<p>The coastal geomorphology of the Raukumara Peninsula, North Island, New Zealand, is investigated with the aim of understanding the geodynamics of this segment of the Hikurangi subduction zone. There are few active faults on the Raukumara Peninsula and there have been no historical subduction earthquakes on this part of the margin. Geodetic and geologic evidence suggests the onshore forearc is extending and sliding trenchward. However, Holocene and Pleistocene terraces that have some of the highest uplift rates and elevation in New Zealand occur in this region. Sediment underplating has previously been suggested as a cause of coastal forearc uplift, yet the inferred gradualness of this process does not reconcile with the inferred episodic nature of uplift suggested by stepped Holocene marine terraces along the Raukumara Peninsula. This study focuses on resolving these apparent inconsistencies in the current knowledge of the Raukumara sector of the Hikurangi margin by using the mechanisms of coastal uplift to improve our general understanding of forearc deformation processes. The mid to late Holocene marine terrace chronology and distribution at the Pakarae River mouth locality on the central Raukumara Peninsula eastern coastline is revised. The coverbed sediments and morphology of the terrace surfaces and straths there are consistent with abandonment of each terrace during a coseismic coastal uplift event. The terraces are offset across the Pakarae Fault, a normal fault, but tilt vectors of the terraces suggest the dominant structure driving uplift is instead an offshore reverse fault in the upper plate of the subduction zone. The highest Holocene marine terrace at the Pakarae River mouth is underlain by a transgressive fluvio-estuarine sequence that documents infilling of the Pakarae River paleo-valley from ~ 10,000 - 7,000 cal. yrs B.P. prior to eustatic sea level stabilisation. This sequence is an example of a valley infill sequence modified by tectonic uplift. By comparing it with facies models developed from stable coastlines a new facies architecture model is developed for incised valley infilling during conditions of synchronous eustatic sea level rise and tectonic uplift. The sedimentology and biostratigraphy of the Pakarae River incised valley infill sequence is also used to reconstruct the paleoenvironmental evolution of the paleovalley during the early Holocene. Two estuarine units display sudden vertical transitions to floodplain sediments implying significant marine regressions. During the depositional period, however, eustatic sea level was rapidly rising. The marine regressions, and associated estuary abandonment, are therefore attributed to coseismic coastal uplift events, occurring at ~9,000 and ~8,500 cal. yrs B.P. A third uplift between 8,500 and ~7,350 cal. yrs B.P. is inferred from a significant difference between the amount of sediment preserved and the predicted sediment thickness according to the eustatic SL curve. This study demonstrates the utility of the stratigraphic analysis of incised valley infill sequences for neotectonic investigations on active coasts. In practice the application of incised valley infill analysis allows extension of coastal paleoseismic histories back prior to the time of eustatic sea level rise culmination. The combination of the Pakarae River mouth marine terrace and sedimentary infill sequence data yields the longest record of coastal paleo-earthquakes yet constructed globally. Along the northeastern coastline of the Raukumara Peninsula a variety of Holocene coastal features have been studied to determine the uplift mechanism of this region. Topographic and coverbed stratigraphic data from previously interpreted coseismic marine terraces at the Horoera and Waipapa localities, east of Te Araroa, indicate that, despite their stepped surface morphology, they are not of marine origin and thus cannot be attributed to coseismic coastal uplift events. Paleoenvironmental analysis of drill cores of early Holocene sediments underlying the Hicks Bay flats shows a typical incised valley infill sequence with gradual transitions from fluvial to estuarine and back to fluvial. However, there are significant differences between the thickness of preserved intertidal infill sediments and the amount of space created by eustatic sea level rise. Therefore, uplift did occur during early Holocene evolution of the Hicks Bay paleo-estuary, but at this location there is no evidence of any sudden or coseismic land elevation changes. The beach ridge sequence of Te Araroa slopes gradually toward the present day coast with no evidence of coseismic steps. It is inferred the evolution of the beach ridges was controlled by a variable sediment supply rate in the context of a background tectonic uplift rate that was geologically continuous rather than punctuated. Although no individual dataset from the Holocene coastal features of the northeastern Raukumara Peninsula can uniquely resolve the mechanism of uplift, a careful evaluation and integration of all available evidence indicates uplift of this region has been driven by a gradual and aseismic mechanism. Pleistocene marine terraces fringing the northern coastline of the Raukumara Peninsula, and overlapping with the northeastern Raukumara Peninsula study area, display a northwest tilt. The tilt vector steepens and rotates gradually northward on the eastern side of the Peninsula. As there are no upper plate active faults that can account for this geometry, it is concluded that the buoyancy of underplated sediment beneath the forearc drives aseismic uplift of this region. The Holocene coastal uplift mechanism data from the Pakarae River mouth and the northeastern Raukumara Peninsula are integrated with other coastal geomorphic data from the Raukumara Peninsula in this thesis. Distinct spatial variations in the coastal uplift rates and mechanisms are seen. The main parameter controlling the distribution of aseismic and coseismic deformation processes is inferred to be the distance of the forearc from the Hikurangi Trough, and depth to the underlying plate interface. Three margin-parallel zones of forearc uplift mechanisms are identified: (1) a zone of coseismic uplift on upper plate contractional faults located within 20 - 80 km of the trench; (2) a passive inner forearc zone at a margin-normal distance of ~80-120 km from the trench, in which vertical tectonic movement is controlled by distal upper plate structures and or slip on the plate interface; and (3) a zone of aseismic uplift driven by sediment underplating located at a margin-normal distance of ~120 - 180 km of the trench, also encompassing the Raukumara Ranges. Each of the zones has specific seismic hazard implications. Reverse faults within the coseismic uplift zone pose a significant seismic and tsunami hazard to Raukumara Peninsula coastal communities, while seismic hazard in the zone of aseismic uplift is probably less than previously thought. The passive forearc zone probably experiences uplift or subsidence during plate interface rupturing earthquake and therefore it may be the only one of the three zones that has the potential to record past subduction earthquakes. This study demonstrates the value of coastal geomorphology and stratigraphic studies in subduction zone neotectonic and paleoseismology studies. Variances in coastal uplift mechanisms across the Raukumara Peninsula are related to spatial position within the Hikurangi subduction margin and this has implications for our understanding of forearc deformation processes along the length of the margin.</p>
