The surface of the earth may be divided into zones each characterised by one of three grades of seismic risk. In the first, earthquakes have been observed to occur either very rarely or never, in the second earthquakes giving some damage occasionally occur and in the third damaging earthquakes frequently occur and become a part of life. The second gives the greatest difficulty for not only is the seismic record likely to be incomplete but the public may not perceive earthquakes as a hazard requiring special, and possibly expensive, engineering precautions; only an “unexpected” and possibly damaging earthquake may sharpen that perception.
However, having established that a hazard exists, engineering design may be employed to mitigate the damage that could result. Every building rests on a foundation and static foundation designs may be modified to allow for dynamic loading. One approach in Japan (Okamoto 1973) has been to divide the country into areas with particular seismic coefficients which relate to expected intensities of shaking. In Table 1 the seismic coefficient k is 0 for static conditions rising to a maximum of 0.4 for maximum shaking. The Terzaghi bearing capacity factors are thence reduced as k increases to give larger area footings (and thus less bearing pressure) to compensate for increased loading under dynamic conditions. The reader can be forgiven for feeling this all sounds too simple a solution and the author admits that here only the basic ideas of 30 years ago are described as at that time the concepts were simple enough for most engineers to understand them.
Earthquake engineers can be said to design against “the last earthquake”, for each new event may reveal the success or failure of some previous approach. For example, one idea to combat damage from liquefaction was to build on piles to a bearing stratum below the liquefiable layer. Hamada et al. (1987) described how a building, damaged in the 1964 Niigata earthquake, was 20 years later scheduled for reconstruction.
As part of this work the old pile foundations were excavated to examine their condition.
Out of a total of 304 piles, 74 were excavated and most of these were found damaged, showing fracture displacements in a direction corresponding to that of the permanent ground movements and suspected to be a consequence of liquefaction. Clearly, piled foundations alone are not the answer to liquefaction; if liquefaction induces lateral movement piled foundations may be damaged and even if this does not occur, the loss of lateral support to the pile in the liquefied layer may cause damage as piles and building shake.
Greater benefits can be obtained by improving the ground where possible so that it is not weakened by shaking. Paradoxically, the most commonly used technique in sands and silts is to increase their relative density by giving them a good shake, where necessary induced by dynamic loading! The objective is to attain a relative density which safeguards the deposit against liquefaction from the earthquake the structure is designed to withstand. This might be achieved by the process of vibroflotation, in which a vibrating cylinder, the vibroflot, is jetted into the ground, and then vibrates the ground so expelling water from its pores and enabling the material around it to consolidate and strengthen. In the 1968 Off-Tokachi earthquake, which gave MM intensity IX at the site of the Hachinohe paper mill, surrounding untreated ground cracked and liquefied, while buildings on the consolidated ground suffered only light damage (Okamoto 1973). Ground improvement may be monitored by in situ testing using the Standard Penetration Test, the static cone test or the Menard pressuremeter, and geophysical methods. The seismic design of engineered structures is continually improving to accommodate the most recent experiences.