TECTONIC GEOMORPHOLOGY AND PRESENT-DAY TECTONICS OF THE ALPINE SHEAR SYSTEM, SOUTH ISLAND, NEW ZEALAND (NEOTECTONICS, FAULTS).

Abstract

Rates of latest Quaternary slip obtained from stream terraces and glacial moraines displaced by faults of the Alpine shear system vary with space and time. Field measurements yield displacement values for faulted geomorphic surfaces, while the rate of thickening of weathering rinds and changes in soil properties, calibrated at sites of known age, yield age estimates. Precisions are 5-20% from weathering rinds and 15-50% from soil data. The oldest surfaces examined have ages of 15-20 ka and right-lateral fault displacements up to 400-600 m. Latest Quaternary lateral-slip rates are 20-45 mm/yr across the Alpine fault in the Southern Alps. To the northeast slip is distributed across a system of faults in Marlborough. The main faults of this shear system--the Wairau, Awatere, Clarence, Hope, and Porters Pass--have latest Quaternary rates of 5-10, 7-10, 7-9, 20-40, and 4-5 mm/yr respectively. Each fault has undergone a substantial decrease in lateral slip in the last 3-5 ka. Long-term rates of horizontal slip across the Australian-Pacific plate boundary--the Alpine shear system in most of the South Island--are 35-50 mm/yr parallel and 8-25 mm/yr normal. Sums of fault-slip rates exceed these plate motions for the early-middle Holocene, but late Holocene fault-slip rates are less than half the long-term average. Rates of geodetic strain and seismic moment release over the last 50-100 yr approximate the long-period rates in Marlborough but are only half in the Southern Alps. The best explanations of these variabilities are that the proportion of plate-boundary motion accommodated by fault slip changes, or that the rate of motion across the plate boundary varies, perhaps over 5 ka intervals. The first hypothesis is not consistent with the early Holocene rates exceeding the long-term average, but the second hypothesis implies that the last 50-100 yr is a period of renewed high tectonic activity. The second hypothesis is more consistent with the data, and the last 15-20 ka may be the time interval necessary to average out shorter, 5 ka episodic variations in plate-boundary motions.