FLOOD09139 - Laserfiche WebLink

Home Browse Search

design earthquake, distance of the site from epicenter of the earthquake, attenuation characteristics of the earthquake to the site, and the source mechanisms of the earthquake (type of fault, depth, etc.). The current library of strong-motion records, in most cases, is not sufficient to complete ly consider all of the abOlTe fac tors, particularly the attenuation charac teristic s and the source mechanisms. Site-dependent response spectra have been developed for various conditions including foundation conditions, earthquake size, and attenuation effects. Site-dependent spectra will generally produce ground- motion parameters that correspond better with those expected on the basis of the seismological and geologic conditions at the site. The limiting factor is that source mechanisms and the attenuation charac teristic s to the site may not be fully represented by the available data. Also, the nuuber of relevant ground motion records may not be sufficient to perform a statistical evaluation. Therefore, care should be exercised in developing a site-dependent re sponse spec tra. 4. Acceleration Time-Histories The acceleration time-histories used in the analysis are usually selected in conjunction with the site-independent or site-dependent response spectra. The time-histories may be real earthquake records or synthetic (mathematical models) records whose response spectra closely represent the site- independent or si te-dependen t re sponse spec tra. The re sponse spectra from the time-histories may envelope site response spec tra or they may repre sent an average; this depends on how the site response spec tra were developed and the amount of conservatism desired in defining the time-histories. The synthetic earthquake rec ord s are developed from mathematical models that use white noise, filtered white noise, and sta tionary and nonsta tionary fi ltered white noise or theoretical seismic source models of failure in the fault zone to genera te time-histories. Selection of real earthquake records to represent the site acceleration time-histories is usually preferable to using synthetic earthquakes. Sometimes, this is not possible due to lack 0 f rec orded da ta . Ac tual and synthetic earthquake time-histories may be modified and/or combined to form time-histories which contain desired ground motion charac terist ic s 0 f peak accelera tion, dura tion, and frequency content. Accelerations can be increased or decreased by scaling the amplitudes, i.e., by llDlltiplying by the ratio of the peak acceleration desired to the peak unmodified acceleration. Frequency content can be modified by -16-