incipient collapse is certainly noteworthy. From such observations can be gleaned lessons on construction to be avoided.
Observations on good performances of buildings are also noteworthy, as they provide lessons on desirable structural systems. We are fortunate that most engineered buildings have performed well during recent earthquakes. Still, within this sea of successes, we as observers are prone to focus on the few failures. An effort will be made in the following to present observations on good structural performances where appropriate.
The subjective nature of observations of earthquake performance should be immediately clear. In most instances, observations must be drawn from incompletely documented performances of buildings with uncertain structural characteristics and subjected to unknown earthquake motions. Conclusions that are drawn from such observations are dubious. Several systematic studies of earthquake performance have been undertaken (2, 3, 4, 5, 6, 7, 8). These can provide conclusions of greater substance, and will be cited where possible in the following.
It has been observed repeatedly that proper selection of the load carrying system is essential to good performance under any loading. A properly selected structural system tends to be relatively forgiving of oversights in analysis, proportion, detail, and construction. On the other hand, extra attention to analysis and detail is not likely to improve significantly the performance of a poorly-conceived structural system. This observation is particularly appropriate in earthquake-resistant design where the intensity and orientation of loading are highly uncertain. Buildings having simple, regular, and compact layouts incorporating a continuous and redundant lateral force resisting system tend to perform well and thus are desirable. Complex structural systems that introduce uncertainties in the analysis and detailing or that rely on effectively non-redundant load paths can lead to unanticipated and potentially undesirable structural behavior. These general observations are further elaborated in the following.
An essential characteristic of any lateral load resisting system is that it must provide a continuous load path to the foundation. Inertial loads that develop due to accelerations of individual elements must be transferred from the individual reactive elements to floor diaphragms, to vertical elements in the lateral load system, to the foundation, and eventually to the ground. Failure to provide adequate strength and toughness of individual elements in the system, or failure to tie individual elements together can result in distress or complete collapse of the system.
One of the earliest lessons from earthquakes was realization that structural and nonstructural elements must be adequately tied to the structural system (9). Numerous examples can be found of detachment of exterior cladding (10,11,12), of parapets (7,13,14), and of various nonstructural elements within buildings (9,16). This observation has resulted in local ordinances and building code provisions (1,17) that specifically require that individual elements of a building be adequately tied to the structure.
Inertial loads that develop in individual elements must be carried to the vertical elements of the lateral load resisting system by horizontal floor diaphragms. Concrete diaphragms, with appropriate struts, ties, and boundary elements should be provided with adequate reinforcement to transmit these forces. Examples of distress in diaphragms and collectors have been reported (14,15).
Failures due to discontinuity of vertical elements of the lateral load resisting system have been among the most notable and spectacular. One common form of this type of discontinuity occurs when shear walls that are present in upper floors are discontinued in the lower floors. The result is frequently formation of a soft first story that concentrates damage. A well-known example (shown below) is the Olive View Hospital, which nearly collapsed due to excessive deformation in the first two stories during the 1972 San Fernando earthquake and subsequently had to be demolished (6,18).
Another well-studied example is the Imperial County Services Building, which nearly collapsed during the 1979 El Centro earthquake (3,4,19). In this latter building, lateral forces that accumulated in exterior walls in the upper stories had to be transmitted by the diaphragms through shear to interior first-story walls. Large overturning forces that developed in the discontinued exterior walls had to be carried by columns supporting the walls, an effect that contributed to severe damage in the columns. Numerous other examples where discontinuity in shear walls contributed to failures have been identified (20,21).
Sudden changes in stiffness, strength, or mass in either vertical or horizontal planes of a building can result in distributions of lateral loads and deformations different from those that are anticipated for uniform structures. Although the effects of such irregularities can and have been designed for with success, the uncertainties associated with the effects of such irregularities are such that they are better avoided if possible.
