Figure 6. (a) Map of maximum contraction directions from GPS data (Haines & Wallace, 2020) and borehole-derived SHmax orientations (orange arrows) and IODP borehole image analysis from McNamara et al. (2021) (blue arrows). (b) Map of SHmax orientations and inferred faulting regime from focal plane mechanisms (0-60 km; square: extensional, circle: strike-slip, and triangle: compressional regime; Townend et al., 2012), seismic anisotropy measurements (0-40 km; black lines). Dashed blue lines show depth of subduction interface from sea level (Williams et al., 2013). (c) The map showing depth to basement map of North Island adapted from FROGTECH (2014). The clockwise rotation of SHmax orientation along the HSM strike follows the basement topography. Black arrows indicate the long-term motion of the Pacific relative to Australian Plate from DeMets et al., (1994) and grey arrows show shortening rates at the Hikurangi Trough from Wallace et al. (2012). Red dots show the borehole locations.
Although borehole-derived SHmax orientations at the HSM are broadly consistent with maximum contraction directions, SHmax orientations are not uniform within the forearc and in places deviate from the observed maximum contraction directions (Figure 6a). Some of these variabilities may be due to longer-term partitioning of oblique relative motion and strain into margin-parallel, margin-perpendicular, and strike-slip components. Similar deviations from maximum contraction directions are observed within the Nankai subduction zone (Chang et al., 2010) and Costa Rica margin (Malinverno et. al, 2016). For example, borehole-derived SHmaxorientations in the accretionary prism across the Nankai subduction zone are subparallel to the plate convergence vector between the Philippine Sea plate and Japan, perpendicular to the plate boundary, and for one borehole in the shelf edge is parallel to the plate boundary (Lin et al., 2010; Chang et al. 2010; Wu et al. 2012). The variations of stress among these sites and the deviation of borehole-derived SHmax orientations from the plate motion vectors derived from GPS data are suggested to be attributed to the strain partitioning into trench‐parallel, right‐lateral slip and thrust tectonics due to oblique plate convergence (Tobin et al. 2009; Lin et al. 2010). Alternatively, such stress variations have also been attributed to other factors such as thrusting and bending within individual geologic domains and local extension deformations derived by gravitation collapse of the prism in the forearc (Chang et al., 2010; Lin et al., 2016).
Townend et al. (2012) derived SHmax orientations from earthquake focal mechanisms using a Bayesian approach for earthquake data between 2004-2011 with ML> 3. Their observations reveal that SHmax orientation changes from NE-SW in the central HSM to a more oblique SHmaxorientation, sub-parallel to Pacific-Australia relative plate motion, in the southern HSM (Figure 6b). Only two focal mechanism solutions are located within the upper plate of the HSM or near the subduction interface; one strike-slip event with a NE-SW SHmaxorientation (latitude 38°S; North of central HSM) at 7 km depth, and one compressional event with a WNW-ESE SHmax orientation (latitude 40.5°S; southern HSM) at 25 km depth (at the top of subducting plate near the subduction interface; Figure 6b). In the central HSM (where the plate boundary is largely creeping) borehole-derived NE-SW SHmax orientations and maximum contraction directions agree with the NE-SW SHmax orientations derived from focal mechanism inversions within the upper plate and subducting plate, implying that far-field plate boundary forces exerted at the HSM primarily control the present-day crustal stresses in the upper plate in this area. In contrast, the southern HSM shows stress field rotation with depth from borehole-derived WNW-ESE/NW-SE SHmaxorientations (roughly margin-perpendicular) in the shallow crust of upper plate to NE-SW/ ENE-WSW focal mechanism-derived SHmax orientations within the subducting plate (Figure 6b). The NE-SW/ ENE-WSW SHmax orientations derived from focal mechanism inversions indicate a normal or strike-slip tectonic regime within the subducting plate of the southern HSM. There is one compressional event at ~25 km depth (near the subduction interface) where the borehole-derived SHmax orientation agrees with a focal mechanism-derived WNW-ESE SHmaxorientation (latitude 40.5°S; Figure 6b). This implies that the overriding plate broadly is in a condition of horizontal compression parallel to convergence direction, potentially as deep as the plate interface (±10 km) (barring localized variations) and the tectonic regime becomes more strike-slip and normal within the subducting plate. The change in stress orientation with depth in the southern HSM is more compatible with a theoretically mechanically weak plate interface with low resolved shear stress and normal effective stress on the plate interface. The fact that the upper plate is in the state of horizontal compression parallel to plate convergence, on the other hand, implies that contemporary SHmax orientation is being driven by subduction of Hikurangi Plateau beneath North Island. Taking these observations into consideration, we propose that plate interface remains mechanically weak along the whole HSM and that a small increase of elastic compression strain and stress caused by the interseismically coupled southern HSM interface is sufficient to rotate SHmax orientation in the upper plate toward the subduction direction.
