Description:

_Journal of Structural Geology 32 (2010) 1101-1113_ Contents lists available at ScienceDirect Journal of Structural Geology journal homepage: www.elsevier.com/locate/jsg Normal fault interactions, paleoearthquakes and growth in an active rift A. Nicol*, J.J. Walsh, P. Villamor, H. Seebeck, K.R. Berryman GNS Science, PO Box 30368, Lower Hutt, New Zealand Fault Analysis Group, School of Geological Sciences, University College Dublin, Belfield, Dublin 4, Ireland Article info Article history: Received 22 December 2009 Received in revised form 31 May 2010 Accepted 25 June 2010 Available online 24 July 2010 Keywords: Fault interactions Taupo rift Paleoearthquakes Kinematic coherence Fault growth Abstract Fault interactions are an essential feature of all fault systems on timescales from individual earthquakes to millions of years. We examine the role of these interactions in the development of an array of normal faults within the active Taupo Rift, New Zealand. Stratigraphic horizons (0-26 ka) exposed in 30 trenches and laterally extensive topographic surfaces (18-340 ka) record displacements during surface-rupturing earthquakes over time intervals up to 100’s of thousands of years. Complementary changes in displacements, displacement rates, and earthquake histories between faults are observed for along-strike displacement profiles and at points on fault traces. Variations of displacement are attributed mainly to fault interactions, and decrease with the aggregation of displacements on progressively more faults and over longer time intervals. Rift-wide displacement rates are near-constant over timescales of <60 kyr and suggest a level of order which is greater than that of individual fault traces. Each fault is nevertheless a vital element of a system that displays a remarkable degree of kinematic coherence, producing and maintaining a hierarchy of fault size throughout the deformation history. As a consequence, on spatial scales greater than an individual fault trace and over temporal scales more than several earthquake cycles, the behavior of individual faults can be relatively predictable. Fault interactions are accompanied by changes in fault system geometries consistent with increases in their maturity arising from strain localization processes, including fault linkage and death. 1. Introduction Fault interactions are an essential feature of fault systems (Walsh and Watterson, 1991). In circumstances where faults intersect (hard-linkage) or form relay ramps (soft-linkage), complementary displacement patterns on the component faults provide unequivocal kinematic evidence for their interaction (e.g., Peacock and Sanderson, 1991; Childs et al., 1995; Cartwright et al., 1995, 1996; Dawers and Anders, 1995; Nicol et al., 1996; Walsh et al., 2003a). As fault growth by increasing fault length or displacement is typically achieved via many increments of slip which accrue during earthquakes (Walsh and Watterson, 1987; Stein et al., 1988; Cowie and Scholz, 1992a; Nicol et al., 2005a), fault interactions require that both finite fault displacement and earthquake slip are complementary between faults (Fig. 1). Whether a fault system is hard-linked or soft-linked, fault interaction reflects the stress concentrations and shadows arising from short-term stress-dependent behavior of earthquakes (e.g., Harris, 1998; Stein, 1999; King and Cocco, 2000; Robinson, 2004; Pondard et al., 2007). These stress studies together with analysis of paleoearthquakes on multiple faults within systems (e.g., Rockwell et al., 2000; Nicol et al., 2006; Dolan et al., 2007) suggest that faults may dynamically and kinematically interact over distances of tens of kilometers. These long-range interactions are, however, sometimes difficult to demonstrate kinematically because they produce displacement variations over the life span of the faults that are more subtle and less obviously complementary than is the case for well-defined arrays of fault segments (compare faults 2 & 3 in Fig. 1). Fault interactions have a profound impact on the dimensions and displacement accumulation of faults in systems over timescales from millions to thousands of years (Fig. 1). Many faults can, however, have near-constant lengths from an early stage in their history, with growth primarily achieved by increases in cumulative displacement rather than changes in fault length (Walsh et al., 2002; Childs et al., 2003; Schlagenhauf et al., 2008). Near-constant fault lengths are attributed to retardation of lateral propagation by interaction between fault tips. In systems where faults reach their final length early, the maximum magnitude of subsequent earthquakes will also be constant. For faults of constant-length temporal variability in slip and recurrence of earthquakes may, in part, reflect migrations in the loci of strain accumulation associated with fault interactions (Wallace, 1987; Nicol et al., 2006). These interactions therefore produce both variations in the earthquake process and, in circumstances where fault system boundary conditions are uniform, a hierarchy in which the longest faults move fastest, but with individual faults accruing displacement at near-constant rates over millions of years (Nicol et al., 1997, 2005b). In this paper we examine the role of fault interactions on fault displacement accumulation over timescales ranging from individual earthquakes to 100’s of thousands of years for