International Journal of the Physical Sciences Vol. 6(1), pp. 27-34, 4 January, 2011 Available online at http://www.academicjournals.org/IJPS DOI: 10.5897/IJPS10.411 ISSN 1992 - 1950 ©2011 Academic Journals

Full Length Research Paper

A conceptual design approach to manage large earthquake forces

Yusuf Ateş

Süleyman Demirel University, Isparta, Turkey. E-mail: yates@sdu.edu.tr. Tel: +90 544 592 5553. Fax: +90 246 2370953.

Accepted 07 December 2010

Design is an area over which the engineer has the most control, both technically and financially, in creating structures, which can withstand destructive forces of earthquakes. The most widely applied philosophy is to design for earthquake-resistant structures with the idea that stronger and stronger building materials are needed to resist larger and larger forces from earthquakes. This approach may be adequate when designing for protection against small earthquakes, but has serious shortcomings when designing for large earthquakes, as is evident from many after-earthquake scenes. In this article, a force-management approach is illustrated where the emphasis is on management of the earthquake forces. The concepts of stress concentration, dispersement and re-direction, and the shape of the structure, play a significant role in this new approach. Using numerical modeling results and field data, it is shown that the shape of buildings can be designed to disperse the earthquake forces on impact minimizing the stress concentration in the buildings, and greatly improving the overall safety. Key words: Design, earthquakes, shape, stress distribution, building safety.

INTRODUCTION In many regions around the world, earthquakes and their sheer destruction have been a major cause of loss of lives and infrastructures. Many disastrous earthquakes have been reported around the world. For example, over 17,000 people lost their lives in the 1999 Đzmit earthquake of M = 7 in Turkey (Atabey, 2000; Gülhan and Güney, 2000). Atabey (2000) also reported that about 120,000 buildings were severely damaged during this earthquake. Many other places on the globe are vulnerable to the forces of earthquakes (Spence, 2007; James, 2006; Shedlock et al., 2000; Uitto, 1998; Ambraseys and Finkel, 1995). Large earthquakes (M≤6) are responsible for the majority of the damage. According to Spence (2007), almost 80% of about 1 million deaths since 1960 have been caused by just ten largest earthquakes (Table 1).

Studies about earthquakes are ongoing around the world and significant progress has been made in understanding their mechanism of occurrence, effects on structures and probability of occurrence in a given region (Aki, 1989; Vere-jones, 1970, Hagiwara, 1974; Bolt 2003; Keilis-Borok and Rotwain, 2002; Sezen et al., 2003). However, it is virtually impossible to forecast the exact

time of happening of an earthquake and it’s magnitude, depth or duration quite well before the event. The prediction of the Haicheng earthquake (M=7.3) that occurred in north-east China on 4 February 1975, where people have been warned and measures taken for civil protection (Adams, 2006), can be noted here as an exception. While research and development is ongoing in all of these fronts, it is crucial to take precautions, starting from the areas over which the human has more control on reducing the damages from catastrophic earthquakes. In the area of design approach, it is viable to design structures to eliminate or minimize the level of damage, and save lives.

There are some well-known methods that can be used to reduce the level of damage to structures (Coburn et al., 1995; Gülkan, 2005). For example, the level of damage can be reduced by selecting building sites appropriately. It is known that a building on rock has better chance of survival and/or sustains less damage than a building whose foundations are in soil, and even better chance than the building on soil with high water table (liquefaction risk) (Mollamahmutoglu et al., 2003). An underground structure is similarly better off than a

28 Int. J. Phys. Sci.

Table 1. Large earthquakes occurred since 1960 and their damage (Spence, 2007).

