Proceedings of the 12th International INQUA meeting on paleoseismology, active tectonic and archaeoseismology

of three recurrence intervals from four events is 396 ± 21 yr. The timing of these turbidites overlaps with ages derived from the trenches and provides some confidence that there are four surface faulting events that can be precisely dated. Very large landslide deposits have also been targeted to provide paleoearthquake shaking proxies. One of these landslides (Timpson’s) occurs southwest of Springs Junction. Logs exposed at the base of the ~5.5 x 106 m3 deposit have been precisely dated, yielding a calibrated age range for death of two trees at 518-538 cal yr BP (Langridge et al., unpublished). Timpson’s Landslide has the same age as the penultimate significant turbidite from Lake Christabel. Collectively these data point toward coeval fault rupture and strong ground motions along this part of the AF-NS. M A R U I A R I V E R AV U L S I O N Examination of the LiDAR data at Marble Hill also highlights an avulsion history for the Maruia River. The highest terrace at Marble Hill (Fig. 2; T6) has channel morphologies that indicate flow direction to the southwest following the current route of State Highway 7 to Springs Junction. The IRSL date CP32 from Pit 3 provides a maximum age for terrace T6. Sometime after this the river avulsed and has geomorphology (risers, channels) consistent with northwest flow to The Sluice Box, which is the current outlet of the Maruia River. Dates within the trenches and pits on terraces 4, 2 and 1 typically post-date the age from T6. The maximum age for cover deposits on T4 is 1185 ± 44 yr BP; and on T2 is 1209 ± 20 yr BP. Displacements observed on T5 to T2 are equivalent. The mean scarp height across these terraces is 1.6 ± 0.6 m and the mean of dextral offsets on risers and channels ranges from 12.0 ± 1.3 m (Langridge et al., 2017). Therefore, it is reasonable to assume that the degradation of terraces from T6 through T1 occurred fairly rapidly because each of T5 through T2 record the same relative amount of displacement, i.e., they were faulted by the same number of paleoearthquakes. In essence, this indicates that the avulsion from southwestward Maruia River flow to northwest flow via The Sluice Box could have occurred following one paleoearthquake rupture and the degradation down to T1 was accomplished by the time of the next paleoearthquake (event IV), i.e., avulsion and degradation within one earthquake cycle. DISCUSSION AND CONCLUSIONS Figure 4 documents the relationships between onfault and off-fault datasets in the Springs Junction area, including potential slip per event relationships at Marble Hill. An obvious outcome from these studies is that the offset terrace features point to paleo-slips occurring during metre- to multi-metre sized coseismic displacements which is supported by lack of aseismic creep observed across ‘Evison’s wall’. The elapsed time back to the MRE, based on the most recent turbidite age (102-183 cal yr BP) is 176-257 years. Slip in the MRE was c. 1.3 m dextral and 0.25 m vertical (Yetton, 2002). These values are subtracted from the mean dextral and vertical offset values across T2 to T5, leaving c. 10.7 m dextral and 1.4 m vertical slip that was