With conventional force-based seismic design, the most important information for the structural f+Sb>$
engineer has been the effective peak ground acceleration. This number is comparatively r]UF<*$
insignificant in displacement-based or performance-based seismic design, where the key data are qQi\/~Y[:
peak spectral response displacement, and the “corner” period at which this occurs. There appears 2h'Wu
qO
to be a disagreement between seismologists on opposite sides of the Atlantic about these data, m^G(qoZ]
which the structural engineer views with concern, since ductile structural response will often be in lOk8VlH<h
this region of the corner period. Dependable information on the reduction of displacement L]|mWyzT
response with damping or ductility is also needed for design approaches utilizing secant stiffness 6FQi=}O 1
characterization. Other areas needing clarification include the issue of how to develop `om+p?j
accelerograms for time-history analysis compatible with the design seismicity using an acceptably %^pi
small number of accelerograms, and how to reconcile the statistical nature of seismicity <"J]u@|
characterization with the structural engineer’s preference for deterministic analysis. There is some j
0pI
evidence that a consequence of this is invalid averaging of response characteristics by structural uI)twry]@
engineers.