Earthquake People

Event Country Date Local

Time

Magnitude

(Mw USGS) Killed Injured Homeless

Ancash Peru 31/05/1970 15:23 7.9 66 794 143 331 -

Guatemala Guatemala 04/07/1976 03:03 7.5 23 000 77 000 1 166 000

Tangshan China 28/07/1976 03:42 7.5 242 419 164 581 -

Armenia Russia 07/12/1988 11:41 6.8 25 000 12 000 530 000

Manjil Iran 21/06/1990 00:30 7.7 40 000 105 000 105 000

Kocaeli Turkey 17/08/1999 03:02 7.6 17 437 43 953 600 000

Bhuj India 26/01/2001 08:46 7.7 13 800 166 812 1 790 000

Bam Iran 26/12/2003 5:26 6.6 32 000 26 628 45 000

Indian ocean Indenosia, Thailand,

Sri Lanka 26/12/2004 07:58 9.3 283 100 41 810 1 033 464

Kashmir Pakistan 08/10/2006 08:50 7.6 73 338 69 142 2 800 000

surface structure directly above it (Sharma and Judd, 1991; Ateş et al., 1994; Wetmiller et al., 1996). However, as is evident from after-earthquake scenes, these measures are not enough to protect structures from large earthquakes and further means of protection are needed.

This article explores beyond these measures and concerns with the structures in similar sites and built with similar materials; so that the emphasis is on one factor: The implication of an overall design approach on the overall safety of the structure. The purpose is to provide an approach where the designers and/or architects are more concentrated on how to minimize the forces that a building element experiences and thus, how to manage and tolerate large earthquake forces that otherwise could not be tolerated.

The shape of a building plays a crucial role in this approach. The hypothesis is that by using proper shapes in design, the on-coming forces to the structure are distributed among the structural elements, thus managed (force-management) approach, rather than exposing the elements to the total impact of the force, thus unmanaged (resist-full-force) approach. Designing with the first approach provides a dramatic improvement in structural safety. THE CONCEPT OF STRESS DISTRIBUTION AND SURVIVABILITY OF STRUCTURES DURING AN EARTHQUAKE During an earthquake, the collapse of a building starts as a result of stresses being concentrated at a location where they grow larger than the strength of the material in this location. If one could distribute the oncoming forces equally over the building elements (that is, average them out), then most of the buildings would survive a large earthquake with minimum damage. This occurs because, by averaging the on-coming forces over many elements of the building, the maximization of stresses in certain localities would be prevented, the stress in these localities remains below the strength of the materials and components, and the failure initiation, which could progress into a total collapse otherwise, could be avoided.

This avoidance is possible by employing certain shapes in building design such that the on-coming waves (forces) are dispersed away from the building first, and then the remaining ones are distributed appropriately throughout the building (Figure 1).

As a result of this distribution, the building elements would ‘see’ less of a force and building survival is improved; otherwise, based on the resist-full-force philosophy, the engineer has to design for much ‘stronger’ components. This will help to a certain extent against smaller earthquakes (M≤6), but as is evident from the after-earthquake scenes (e.g., Kocaeli, Turkey), it does not provide a good solution to the problem. By employing the earthquake-resistant approach, one needs to design all elements of the building based on maximum stress to avoid collapse. Thus, this approach relies heavily on the materials and construction techniques for the stability of the building during an earthquake. While it is possible to employ stronger and stronger materials, it has certain crucial disadvantages. First of all these materials would cost more, and secondly stronger material is a stiffer material; and using stiff materials for earthquake design is generally not recommended for the reason that it cannot stand the forces generated by large earthquakes. According to the approach most commonly used, the building needs to sway to some degree if it is to accommodate the heavy forces. On the other hand, because the earthquake-tolerant design (manage the forces first approach) aims for dispersion of the forces first, it does not rely on as strong materials or the necessity of swaying.

The following three multi-disciplinary examples illustrate stress distribution and its role in survival of different surface and subsurface structures during an earthquake. The first example illustrates damage in some apartment blocks of similar size, but having different orientation to the main earthquake wave direction. It is postulated that the apartment blocks of the same size oriented in different direction from different shapes under the same earthquake forces and create an opportunity to examine the effect of shape. In these blocks, the level of damage is quite different depending on shape. Here, the buildings that have a larger area facing the earthquake wave direction suffer the most damage. The second example compares the stability of three different tunnels excavated in the same stress environment (force field). The tunnels are constructed using similar technology. The main difference between them is that they have three different shapes. In this same environment, while one tunnel exhibits fracturing right away and develops v-shaped notches over time, others can remain quite stable. In the third and last example, classic structures, which

Figure 1. Two force-managing, smooth-shaped

structures (top) compared with two force resisting, barrier-type structures (bottom).

endure severe forces due to their harsh environment, are examined for their shape and stress-distribution relation and survival under these conditions.

