Strain accumulation across the central Southern Alps, New Zealand

Abstract

The Alpine Fault is a major New Zealand geological feature. The fault boundary zone either side of the fault trace is deforming and it is this deformation in the central Southern Alps of New Zealand that we have measured and characterised in this project.

The region of interest includes a small network of permanently tracking or continuous GPS (CGPS and semi-CGPS) receivers located 30km south of Fox Glacier that extends between Karangarua (West Coast) and Mt Cook Village (East Coast). The network crosses both the Alpine fault and the main divide of the central Southern Alps. The Southern Alps Geodetic Experiment (SAGENZ) established the network of 12 sites in 2000 and it is the GPS data from the SAGENZ project that has been used to measure the horizontal deformation or more specifically, the strain across this tectonically active region.

In addition, GPS data observed during four epoch campaigns between 1994 and 2002 has been used. This data is part of the Institute of Geological and Nuclear Sciences (GNS) Global Plate Tectonics programme and covers a region of approximately 30,000km2 of the Central South Island (CSI), roughly from the Rakaia River to the Waitaki River.

We have been able to measure horizontal velocities to better than ±1 millimetre per year (mm/yr). At this level of precision it is possible to investigate the velocity variation (ie, velocity field) across the region and hence deduce the strain variation.

The data collected from the permanently tracking GPS sites has demonstrated that there are significant localised site displacements that appear to be induced by seasonal changes in the environment. The amplitude of the site displacements at several sites is as great as 5-8 mm, the largest is in the order of 20 mm. Although these displacements are not a direct function of the underlying tectonic forces, they do demonstrate the existence of localised train variation and how dynamic the Southern Alps are.

The strain rates determined from the permanently tracking SAGENZ network were compared with the regional strain rates as determined from the CSI network. Generally the strain rates are in agreement, with the SAGENZ strain rate estimates being an order of magnitude more precise than the CSI rates. Although a direct comparison is difficult to make owing to the different sites used in each network, there are significant strain variations on a regional scale. In particular, from the smaller but greater density SAGENZ network, there is evidence of a zone of extension (which agrees with elastic dislocation theory).

Temporal strain variation was also investigated by determining the day to day strain variation relative to the long term site velocities, after removal of the seasonal site displacements. Although the signal is small, there appears to be times during the year when there is a greater degree of strain variability. For example, the spring to early summer period clearly has greater variability corresponding to an increase in daily temperatures and a corresponding increase in snow/ice melt. Conversely, the winter months tend to have less variability possibly due to snow and ice providing an insulating the underlying rock.

The average strain accumulation is consistent with the regional pattern, but there is evidence of regional variations that will not be observed using campaign measurements. Although the largest strain accumulation is derived from the collision at the plate boundary, there are more subtle strain variations being caused by seasonally induced processes.