Sudden changes in stiffness and strength between adjacent stories are commonly observed. Such changes are associated with setbacks (including penthouses and other small appendages), changes in structural system over height (eg., discontinuous shear walls), changes in story height, changes in materials, and unanticipated participation of nonstructural components. A common problem of such discontinuities is that inelastic deformations tend to concentrate in or around the discontinuity. Examples where discontinuity apparently resulted in severe damage or collapse are commonplace in the earthquake reconnaissance literature (4,7,14,18,21,22,23,24,25,26).
Apparent vertical irregularities can occur due to the interaction between adjacent structures having inadequate separation. A tall building adjacent to a shorter building may experience irregular response due to effects of impact between the two structures. The effect can be exacerbated by local column damage due to pounding of the roof of the small building against the columns of the taller one. Examples of distress due to this phenomenon were observed following the 1985 Mexico earthquake (22).
The most common form of vertical discontinuity arises because of unintended effects of nonstructural elements. The problem is most severe in structures having relatively flexible lateral load resisting systems because in that case the nonstructural component can compose a significant portion of the total stiffness. A common cause of failure occurs in infilled frames. If properly designed, the infill can improve the performance of the frame due to its stiffening and strengthening action (27). However, soft stories can result if infills are omitted in a single story (often the first story), as apparently has occurred in several instances (7,9,20,21,22). Even if placed continuously and symmetrically throughout the structure, a soft story can form if one or more infill panels should fail (7,13,22,28).
Partial-height frame infills are also common. In this form of construction, an infill extends between columns from the floor level to the bottom of the window line, leaving a relatively short portion of the column exposed in the upper portion of the story (as shown below). The shear required to develop flexural yield in the effectively shortened column can be substantially higher than that which would develop for flexural yield of the full-length column. If the design has not considered this effect of the infill, shear failure of this so-called 'captive column' can result before flexural yield. Complete collapse of the column (and building) can occur if it is not well equipped with transverse steel. This form of distress is a common cause of building damage and collapse during earthquakes (7,14,21,22,26,29).
Mass, stiffness, and strength plan irregularities can result in significant torsional response. Inelastic torsional response cannot at present be rectified with results of elastic analysis, and techniques for inelastic analysis of complete building systems considering torsion are largely unavailable and unverified. Given such uncertainties and difficulties with analytical techniques, prudence will direct the engineer to design buildings to have substantial torsional resistance, near symmetry, and compactness of plan. Examples of structural distress attributed to torsion abound. For example, collapses have been attributed to torsional effects associated with L and U-shaped plans (21). Asymmetric layouts of lateral load resisting elements has also been cited (22, 23). Asymmetric layout of infill panels has contributed to many structural failures. Of special note is the high proportion of failure of corner buildings that have infill panels on the two inner perimeter walls and open frames on the street-side perimeter walls (22). Torsion due to asymmetric failure of infill panels also appears to have contributed to building failures (22). Following the 1985 Mexico City earthquake, torsion was identified (22) as one of the most prevalent contributors to failure.
Although not currently the focal point of most seismic design codes, control of lateral drift should be a central element of any seismic design. Excess drift can lead to excessive distortion, and thence, damage, of structural and nonstructural components. Because repair cost is the primary measure of the success of a building that has survived an earthquake, damage control is essential. Control of damage in nonstructural components is important because they typically comprise a significant majority of the total value of a building and because falling nonstructural elements can cause injury and death to building inhabitants. Widespread environmental disaster due to nonstructural damage is not inconceivable for certain industrial facilities. Numerous examples can be found where a structure has survived an earthquake without damage but significant repairs to the architectural components and contents have been required (16).
Control of drift is also important because nonductile or moderately-ductile elements of the structural system can be damaged by excessive distortion. Control of drift by structural walls is a proven means of reducing damage to weak, low-ductile framing. For example, in Chile multistory RC structures are provided with a stiff, continuous system of structural walls to control drift. Conventional detailing practice in that country generally does not follow the ductile detailing conventions of seismic regions of the U.S., but rather follows those of the nonseismic provisions of ACI 318-89 (35). The exceptionally good performance of these buildings during the 1985 Chile earthquake bears testimony that drift control by structural walls can protect relatively nonductile framing elements (8). Similar conclusions can be drawn by comparing earthquake performances of various buildings reported in the literature. The relative performances of the Banco de America and Banco Central (shear wall versus frame buildings) during the Managua earthquake (14) are commonly cited in this regard.
