ACCELERATION SPECTRA FROM GISBORNE - 2ZG SITE Figure 1: Three component spectra for Gisborne 2ZG record The ground motion record from the above station has been reduced to motion spectra in the E/W orientation (primary component PGA=0.25g) and the N/S orientation (secondary component PGS 0.2g) with the vertical spectral response also presented for completeness). (Note: the orientation selected are not necessarily the most severe from any orientation – work is ongoing on this aspect). The location of the recorder is two blocks south of the CDB and approximately 3 blocks east of the river. Site soil classification is considered to be Class D, although it is close to being Class E). Notable Features the sharp spikes present at around 0.35 seconds in the E/W component where Sa(0.35)1.15g. A less intense spike occurs in both records at approx T=0.5 sec with components Sa(0.5)0.87g E/W and Sa(0.5)0.93g N/S. Between the period ranges T=0.25 and T=0.7 both records indicate acceleration spectral levels generally in excess of 0.6g. Between T=1 and T=1.5 second, acceleration levels fall from 0.2g to 0.15g and rapidly continue to decline into the higher period ranges. Recorded vs Code 5% Damped Deep Soil Spectra 1.2000 1.0000 SA(T) (g) 0.8000 2ZG EW 2ZG NS NZS1170 Deep Soil Z=0.36 R=1 NZS1170 Deep Soil Z=0.36 R=0.67 NZS1170 Deep Soil Z=0.36 R=0.5 0.6000 0.4000 0.2000 0.0000 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 Period T(s) Figure 2: Recorded vs Code The two components of the recorded record are compared with the Deep Soil ground motion spectra with R=1 (1/500), R=0.67 (1/150) and R=0.5 (1/100) (Note bracketed figures representing annual probabilities or exceedence). Notable features: Within the period range T=0.25 to T=0.8, the measured response components are reasonably aligned with the R=0.67 ground motion projection, indicating that the motion could be considered to be approximately coincident with motions associated with a 1/150 annual probability or exceedence or motions approximately 2/3 those of ULS design level. The Sa(0.35) peak value is exception in that it approaches the 1/500 design value although is limited to a very narrow period band. Beyond T=1 second motions approximately align to R=0.25 (1/25) being levels of motion at which the onset of damage could be expected (while recognising that few elements and components will experience significant excitation over that period band. Recorded vs Modelled 5% Damped Deep Soil Spectra 1.2 Magnitude 6.9 Intraslab earthquake Epicentral Distance 47 km Depth 44 km Hypocentral distance 64 km 1 SA(T) (g) 0.8 Deep Soil 50-percentile Deep Soil 84-percentile 0.6 2ZG EW 2ZG NS 0.4 0.2 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 Period T(s) Figure 3: Recorded vs Modelled The two component records are compared with those projected by running the event (location, depth, magnitude and earthquake type) through the NZ seismic hazard model. Notable features: The recorded accelerations match relatively closely the 84% deep soil projections in the short period range T<0.4 seconds, exceed the projected motions over the mid period range 0.4<T<1 second and are below projected motions for the long period range T>1 second. The stronger motions within the mid-period range are abnormal and not included in the model Recorded vs Hazard 5% Damped Deep Soil Spectra 10 10yrs 25yrs 50yrs 100yrs 250yrs 500yrs 1000yrs 2500yrs 5000yrs 10000yrs 2ZG EW 2ZG NS SA(T) (g) 1 0.1 0.01 0.01 1 0.1 10 Period T(s) Figure 4: Recorded vs Hazard The recorded motions have been superimposed over various hazard spectra for visual comparison. Notable features: The recorded values are deficient in both short (T<0.4 sec) and long (T>1 sec) ranges, but relative strong within the mid-period range where motions align approximately with motions expected from an event with an annual probability of exceedence of 1/250.