In the last few years there have been several probabilistic seismic hazard assessments (PSHA) of Adelaide. The resulting 500 year PGA obtained are 0.059, 0.067, 0.109 and 0.141. The differences between the first three are readily accounted for by choice of GMPE, how faults are included and differences in recurrence estimation, with each of these having a similar level of importance. As no GMPEs exist for the Mt Lofty and Flingers Ranges the choices of GMPEs were all based on geological analogies. The choice of at what weighting to include low attenuation, that is a stable continental crust, GMPE was most important. At a return period of 500 year the inclusion of faults was not necessarily significant. The choice of whether the faults behaved with Characteristic or Gutenberg-Richter recurrence statistics had the highest impact on the hazard with the choice of slip rate the next most important. A low slip rate Characteristic fault, while increasing the hazard for longer return periods (i.e. >=2500 years), results in only a minor increase at 500 years. The magnitude frequency distribution b-value for the four studies were 1.043, 0.88, 0.915 and 0.724. For the same activity in the magnitude range of 3.0 to 3.5, the activity level at M 6.0 is an order of magnitude higher for a b-value of 0.724 compared to a b-value of 1.043. This increase in activity rate of larger earthquakes significantly increases the hazard. The average of the first three studies is 0.078{\textpm}0.022 (0.056 {\textendash}0.100) g. This range is reflecting the intrinsic uncertainty in calculating PSHAs where many of the inputs are poorly constrained. The results for the highest hazard level PSHA study (i.e. 0.141g) can be explained by their use of a low b-value (i.e. 0.724).

}, author = {M. Leonard and Ryan D. Hoult and Somerville, P and G. Gibson and D. Sandiford and Helen M. Goldsworthy and Elisa Lumantarna and S. Spiliopoulos} }