Drift control is important also in preserving the vertical stability of a structural system. If a structural system is excessively flexible, and particularly if it is also massive, collapse can occur due to P-delta effects. Observations following the 1985 Mexico earthquake demonstrated that this was a particularly prominent problem with flat-slab structures because of their relatively low lateral load stiffness (22).
Where buildings are constructed in close proximity to one another, damage due to pounding between the buildings is possible. Several examples of building failures due to pounding have been observed (9,11,14,22,28,30,31). Pounding may result in irregular response of buildings of different heights, local damage to columns as the floor of one building collides with columns of another, collapse of damaged floors, and in many cases collapse of entire structures (22). Damage due to pounding can be minimized by drift control, building separation, or as a last resort, aligning floors in adjacent buildings so that columns do not bear the blows of oncoming floor slabs.
Several examples can be cited where buildings in close proximity apparently supported one- another rather than resulting in damage (9). In most cases, the buildings are of similar story and total height, of similar stiffness, and located sufficiently close that pounding impacts are of relatively low energy.
Excess mass can lead to unnecessary increases in lateral inertial forces, reduced ductility of vertical load resisting elements, and increased propensity toward collapse due to P-delta effects (6,21,22). For these reasons, efforts should be made to achieve a system that is as lightweight as possible. This is not to suggest that lightweight concrete aggregates be used for all concretes, as laboratory research has shown that performance of elements of lightweight concrete can be inferior to that of normal weight concretes (32). However, concrete as an architectural fill and soil for landscaping on top of structural slabs provide unnecessary mass without structural benefit. Numerous examples of buildings that collapsed due to the presence of excessive vertical loads were identified following the 1985 Mexico city earthquake (22). In many cases, the excesses were attributable to dead loads that exceeded the specified values; more often excessive live loads resulted from change in occupancy of a building.
Irregularity of mass distribution in vertical and horizontal planes can result in irregular responses and complex dynamics (6). These should be avoided where practicable.
Structural systems that combine several lateral load resisting elements or subsystems generally have been observed to perform well during earthquakes (2,9,23). Redundancy in the structural system permits redistribution of internal forces in the event of failure of key elements. Without capacity for redistribution, global structural collapse can result from failure of individual members or connections. Redundancy can be provided by several means; a dual system, a system of interconnected frames that enable redistribution between frames after yield has initiated in individual frames, and multiple shear walls. Redundancy, combined with adequate strength, stiffness, and continuity, can alleviate the need for excesses in ductile detailing. This aspect was discussed previously with regard to the redundant shear-wall systems used in buildings in Chile (2). The benefits of redundancy are also apparent by comparing the successful performance of bearing wall buildings with the generally poor performance of nearby precast frame buildings as reported for the 1988 Armenia earthquake (15).
Cyclic deterioration in the hysteretic response has been observed to occur in reinforced concrete structures in the laboratory under certain conditions, in particular, where behavior is influenced by concrete in compression, by shear, and by anchorage. This deterioration is believed to be the cause for damage in buildings that have experienced successive earthquakes. Several examples of buildings that had been damaged during previous earthquakes and were damaged more severely in subsequent earthquakes have been reported (2,21,22,33). In many cases, following the previous earthquake these buildings had not been repaired, had received only cosmetic repair, or had been repaired in a manner that induced more severe damage in subsequent earthquakes. Detailed evaluations of the damage state and implementation of appropriate reparative measures are clearly important.
Conventional earthquake resistant design of buildings relies on element ductility to enable redistribution and reduction of internal actions, and dissipation of earthquake energy. Observations have shown repeatedly the necessity of attention to proportioning, to ensure that inelastic action occurs at appropriate locations, and detailing, to ensure adequate ductility in these locations that yield. Some of the more prominent observations and lessons are summarized in the following.