EXAMPLES/RESULTS Example 1: Shape and damage level in apartment blocks during and earthquake The level of earthquake damage to buildings can be differentiated based on the shape of the structures. Lekkas (2002) reports on damage inflicted on multi-story buildings in Ceyhan during the Adana (Turkey) earthquake, which severely hit the broader area: ‘..More specifically, a series of twenty rectangular-based apartment blocks, used for housing purposes have been built at the eastern sector of the city. All of them are identical in type and form, but are oriented in two different ways. Half of them are oriented along the NE-SW direction and the other half is perpendicular to the former, oriented NW-SE. Among the striking features of those buildings were the elongated supporting columns, whose long side was perpendicular to the long side of the building.., that is to say, NE-SW supporting columns for the NW-SE oriented apartment blocks and vice versa. Of all the apartment blocks, the ones that collapsed were those oriented NESW (with NW-SE supporting columns). On the contrary, no collapse was observed at the NW-SE oriented buildings (With NE-SW supporting columns).”

Ateş 29 Lekkas et al. (2002) went on to conclude that the difference was due to “(i) the orientation of the long side of the buildings in relation to the epicentre and (ii) the increased strength of the elongated supporting columns along the direction parallel to their long side. The ones that were oriented parallel to the direction of the seismic wave propagation withstood the shock, while the others did not.“

The damage and orientation relationship that Lakkas et al. (2002) described, can also be explained in terms of the building shape such that the orientation, which is parallel to the direction of earthquake has a shape that can tolerate the earthquake shocks because it has lesser of an wave-impact area and more of an area where the waves can be directed along or dispersed. The apartment blocks that are perpendicular to the earthquake waves receive the full force and must resist the waves as it has no feature to redirect or disperse the waves. In other words, the buildings that have larger areas facing the earthquake wave direction suffer the most damage. To account for a vertical ground motion, the shape factor can be considered in foundations design as well. Example 2: Shape and stability of underground openings Generally, design of underground structures, such as tunnels, starts with the determination of vertical forces that are largely due to gravity, and horizontal forces that are largely due to the tectonic movements in the earth. Once these forces are known, initial determination of the shape of tunnel is made. The process may be iterative to consider the size required for the use of the tunnel, but ultimately the shape determines the level of stress concentration on the boundary of the tunnel. The comparison of stresses and the strength of the rock determine the stability of the tunnel at specific locations.

In tunnels with long service life, rectangular-shaped tunnels are avoided; or soon, the corners will start to collapse. This is due to concentration of stresses in these areas, which can reach levels that are many times larger than original in-situ stresses, which existed before opening the tunnel. The concentrated forces on corners are usually compressive and their magnitudes could exceed the strength of the rock.

Figure 2 shows stress distributions on the boundaries of three shapes representing three structures of similar sizes. All three are subjected to a 150 MPa far-field force. The first shape (Figure 2A) is rectangular prism, 5 m wide and 20 m long. The second shape (Figure 2B) is a 5 m in diameter and 20 m cylindrical body, and the third figure (Figure 2C) represents an elliptical body, 5 m wide and 20 m long. Stress distribution is calculated using the Examine 2D (Rocscience, 2008) program, which is based on boundary element method. The height for all three

30 Int. J. Phys. Sci.

Figure 2. Stress distribution around three different shapes.

Ateş 31

(a) Circular opening without support shortly after opening (crumbling at the roof and floor)

(b) Circular opening sometime after opening (c) Elliptical opening (no failure)

Figure 3. Shape dependant stability conditions (a and b) Circular (c) Elliptical (Pictures from the Underground Research Laboratory, Pinawa, Manitoba, Canada).

shapes is perpendicular to the paper.

