|
NORSAR provides services in computing seismic hazard along state-of-the-art principles. Essentially this involves computation of annual exceedance probabilities for given ground motions at a given site. Some features of the probabilistic computational method are provided below.
The computation of earthquake hazard is essentially a process in which one integrates the contribution at a particular site from earthquakes at different distances from the site. This requires, in addition to the source model, which delineates the characteristics of the expected occurrence in time and space of future earthquakes, detailed models for how the seismic waves that are generated at the source will attenuate with distance, for different types of ground motion (often PGA or PSV) and for different frequencies.
The variability in terms of scatter around the mean values for many of the parameters can be modeled realistically in a probabilistic way (Cornell, 1968). This is done through the so-called probabilistic hazard computation.
The model is often computed using a logic-tree formulation in which each of the defining parameters of the earthquake activity is modeled with a weighted distribution:
Logic tree formulation
The results of earthquake hazard are often provided in terms of Peak Ground Acceleration for diffferent annual exceedance probabilities:
Peak ground acceleration vs. probability of exceedance
The hazard results may be subject to a subsequent deaggregation analysis aimed at reavealing the factors that have the major determining impact on the seismic hazard.
|
 |
|
|
|