Structures should be proportioned to yield in locations most capable of sustaining inelastic deformations. In reinforced concrete frame buildings, attempts should be made to minimize yielding in columns because of the difficulty of detailing for ductile response in the presence of high axial loads and because of the possibility that column yielding may result in formation of demanding story sway mechanisms and collapse. Examples of collapse of stories or complete structures due to column weakness and limited ductility can be found following many destructive earthquakes (7,14,20,21,22,25,26). The problem of yielding in columns rather than beams is particularly pronounced in structures for which gravity load effects control proportions and strengths, resulting in beam flexural strengths exceeding by some margin the column flexural strengths. This situation typically occurs in buildings having long beam spans, and in the upper floors of buildings where design seismic effects are relatively low. The latter situation may be a contributing factor for failures in upper stories of frames (20,22,26).
Observations of failures due to yielding in columns have led to formulation of the weak beam- strong column design philosophy in which column strengths are made at least equal to beam strengths. The intended result is columns that form a stiff, unyielding spine over the height of the building with inelastic action limited largely to beams. Even in structures so designed, yielding in first-story columns should be anticipated and appropriate details provided (34). In buildings where architectural requirements require wide bays with resulting strong girders, the strong column-weak beam design philosophy may be difficult to implement. In such cases, columns should be detailed to sustain inelastic action or, preferably, continuity of deformations over height should be enforced by providing continuous structural walls.
Coupled wall systems are generally proportioned so that a considerable portion of the inelastic energy dissipation occurs in the coupling beams. Performance of such systems generally has been good, although damage to coupling beams and slabs is not uncommon (2). Closely spaced transverse reinforcement or special reinforcement details (34,35) are recommended.
The structure should be proportioned and detailed in a manner that is consistent with the expected inelastic deformation mode. If inelastic flexure is preferred in selected elements, design actions and appropriate proportions should be selected to ensure that the elements can achieve the flexural strengths. For beams and columns, shear failures have occurred because design shear forces were determined on the basis of design lateral forces rather than the shear required to equilibrate the plastic moment capacities of the member (6,7). Consequently, most modern codes stipulate that design shears be evaluated on the basis of likely plastic hinge locations with appropriate factors of safety applied to member strengths and transverse loading.
Failures have also apparently occurred because bar cutoffs were inconsistent with the moment distribution that develops when flexural strengths are reached at member ends (14,20). Uncertainties in determining these moment distributions has in part motivated most model code writing bodies to recommend continuous nominal reinforcement on both faces of all structural elements.
Nonstructural components have been observed to substantially alter structural behavior. For example, slabs on grade can change assumed base fixity conditions, and stairways and partial infills in frames can alter the member actions. These interactions can result in increased member shear demands as well as formation of plastic hinges away from those regions detailed for ductile action on the basis of intended behavior. Numerous examples of this type of damage (3,5,14,26) emphasizes the need for realistic assessment of member behavior.
Corner columns have statistically greater damage rates than other columns in moment resisting frames (9). An apparent cause is the combined effect of actions from perimeter frames oriented perpendicular to one another and connecting at the corner column. Extra attention appears warranted in selection of design actions considering orthogonal effects and in detailing and construction.
Generous supply and appropriate placement of transverse reinforcement in reinforced concrete beams, columns, beam-column connections, and walls have proven to be desirable. Such reinforcement is useful for concrete confinement, resistance to shear, restraint of longitudinal reinforcement buckling, and improved anchorage. Failure to provide transverse reinforcement at member ends where plastic hinges are anticipated results in reduced flexural strength (particularly sustained strength) and ductility, as well as degradation of shear resistance. Closely spaced transverse reinforcement is particularly recommended for the unrestrained length of captive columns where inelastic flexure is combined with high shear force. Boundary elements of walls where significant inelastic action is anticipated should be well confined to provide ductility under axial compression. Columns supporting discontinuous walls should be confined over their entire height (35).
Distress in beam-column joints, in some cases leading to building collapse, has been attributed to inadequate joint confinement (7,13,22,29). Several disastrous failures during the 1985 Mexico City earthquake could apparently be attributed to joint failure in cases where heavy spiral or rectilinear confinement in columns above and below a joint was discontinued at the joint. In general, confinement in columns should continue through the connection region.