As displayed in Figure 2, the maximum compressive stress concentration generated on the surface of the rectangular prism-shaped structure is 340 MPa, and maximum tensile stress concentration is 20 MPa. In comparison, maximum compressive and tensile stresses generated on the surface of cylindrical structure are 330 and 30 MPa, respectively. The elliptical shaped structure is faced with the least amount of stresses: 280 MPa compressional and 0 MPa tensile.

The effect of shape on structural stability is demonstrated in real case tunnels as well (Read, 2004). In this case, two sets of tunnels are excavated at 420 m depth in the Underground Research Laboratory of the Atomic Energy of Canada (Pinawa, Manitoba, Canada) in the same stress environment (The highest horizontal forces are around 65 MPa and the vertical forces are around 27 MPa - σ1/σ3 ratio is 65/27 = 2.4) and in the

same orientation: One circular and one elliptical. The only significant difference between two tunnels is that the first tunnel is excavated in circular form, while the second tunnel is excavated in elliptical form. The ratio of dimensions of elliptical tunnel are selected in similar proportions (width/height = 2.4) as the in-situ stress ratio; an aspect which is critical in stability of tunnels (Ateş and Baumgartner, 1995). In the circular tunnel, the roof of the tunnel is fractured immediately after opening (Figure 3a). Similar fracturing occurs in the floor of the tunnel. This fracturing process continues and eventually develops into notches as shown in Figure 3b. On the other hand, a ‘smoother’ stress distribution, and therefore less local stress concentration, is obtained by using an elliptical shape in the same location. Figure 3c shows the elliptical-shaped tunnel in the same location. It can be seen that there is no initial fracturing ‘notch’ type failure, which occurred with the circular-shaped tunnel. These

32 Int. J. Phys. Sci. analyses and experiments show that certain shapes are inherently better able in distributing on-coming forces around structures than others. Example 3: Shape and survivability of classic structures under harsh conditions There are several classic structures, which have shapes that can manage severe stresses and survive long times. Tents, pyramids, and domes are among these structures that have long endured and survived harsh environments (Figure 4). An examination of these shapes reveals that the most common feature among them is their inherent ability to disperse forces. Other structures of interest, classic or modern, are: A bridge pier, which is shown to illustrate its shape and forces acting on it represented by action of water around it. A rectangular bridge pier would have no chance ‘resisting’ the active forces of the environment in many places, for example, Confederation Bridge – Canada (2004). This bridge is the longest bridge in the world to span ice covered waters of Northumberland Strait in Eastern Canada. There is ice covering the strait for five months of the year. A special design was made to accommodate this, the bridge designers developed a 52 degree conical ice shield located on the pier shaft to break up the ice. This ice shield actually lifts the ice flow up, so, it breaks on its own weight. The shape of tall towers is also of interest, because they endure complex forces, such as turbulences (wind/air resistance) similar to the earthquake generated surface waves. Examples provided in Figures 4 depict these structures and illustrate the paths of earthquake waves acting on them. All of these structures cannot withstand in their harsh environments had it not been for their inherent ability to disperse the oncoming forces. DISCUSSION Almost after every earthquake, there is an argument among the public, the media, and experts about why certain buildings survived the earthquakes and others in the nearby localities did not. The immediate blame usually goes to the use of sub-standard materials (or not enough use of certain materials- e.g., steel bars), or that the buildings were not constructed according to the latest design codes (Ekici, 2000). This may be the case in some instances. However, the issue of the use of sub-standard materials or not following the building code is a legal and/or ethical issue. There are numerous cases where the buildings are designed and constructed according to the latest codes and yet still collapse (Homayun, 2000; Lakkas et al., 2002). This is a clear sign that there are gaps in the current design philosophy

Figure 4. Force managing structures A) Tent, B) Pyramid, C) Bridge pier, D) Tower, E) Dome. An on-coming force (such as the force of destructive surface waves created by an earthquake) is depicted to have been dispersed at the point of impact by the shape of the force- managing structures. Note that the same force must be faced totally by the force resisting structure (Figure 1, bottom), making the survival of the force-resisting structure difficult or, depending the force, impossible.

and that new and practical approaches are needed. The ‘new’ does not have to be complex or revolutionary, but needs to consider fundamentals and can be as simple as possible.

