Groundwater resources critical to North Canterbury's agriculture, and industrial sustainability and development are predominately located within the Late Quaternary age (0- 0.78 myr) sediments that underlie, and form the North Canterbury Plains. The aim of this research is to determine if shallow seismic reflection P-wave surveying is capable of delineating sedimentary architecture related to groundwater resources in the glaciofluvial and fluvial sediments of the North Canterbury Plains and in particular, if (high porosity/permeability pathways) and aquitard (low permeability) units can be differentiated and so be applied to understanding groundwater resource in the North Canterbury region. The research was undertaken in three geographic areas in North Canterbury, which represent the main environments for Late Quaternary sedimentary deposition. i) The Omihi Valley a thrust controlled foothill valley setting. ii) The Racecourse Hill-Burnt Hill area which represents a basin-marginal setting close to the emergence of the glacial outwash fans that constitutes the Northwest Canterbury composite aggradation surfaces, and iii) Pines Beach, a more extensively reworked prograding postglacial braidplain with repeated episodes of marine incursion. Three methodologies have been developed to identifY aquifer units in the Late Quaternary sediments of North Canterbury using shallow seismic reflection surveying.

1. Combined structural delineation seismic surveys and borehole logging to define medium scale (lOrn- 2 km) structures and lithology which affect aquifer and aquitard location and extents.
2. Tailored two and three dimensional shallow seismic reflection surveys to define sedimentary architecture within the larger scale sediment packages delineating high porosity/permeability pathways e.g. paleo-channels in the Omihi Valley and paleo-cut and fill valleys on the Northwestern Canterbury Plains margin.
3. Changes in seismic reflection attributes which directly reflect sediment changes in permeability and porosity such as the decrease in interval velocity with reduced matrix clay/silt content in the Omihi Valley Late Quaternary sediments. The main research outcome achieved from the 33 km2 Omihi Valley was the integration of geological/geophysical mapping with lithological borehole logging to develop a groundwater resource model allowing a predictive approach to its exploration and utilization. Structural geometry and styles of deformation of the Valley have also been characterised . The seismic reflection survey results for Burnt Hill and Racecourse Hill, both basin-margin areas, indicate that seismic surveying can be successfully used to image alluvial and fluvial architecture at scales from 10m- 1000's m, but with lower lateral ( ~28m@ 100m depth) and vertical resolution ( 4 m) than that of the Omihi Valley surveys. The interpretation of the intra-gravel seismic reflections is complex, and likely only to be possible with three dimensional seismic reflection surveying. In the third site, only a limited survey of one line was undertaken at Pines Beach (Canterbury Plains/Pegasus Bay junction). This survey indicates that interfingering of reworked finergrained fluvial sediments and coastal marine sediments can be successfully characterized with a lateral resolution of< 11 m and vertical resolution of 1.5 m for sediments within the top 50 m. The research generally demonstrates that shallow seismic reflection P-wave surveying can be extended to the more laterally extensive North Canterbury Plains as a whole, where it is capable of delineating subsurface sedimentary facies, including hydrologically important architecture, down to sub-two metre vertical resolution, in glacio-fluvial, fluvial and shallow marine derived sediments in the 20 - 500 m depth range. The paleo-sedimentary structures successfully delineated in the three field areas include paleo-channels, channel fill, large scale erosional cut-and-fill valleys, floodplain surfaces, and alluvial fans. It is shown that the aquifer and aquitard contrast in North Canterbury is not characterised by porosity, but by permeability differences (which are also affected by porosity). Aquifers are saturated high porosity/permeable sediments with low silt/clay content while aquitards are saturated units with low permeability which appear (in the Omihi Valley) to be controlled by silt/clay content, but not necessary low porosity. P-wave seismic reflection surveying is shown to be insensitive to sediment permeability variations, but limited data from the Omihi Valley indicates that seismic P-wave velocities may be sensitive to matrix clay/silt content. If proven elsewhere in the Canterbury Plains region, this may lead to a method of defining aquifer/aquitard geometry directly, without the need to use indirect methods such as paleo-fluvial facies architecture delineation. This thesis concludes that shallow seismic reflection surveying is capable of characterizing the North Canterbury Plains and foothills valleys sedimentary lithofacies architecture, and also shallow, tectonically-driven structural deformation, when used as part of a multi-faceted programme of investigation. These data can be used to delineate the main groundwater resources. Further research is required to determine if direct aquifer geometry identification is possible by seismic P-wave interval velocity inversions and to increase seismic reflection acquisition speed to allow efficient coverage of large areas.