Effective concrete confinement can be obtained using either spiral or rectilinear reinforcement. The former is generally the more effective form of confinement reinforcement (18). To be effective, transverse reinforcement must be coupled with well distributed longitudinal reinforcement. Poorly spaced and bundled column bars have been displayed by past earthquakes in a manner suggesting that they contributed to the structural distress (22). In addition, the transverse reinforcement must be anchored so as to remain effective in the event that concrete cover spalls. Perimeter hoops without hook anchorage into the core concrete were observed to be largely ineffective following the 1985 Mexico earthquake, as judged by the fact that they could be easily removed by hand following the earthquake.
Strength and toughness must be developed not only within members themselves but also in the connections between members. Numerous examples can be found where beam-column connections with inadequate transverse reinforcement failed (22,25,29). Few examples, if any, have been found where joints with transverse reinforcement approaching current recommendations have failed. Problems have been observed in joints where the members frame eccentric to the joint, and where members having non-coincident longitudinal axes frame into a single joint (25).
Slab-column connections have suffered distress in numerous earthquakes, and in several cases have contributed to collapse (21,22,28,36). In the 1985 Mexico City earthquake, the presence of heavy vertical loads is believed to have resulted in excessive shear stress on connections, with resultant decrease in connection moment capacity and ductility and increase in P-delta moments. Combined with the relatively large flexibility of slab-column frames, several collapses resulted (22,26). In the event of punching failure at a connection, bottom slab reinforcement anchored through the columns has been observed (Mexico, 1985) to be an effective means of preventing or delaying collapse; lack of such reinforcement has been observed to result in catastrophic failures.
Continuity between members and between members and joints is also essential. Following the 1985 Mexico City earthquake, several collapsed buildings were found where columns had not been adequately interconnected through the joints by continuous longitudinal reinforcement. Failures have also been observed where reinforcement splices within members were of insufficient length or were inadequately located (28). The ability of continuous bottom reinforcement passing through connections to support beams and slabs by catenary action following initial shear or connection failures was evident following the 1985 Mexico earthquake and 1987 Whittier Narrows earthquake (10). These observations point to the general need of providing continuous ties within the structure, and the specific need of considering distributions of internal member actions that can occur under severe seismic loading.
Proper anchorage of reinforcement under the action of cyclic inelastic load reversals requires that the reinforcement to be developed have adequate transverse reinforcement and concrete surrounding each bar. In particular, inadequate bond of bundled column reinforcement is believed to have been the cause of damage to and collapse of several buildings in the 1985 Mexico City earthquake.
Observations of earthquake performance demonstrate clearly and repeatedly the need for punctilious attention to design, detail, and construction. In particular, structures that rely on ductile response and that do not provide multiple load paths to the foundation require dedicated, professional inspection to ensure that required ductile details are properly implemented. The designer must ensure that construction drawings and documents are clear and unambiguous, and that actual conditions of construction do not interfere with the behavior intended in design.
As examples of the importance of construction, inadequate details at the base of columns in the Imperial County Service Building (shown below) are believed to have been contributory to the column failure (3).
Improper anchorage of transverse reinforcement has resulted in failure of confinement in columns during the 1985 Mexico earthquake, and improperly executed construction joints in shear walls have resulted in movement and damage along the joints (23,28). Numerous other examples where poor construction and material quality contributed to building failures can be found (7,9).
Of particular note is the practice of providing reliable strength, redundancy, and continuity in lieu of requiring ductile detailing and inspection. In Chile, this practice has resulted in a high-rise reinforced concrete building form that has performed exceptionally well in recent earthquakes (2). Although still under study, this simple and apparently effective alternative to current practice in the U.S. is among the important lessons to be derived from recent observations.The occurrence of inelastic action in some buildings during strong earthquakes is evident in the observations that have been made thus far. Perhaps less evident, and deserving emphasis, is the observation that the inelastic action often is attributable to effects that were not foreseen during design. Procedures for inelastic design are of little use if the analyst is unaware of the potential modes of inelastic action. The less than omniscient designer should not attempt to replace simplicity, continuity, redundancy, and detail with computed predictions that such qualities are unnecessary.
