Timing of late Holocene paleoearthquakes on the Hurunui segment of the Hope fault: Implications for plate boundary strain release through South Island, New Zealand

摘要

The Hurunui segment of the dextral-slip Hope fault extends for ∼42 km between the Hope and Hurunui River catchments and westward to Harper Pass at the Main Divide of South Island. We conducted paleoseismic, soil, and landscape dating studies in the upper Hurunui valley to determine the timing of past earthquake ruptures along this geometric fault segment. On a late Holocene alluvial surface at Matagouri Flat, a young channel system is displaced dextrally by ∼4.5 ± 0.6 m, which is attributed to the most recent faulting event. A trench and soil pits excavated nearby yield evidence for the timing of the last two surface-faulting events. These events, dated from a combination of radiocarbon dates, relative soil, tree, and historical constraints, modified within an OxCal analysis, occurred at ca. A.D. 1655–1835 and 1425–1625. The occurrence of two rupture events during the last ∼600 yr is consistent with previous estimates stating a short recurrence interval for the Hurunui segment. The most recent faulting event in our trenches predates the historic 1888 North Canterbury (Amuri) earthquake, which ruptured the Hope River segment to the east. A comparison of the record from the Hurunui segment with paleoseismic records from other segments along the Hope fault zone indicates evidence for two or three rupture events within the last 700–900 yr. Within the age resolution of the current dates, the data suggest that along-strike rupture occurred along the majority of the Hope fault zone between 120 and 360 yr ago (i.e., A.D. 1650–1888), and may all postdate the most recent rupture of the Alpine fault in ca. A.D. 1717. In addition, the timings of penultimate faulting events along the Hurunui, Hope River, and Conway segments and the Hanmer fault all fall in the range ∼400–700 yr ago. These preliminary observations indicate that rupture of segments of the Hope fault, and indeed the Alpine fault, could occur as clusters that are years to decades apart. Rupture of the major plate boundary faults in the northern South Island may therefore occur in sequences that are enhanced, or even retarded, by stress triggering and fault interactions.