More recently, Illsley-Kemp et al. (2019) measured seismic anisotropy using shear wave splitting fast orientations from earthquake data in 2005–July 2018 with ML>0. Some studies consider fast orientations as a proxy for SHmaxorientation if other significant crustal anisotropies are not present (e.g. fracturing, faulting, grain and crustal preferred orientations). The seismic anisotropy was derived from events ≤40 km deep and fast orientations were dominantly NE-SW (margin-parallel) for most of the HSM forearc, and more ENE-WSW in the northern and central HSM (Figure 6b). With the exception of some portions of the central HSM, the fast directions from shear-wave splitting studies do not agree well with the SHmax orientations we measure in the shallow boreholes. This could mean either (a) that the shear wave splitting fast directions are not controlled by contemporary SHmax orientation, or (b) that the fast directions are mainly controlled by structural features such as faults and the stress state at greater depths, which may be different than the near-surface which our study samples. In fact, propagating waves sample the volume of crust above the hypocenter without taking the variation of anisotropic with depth into account.
Long-term tectonic deformation resulting from rapid clockwise rotation of the Hikurangi forearc (Wallace et al. 2004) may also possibly explain the along-strike variation in SHmax orientation. The clockwise rotation of forearc, which accommodates the margin-parallel component of oblique Pacific-Australian plate motion, results in significant tectonic transitions along the strike (Nicol et al., 2007; Wallace et al., 2004). Tectonic regime varies from back-arc rifting within the TVZ in the central part of North Island, strike-slip and normal faulting onshore of the northern and central HSM to transpression and reverse faulting at the southern HSM (Figure 5). The rotation of SHmax orientations from margin-parallel in the onshore central HSM to margin-perpendicular in the southern HSM likely reflects this transition in the tectonic regime, Such that the NE-SW SHmax orientation in the central HSM reflects margin-normal extension due to the transmission of slab rollback forces across the forearc and into the adjacent extensional back-arc rift (Wallace et al., 2012b). The clockwise rotation of the Hikurangi forearc may also explain why our observed contemporary SHmaxorientation of NE-SW is not consistent with active large-scale NE-SW striking compressional faults in the central HSM. The imaged intervals of the central HSM boreholes are located within the hanging walls of active NE-SW striking compressional faults and within NE-SW oriented fault-bend fold axes identified from seismic reflection profiles (Western Energy New Zealand, 2001; Barnes et al., 2002). It is possible that the deformation and growing rate of these reverse faults within the central HSM have changed through time, allowing forces exerted by forearc rotation surpass the NW-SE compression stress on these active faults. In this case, the rotation of the forearc, which accommodates the margin-parallel component, induces a NE slip motion on these pre-existing faults. This also suggests that these reverse faults in the context of current stress field could be currently active as strike-/oblique-slip faults. Another possibility is that due to frequent earthquakes in this region, the SHmax orientation within the central HSM has changed from a potential NW-SE orientation perpendicular to the margin, compatible with the orientation of observed reverse faults and fold axes, to the contemporary NE-SW orientation subparallel to the margin. During the interseimic period, strain and stress accumulation along the reverse faults and subduction interface drives compression in the upper plate parallel to convergence direction. Following great earthquakes or frequent earthquakes, the margin-perpendicular component is reduced due to post-seismic stress release, allowing a strike-slip motion in direction of long-term oblique relative plate motion on these faults. This implies that the stress state associated with seismicity perturbations is quite long lasting (these boreholes are drilled over the course of 4 years) such that stress state require a long time to recover or the perturbations were significant enough to permanently reorient the stress. In this scenario, the intermediate principal stress (Shmin) and minimum principal stress (Sv) would switch and principal stresses re-orient in response to forces exerted by long-term relative plate motion and fluctuations in the magnitude of stress parallel to plate convergence, modulated by the seismic cycles. Sacks et al. (2013) and Moore et al. (2013) used a similar concept to explain the contrast between SHmax orientations and normal fault strikes within Kumano Basin of the Nankai subduction zone.