Furthermore, behavioral aspects of many forms of inelastic response (for example, joint deformations, failures in shear and anchorage, severe discontinuities, and three-dimensional inelastic response including torsion) at present cannot be modeled confidently even when identified. The influence of such behaviors should be minimized by layout of the structural system, and proportion and detail of its components. Simple design techniques usually are sufficient for this purpose. Inelastic design procedures should steer clear of responses involving the uncertain modes mentioned above.
1. "Recommended Lateral Force Requirements and Commentary," Seismology Committee, Structural Engineers Association of California, San Francisco, California, 1989.
2. Wood, S. L., Wight, J. K., and Moehle, J. P., "The 1985 Chile Earthquake: Observations on Earthquake-Resistant Construction in Vina del Mar," Civil Engineering Studies, Structural Research Series No. 532, University of Illinois at Urbana-Champaign, Urbana, Illinois, February 1987, 176 pp.
3. Zeris, C., Mahin, S. A., and Bertero, V. V., "Analysis of the Seismic Performance of the Imperial County Services Building," Proceedings, Eighth World Conference on Earthquake Engineering, San Francisco, California, 1984.
4. Kreger, M. E., and Sozen, M. A., "A Study of the Causes of Column Failures in the Imperial County Services Building During the 15 October 1979 Imperial Valley Earthquake," Civil Engineering Studies, Structural Research Series No. 509, University of Illinois, Urbana, Illinois, August, 1983.
5. Bertero, V. V., et. al., "Seismic Response of the Charaima Building," Report No. EERC- 70/4, Earthquake Engineering Research Center, University of California, Berkeley, California, 1970 .
6. Mahin, S. A., et. al., "Response of the Olive View Hospital Main Building during the San Fernando Earthquake," Report No. EERC-76/22, Earthquake Engineering Research Center, University of California, Berkeley, California, 1976.
7. Leeds, A., ed., "El-Asnam, Algeria Earthquake, October 10, 1980," Earthquake Engineering Research Institute, El Cerrito, California, January 1983.
8. Wallace, J. W., and Moehle, J. P., "The 1985 Chile Earthquake: A study of Requirements for Bearing Wall Buildings," Report No. UCB/EERC-89/05, Earthquake Engineering Research Center, University of California at Berkeley, Berkeley, California, 1985.
9. "Reducing Earthquake Hazards: Lessons Learned from Earthquakes," Publication No. 86- 02, Earthquake Engineering Research Institute, El Cerrito, California, November 1986, 208 pp.
10. The Whittier Narrows Earthquake, October 1. 1987, H.J. Degenkolb Associates, Engineers, San Francisco, 1988, 65 pp.
11. Berg, G. V., and Stratta, J. L., "Anchorage and the Alaskan Earthquake of March 27, 1964," American Iron and Steel Institute, Washington, D. C., 1964.
12. Degenkolb, H. J. and Wyllie, L. A., "Foothill Medical Center," in San Fernando, California, Earthquake of February 9. 1971, Vol. I, Part A, U.S. Department of Commerce, Washington, D. C ., 1973.
13. Stratta, J. L., et. al., "Earthquake in Campania-Basilicata, Italy, November 23, 1980," National Research Council and Earthquake Engineering Research Institute, El Cerrito, California, 1981, 100 pp.
14. Wyllie, L. A., et. al., "Effects on Structures of the Managua Earthquake of December 23, 1972," Bulletin of the Seismological Society of America, Vol. 64, No. 4, 1974.
15. "Armenia Earthquake Reconnaissance Report," L. Wyllie and J. Filson, eds., Earthquake Spectra, Earthquake Engineering Research Institute, El Cerrito, California, August 1989, 175 pp.
16. "Nonstructural Issues of Seismic Design and Construction," (selected papers from a workshop), Earthquake Engineering Research Institute, El Cerrito, California, June 1984, 122 pp.