a system of normal faults in the Taupo Rift, New Zealand (Fig. 2). This active rift, which is delineated by active fault traces and historical seismicity, generally forms a well-defined graben 15-40 km wide (Fig. 2 cross-section) that started to form at about 1-2 Ma (Wilson et al., 1995) and comprises a dense array of active normal fault traces (Figs. 2, 3a and 4). The large number of active fault traces (N > 300) makes the Taupo Rift an excellent location to examine fault growth and the role of fault interaction. These interactions and associated fault growth are recorded using displacements of 0.2 m-550 m for time intervals of 1-340 kyr from offset topographic surfaces and near-surface stratigraphy in fault trenches (Fig. 3). Displacements provide information on the timing and slip of surface-rupturing earthquakes and on the accrual of displacement over time intervals up to 100’s of thousands years. The high density, short lengths, and segmented nature of active fault traces are all hallmarks of a relatively immature fault system at the ground surface in the Taupo Rift (see Wesnousky, 1988; Cowie et al., 1993; Meyer et al., 2002; Schlagenhauf et al., 2008). This immaturity strongly impacts the fault displacement-length population and appears to decrease with depth (Seebeck, 2008), suggesting that the fault system has evolved through time. Our analysis also demonstrates that the accumulation of displacement of all faults in the system and the history of earthquakes that produce displacement increases are interdependent. Interactions between faults have a profound impact on their growth, a conclusion which we believe has wide application to other fault systems. 2. Fault system geology and data Normal faults of the Taupo Rift accommodate near-orthogonal extension of up to about 15 mm/yr in the central North Island of New Zealand (Villamor and Berryman, 2001; Wallace et al., 2004; Lamarche et al., 2006; Mouslopoulou et al., 2008; Begg and Mouslopoulou, 2010). The rift passes through two rhyolitic volcanic centers outside the study area. However, this volcanism (including dike intrusion) does not appear to induce fault slip outside the volcanic centers where the faults are interpreted to be of tectonic origin (Seebeck and Nicol, 2009). Back-arc extension may have formed in response to clockwise rotation of the eastern North Island (relative to western North Island) associated with rollback of the subducting Pacific Plate, continental collision at the southern end of the Hikurangi margin, and mantle buoyancy force (e.g., Walcott, 1987; Beanland and Haines, 1998; Wallace et al., 2004; Stratford and Stern, 2006; Nicol et al., 2007) (Fig. 2 inset). Normal faulting is thought to have initiated 1-2 Myr ago (Wilson et al., 1995) and produced a total crustal extension of ~20-60_ on the syn-rift to basement boundary (Nicol et al., 2007; see also cross-section on Fig. 2). Ключевые слова: cartwright, fault interaction, increase, range, evolution, north island, single fault, geophysical, watterson walsh, flat, surface, earthquake history, seebeck, walsh watterson, zealand, childs, island, slope, wesnousky, ignimbrite sheet, harris, kinematic coherence, time, array, society, surface-rupturing earthquake, fault journal, profound impact, segmented fault, normal fault, larger fault, tectonics doi, walcott, schultz, earthquake flat, geophysics, journal structural, northeastward shift, horizon, plot, nicol walsh, individual fault, elsevier, fault, earthquake cycle, highest displacement, immature fault, recurrence interval, spatial migration, relay ramp, grey cross, seismological society, extension, dawers, cowie, process, mcclymont, model, map area, earthquake, single-event displacement, change, rift, point sample, villamor, displacement accumulation, dense array, volcanic, interaction, lowe, large number, figs, geological timescales, ground surface, berryman, berryman nicol, gradient, maximum, individual, displacement, relation, displacementetime curve, propagation, essential feature, stein, slip, journal structural geology, journal, location, rate, trace, linkage, temporal, taupo rift, interval, north, segmented nature, abandoned channel, individual earthquake, trench log, fault trace, nicol journal, earthquake activity, wilson, trench, shadow arising, del, data, del relation, channel, smaller fault, berryman berryman, year, deformation history, kinematically coherent, strain accumulation, point, sampling bias, ancient fault, number, time interval, recurrence, lateral propagation, zone, displacement proles, structural geology, accumulation, topographic, strain, active fault, lamarche, map location, displacement rate, preferential linkage, journal geophysical, active, tectonics, del slope, maximum displacement, structural, fault size, robinson, timescales, geological, single, villamor berryman, length, del data, stratigraphic position, watterson, walsh, taupo, segment linkage, fault intersection, pro?les, cumulative displacement, basin, geology, scholz, active rift, history, kyr, event, mouslopoulou, earthquake slip, mans?eld, paleoearthquakes, beavan, wallace, growth, variation, doi, structural evolution, single-event, age, proles, king, prehistoric, consistent, nature, displacement pro?les, ka, near-constant rate, bulletin, nairn, watterson cowie, normal, fault fault, fault length, trench excavated, nicol, zealand journal, map, fault growth, entire rift, fault displacement