Lateral variations in surface and basement topography may also play a role in the observed along-strike variation in upper plate borehole-derived SHmax orientations (Figure 6c). Basement structures such as faults, folds, and seamounts that are not covered by thick sediments can cause significant stress rotation at both regional and large scales by introducing geomechanical inhomogeneities and lateral discontinuities (Gale et al., 1984; Enever et al. 1999; Rajabi et al., 2016a). In fact, the presence of seamounts on the incoming plate have been suggested as a possible localized effect on stress field orientations and shear-wave splitting fast directions in the offshore northern HSM (McNamara et al., 2021; Zal et al., 2020). The SEEBASE™ model of New Zealand, a depth to basement map based on the integrated interpretation of gravity, magnetic, borehole, and seismic data (Figure 6c) clearly shows a significant change in sediment thickness overlying the basement rocks, and basement topography along the HSM margin (FROGTECH , 2014). The onshore northern and central HSM basement is located at depths ~9-15 km and covered by thick Neogene sediments, whereas at the onshore and offshore southern HSM the Mesozoic greywacke basement is uplifted and located at relatively shallow depths (<5 km), and in some locations above sea level. Thin sediment packages directly overlying mechanically strong basement rocks, may be more reflective of the tectonic stresses experienced by the basement. Hence, it is possible that the basement compressional structures striking NNE/NE-SSW/SW control the observed margin-perpendicular, NW-SE SHmax orientation within the shallow crust of the southern HSM.
In addition to large-scale stress variations, localized variations have been observed between and within individual boreholes which could be attributed to local active geological structures such as faults, surface topography, and variable mechanical properties. For instance a localized NW-SE SHmax orientation (144° ± 11°) is noted from four-arm caliper data in borehole Whakatu-1 (south of Napier; Figure 5), perpendicular to the dominant NE-SW SHmax orientation of the central HSM. The NW-SE SHmax orientation in Whakatu-1 is margin-perpendicular implying a localized compressional tectonic regime at this locale, which agrees well with observed onshore and offshore active, NNE-SSW striking, compressional faults in this region (Figure 5; Litchfield et al. (2014); Langridge et al, (2016)), such as the reverse fault responsible for growth of the Cape Kidnapper’s anticline (Hull, 1987). Furthermore, Dimitrova et al. (2016) reported a possible locked patch in southern central HSM between shallow SSEs on the East Coast and deep Kaimanawa SSEs west of the East Coast (Figure 1b). Downdip changes in interseismic coupling behavior of the subduction interface and the accumulated strain associated with the locked patch could also be influencing the compressional tectonic regime in this region and borehole Whakatu-1.
6 Conclusions
This paper presents the first comprehensive analysis of contemporary SHmax orientations along the HSM, and discusses stress field orientation variability within the context of variable tectonics and slip behavior of this subduction margin. SHmaxorientations in the central HSM are predominately NE-SW (sub-parallel to plate boundary), which rotates to a dominantly WNW-ESE SHmax orientation in the southern HSM (approximately perpendicular to the plate boundary). Our borehole-derived SHmax orientations agree with the maximum contraction strain directions from GPS measurements along the HSM suggesting that the observed stress orientations along the HSM are influenced by elastic strain accumulation due to interseismic coupling on the Hikurangi subduction interface. The long-term tectonic deformation arising from rapid rotation of the Hikurangi forearc, causing reverse faulting and strike-slip in the southern part of the margin and a combination of extension and strike-slip in the northern and central margin and the basement topography may also be at play in influencing the along-strike variations in observed stress orientations. In the southern HSM, borehole-derived SHmax orientations are inconsistent with SHmax orientations derived from focal mechanism solutions in the subducting plate, implying that the southern HSM interface is mechanically weak. Further interpretation of HSM stress state could be achieved by constraining stress magnitudes, which will be the focus of future studies.