17. "Uniform Building Code," International Conference of Building Officials, Whittier, California, 1988.
18. Johnston, R.G., and Strand, D. R., "Olive View Hospital," San Fernando, California, Earthquake of February 9, 1971, Vol. 1, Part A, U. S. Department of Commerce, Washington, D. C., 1973.
19. Leeds, D. J., ed., "Imperial County, California, Earthquake of October 15, 1979," Earthquake Engineering Research Institute, El Cerrito, California, 1980.
20. Hanson, R. D., and Degenkolb, H. J., "The Venezuela Earthquake of July 29, 1967," American Iron and Steel Institute, Washington, D.C., 1969.
21. Shepherd, R., ed., "The San Salvador Earthquake of October 10, 1986," Earthquake Spectra, Vol. 3, No. 3, August 1987.
22. Rosenblueth, E., and Meli, R., "The 1985 Earthquake: Causes and Effects in Mexico City," Concrete International, Vol. 8, No. 5, May 1986, pp. 23-34./
23. Wyllie, L. A., ed., "The Chile Earthquake of March 3, 1985," Earthquake Spectra, Vol. 2, No. 2, February 1986.
24. Lagorio, H. L., and Mader, G. G., "Earthquake in Campania Basilicata, Italy, November 23, 1980, - Architectural and Planning Aspects," Earthquake Engineering Research Institute, El Cerrito, California, July, 1981, 88 pp.
25. Carydis, P. G., et. al., "The Central Greece Earthquakes of February-March 1981," Earthquake Engineering Research Institute and the National Research Council, El Cerrito, California, 1982.
26. Mitchell, D., et. al., "Lessons from the 1985 Mexican Earthquake," Canadian Journal of Civil Engineering, Vol. 13, No. 5, 1986, pp.535-557.
27. Bertero, V. V., and Brokken, S., "Infills in Seismic Resistant Building," Journal of Structural Engineering, ASCE, Vol. 109, No. 6, June 1983, pp. 1337-1361.
28. Forell, N. F., and Nicoletti, J. P., "Mexico Earthquakes, Oaxaca- November 29, 1978, and Guerrero - March 14, 1979," Earthquake Engineering Research Institute, El Cerrito, California, October 1980, 89 pp.
29. Yanev, P. I., ed., "Miyagi-Ken-Oki, Japan, Earthquake, June 12, 1978," Earthquake Engineering Research Institute, El Cerrito, California, 1978.
30. "Earthquake in Romania, March 4, 1977," EERI Newsletter, Vol. 11:3B, Earthquake Engineering Research Institute, El Cerrito. California, 1977.
31. Berg, G. V., et. al., "Earthquake in Romania, March 4, 1977," Earthquake Engineering Research Institute, El Cerrito, California 1980, 39 pp.
32. Soleimani, D., Popov, E. P., and Bertero, V. V., "Hysteretic Behavior of Reinforced Concrete Beam-Column Subassemblages," Journal of the American Concrete Institute, Vol. 76, No. 11, November 1979, pp. 1159-1178.
33. Stratta, J. L., and Wyllie, L. A., "Friuli, Italy, Earthquake of 1976," Earthquake Engineering Research Institute, El Cerrito, California, 1979.
34. Park, R., and Paulay, T., Reinforced Concrete Structures, John Wiley and Sons, Inc., New York, 1975, 769 pp.
35. "Building Code Requirements for Reinforced Concrete, and Commentary" (ACI 318-89), American Concrete Institute, Detroit, Michigan, 1989.
36. Rosenblueth, E., "The Earthquake of 28 July 1957 in Mexico City, " Proceedings, Second World Conference on Earthquake Engineering, Tokyo, July, 1960, pp. 359-378.
37. Housner, G. W. , and Jennings, P. C. , Earthquake Design Criteria, EERI Monograph, El Cerrito, CA.
38. Arnold, C., and Reitherman, R., Building Configuration and Seismic Design, Wiley- Interscience, New York, 1982.
Updated January 23, 1998.
Send comments or questions to
EERC Library