Mechanisms of Cliff Retreat on a Developing Rocky Coast

Abstract

Morphological models of long term coastal cliff retreat and shore platform development remain poorly constrained owing to poor understanding of process-form relationships in particular, the nature of cliff retreat mechanisms and the hydrodynamics of shore platforms. Discontinuities clearly provide the failure surfaces for mass movement in harder lithologies, but the spatial and temporal controls on failure frequency and magnitude have not been fully investigated. This thesis applies geotechnical slope stability modelling techniques to understand the cycle of cliff toe erosion and mass movement and how this cycle can be interrupted by external and internal changes to slope stability. Prior to this thesis geotechnical models have not been used in geomorphological studies of cliff retreat in harder rocks. Case studies are taken from two sites near Gisborne, New Zealand from which previous measurements of wave transformation are available. Differences in structural geology help to explain the frequency and magnitude of cliff retreat at each site. Conceptual modelling of Holocene cliff retreat into a hillslope suggests that in early stages of cliff retreat, rapid cliff toe erosion in the high wave energy environment may pre-condition slope geometry to be highly susceptible to external/internal changes. As the cliff retreats into a hillslope, slope stability decreases and shear strain localise in the lower slope causing a ‘detachment’ of the upper slope and more complex forms of retreat. At present, cliff toe erosion rate averages between 12-16 cm/yr nearer the headland tips. Talus deposits are currently eroded at up to 162 cm/yr (based on site survey) but can be up to 285 cm/yr based on assessment of aerial photography. As a result of new insights obtained, the thesis challenges the prevailing static equilibrium theory of shore platform development. A new non-equilibrium recession model is proposed for micro-tidal environments comprising tertiary sediments, in which retreat of the landward cliff proceeds much further than previously imagined. Three lines of evidence support the model: 1) the cliffs at both study sites are still eroding despite being fronted by unusually wide platforms of over 400 m width; 2) wave energy does not decay exponentially across a shore platform as previously thought, and infragravity waves increase in height across the platform; 3) tidal cycles and infragravity waves provide a mechanism for weathering of flysch material, involving swelling of clay minerals and removal of the fine grained products of mass movement from the cliff toe.