The kinematics of the plate boundary zone through New Zealand: a comparison of short‐and long‐term deformations

Summary Surveys of shear strain rates derived from repeated triangulation cover about one‐quarter of the area of the plate boundary zone through New Zealand and are well distributed within it. Using a method developed by Haines relating shear strains to the total strain and velocity fields, the dilatational and rotational strain rates and the velocities within the deformed zone are derived. The plate boundary zone is shown to be two‐dimensional, for the most part, in that gradients of the velocity along the length of the zone are small or zero. The shear strain rates are not uniform, however; they reach a maximum along a central axis and decrease gradually to zero toward the sides. Nor are the two independent components of strain, one involving shortening (or extension) perpendicular, and the other shear parallel, to the zone equally developed everywhere; there is a clear geographic separation with shortening developed over the subduction thrust and shear developed behind in a zone of wrench faulting. Time variations in the strain field in part of the region above the subduction interface are related to changes in the degree of coupling across the interface. A velocity field is computed: the relative velocity of the two plates from this data agree very well with that derived from seafloor spreading data. These strain and velocity fields derived from geodetic studies give the short‐term kinematics of the plate boundary zone. The long‐term kinematics of the zone are estimated from: (1) uplift and subsidence data which are related to changes in crustal thickness and hence to dilatational strain, and (2) variation in the declination of primary magnetization in rocks of known age, related to the rotational strain. The rate of rotational strain has increased in time from about 2 to 3° Myr ‐1 , 10 Myr ago, to about 9° Myr ‐1 today. The increase can be understood as a secondary effect of a developing compressional component increasing the coupling between the plates so that more of the motion parallel to the zone is accommodated by distributed shear and less by aseismic slip. The rates of strain averaged over the last 1 Myr or so, and at a scale length of a few tens of kilometres, are closely similar in distribution and magnitude to the strain rates derived from short‐term kinematics. While, therefore, at scale lengths of a few hundred metres or so faulting is very important and the deformation is discontinuous with unstrained blocks faulted, with large but uncertain slip, against adjacent blocks, the deformation at these longer‐scale lengths is continuous and regular. The two consequences of this study are that: (1) strain from geodetic data can be directly related to the average long‐term deformation and therefore can be an important tool in the study of deformation in regions of current tectonic activity, and (2) at scale lengths of a few tens of kilometres the deformation of a region can be described as the deformation of a continuum and we can ignore the effects of faulting.