The multi-disciplinary examples and analyses provided in this study demonstrate that if the force-management approach, as opposed to the resist-full-force approach can be adopted in early design phase whereby the shape of buildings can be designed to disperse the earthquake

forces, then, the building safety is dramatically improved. The recommended approach is simple and may be economical as well. The complexities introduced to modern buildings and the client's desire for swift use of facilities (Austin et al., 1996) may have forced designers to overlook these benefits.

The shape of building can be designed such that the earthquake waves are re-directed to minimize the stress concentration at certain locations in the building. In most cases, it is the maximum stress, not the average stress, which causes the failure to initiate. If the initiation of failure can be delayed and/or avoided, then the collapse of building can be avoided and/or delayed and the chances of saving lives during an earthquake increases. Certain shapes will try to ’resist’ or fight the oncoming forces by facing the on-coming force in full, while other shapes ‘manage’ the forces by channeling them appropriately around them first, and resist only the remaining ones that they cannot redirect.

The ‘full resist’ mode requires an all around resistive building, while the ‘manage first’ mode requires a building shape that helps to divert the forces of earthquake. The ‘full resist’ mode requires a building that is ‘strong’, thus designed to demand the maximum strength from its materials. This is a very challenging task at the best of the circumstances. Besides, meeting this demand often requires the use of stiff materials at the expense of flexible materials. The ‘manage first’ mode helps to provide pathways to divert the forces around and away from the building. Because the forces that a building sees are less in this case, there is a much greater/higher chance of the building’s survival. This mode of design should also allow more flexible materials than the traditional stiff and ‘strong’ designs.

In this study, only simple examples are provided and basic analyses are carried out. Also, the topic crosses the areas of many disciplines. This, on a detailed level, introduces some short-comings. The work presented is intended to be at a conceptual level. Thus, the analyses are static and/or pseudo-dynamic as opposes to dynamic wave analyses. The concepts should still apply; and dynamic analyses employing earthquake waves can be carried out separately. Also the cost comparison of building in various shapes vs. the cost of off-the-shelf design is not done. However, manufacturing stronger and stronger building elements – required if the force-management-approach is not employed- to withstand a large earthquake is also a very challenging and costly task. Thus adopting the force-management approach can be more economical. Conclusions The real case, multi-disciplinary, examples and analyses provided in this study demonstrate that the approach of

Ateş 33 managing earthquake forces provides more safety. Therefore, this concept should be used more often and more consciously by all disciplines involved in the overall design of structures, specifically in the earthquake-prone zones. As an added benefit, this approach can also improve and diversify the overall landscape of some urban scenes where the rectangle appears the most dominant shape.

REFERENCES

Adams RD (2006). The Haicheng, China, earthquake of 4 February

1975: The first successfully predicted major earthquake. Earthquake Eng. Struc. Dyn., 4: 5.

Aki K (1989). Ideal probabilistic earthquake prediction. Tectonophysics, 1-3(10): 197-198.

Ambraseys NN, Finkel CF (1995). The Seismicity of Turkey and adjacent areas (A historical review, 1500-1800). Published by Muhittin Salih Eren, Eren Publications, Beyoğlu, Đstanbul, 34-95-Y-70-061. ISBN 975-7622-38-9, p. 240.

Atabey E (2000). Earthquakes (in Turkish). MTA General Directorate. Educ. Ser., No. 34.

Ateş Y, Baumgartner P (1995). Preliminary excavation design under thermal-mechanical stress conditions. In: Proceedings of the Second International Workshop on Design and Construction of Final Repositories. Held in Winnipeg, Canada, February 15-17, 1994. Theme: Factors influencing repository design and layout. Compiled by Gary R. Simmons. Atomic Energy of Canada Limited Report, AECL - 11480.

Ateş Y, Ridgway WR, Bruneau D (1994). An evaluation of the potential effects of seismic events on the used fuel disposal vault. Atomic Energy of Canada Limited Technical Record TR 623, COG-94-257.

Austin S, Baldwin A, Newton A (1996). A data flow model to plan and manage the building design process. J. Eng. Des., 7(1): 3-25.

Bolt BA (2003). Earthquakes. WH Freeman, 5th ed. Coburn A, Hughes R, Pomonis A, Spence R (1995). Technical

principles of building for safety. Intermediate Technology Publications.

Confederation Bridge (2004). http://bridgepros.com/projects/Confederation%20Bridge/. Last visited 02/06/2004.

Ekici O (2000). Cumhuriyet (Istanbul,Turkey) Newspaper articles about the ‘The International Conference on the Seismic Performance of Traditional Building, Istanbul, Turkey, Nov 16-18. Cumhuriyet, November 22.

Gülhan D, Güney IO (2000). The behaviour of traditional building systems against earthquake and its comparison to reinforced concrete frame systems. Experience of Marmara earthquake damage assessment studies in Kocaeli and Sakarya. The International Conference on the Seismic Performance of Traditional Building, Istanbul, Turkey, Nov. 16-18.

Gülkan P (2005). An analysis of risk mitigation considerations in regional reconstruction in Turkey: The missing link. Mitigation Adaptation Strategies Global Change, 10: 525–540.

Hagiwara Y (1974). Probability of earthquake occurrence as obtained from a Weibull distribution analysis of crustal strain. Tectonophys. Focal Process. Prediction Earthquakes, 23(3): 313-318.

Homayun A (2000). The role of architects in seismic design. In: The International Conference on the Seismic Performance of Traditional Building, Đstanbul, Turkey, Nov. 16-18.

James J (2006). Fatal attraction: Living with earthquakes, the growth of villages into megacities, and earthquake vulnerability in the modern world. Phil. Trans. R. Soc. A., 15(364): 1911-1925.

Keilis-Borok VI, Rotwain IM (1990). Diagnosis of time of increased probability of strong earthquakes in different regions of the world: Algorithm CN. Phys. Earth Planet. Interiors, 61(1-2): 57-72.

34 Int. J. Phys. Sci. Lekkas EL (2002). The role of earthquake-related effects in urban

complexes, Nat. Hazards, 25(1): 23–35. Mollamahmutoğlu M, Kayabalı K, Beyaz T, Kolay E (2003).

Liquefaction-related building damage in Adapazarı during the Turkey earthquake of August 17, 1999. Eng. Geol., 67(3-4): 297-307.

Read RS (2004). 20 years of excavation response studies at AECL’s Underground Research Laboratory. Int. J. Rock Mech. Min. Sci., 41: 1251-1275.

Rocscience (2008). Examine2D – 2D Stress analysis for excavations. Rocscience Inc., Toronto, Canada.

Sezen H, Whittaker AS., Elwood KJ, Mosalam KM (2003). Performance of reinforced concrete buildings during the August 17, 1999 Kocaeli, Turkey earthquake, and seismic design and construction practise in Turkey. Eng. Struct., 25(1): 103-114.

Sharma S, Judd WR (1991). Underground opening damage from earthquakes. Eng. Geol., 30(3-4): 263-276.

Shedlock KM, Giardini D, Grünthal G, Zhang P (2000): The GSHAP

global seismic hazard map. Seismolog. Res. Lett., 71(6): 679-689. Spence R (2007). Saving lives in earthquakes. Successes and failures

in seismic protection since 1960. Bull. Earthquake Eng., 5: 2. Uitto JI (1998). The geography of disaster vulnerability in megacities. A

theoretical framework. Appl. Geogr., 18(1). Vere-Jones D (1970). Stochastic models for earthquake occurrence. J.

Royal Statistical Soc. Ser. B (Methodol.), 32(1): 1-62. Wetmiller RJ, Cajka MG, Lodha GS, Ateş Y (1996). Comparison of

seismic ground motions at surface and underground for understanding/developing design of nuclear fuel waste repositories. In: Proceedings of the 1996 International Conference on Deep Geological Disposal of Radioactive Waste. Winnipeg, Canada September 16-19.