Shoreline change on Ngarunui Beach, Raglan: From measurement to prediction

Abstract

With improvements in video shoreline measurement, and the introduction of Structure-from- Motion (SfM) surveying, the spatiotemporal resolution of beach monitoring data has increased dramatically over the last 30 years. Beach equilibrium modelling has also brought prediction of weekly to multiyear shoreline change seemingly within reach. However, although the accuracy of both video monitoring and equilibrium models has been demonstrated on intermediate beaches, neither method has been thoroughly assessed at dissipative sites. Similarly, despite rapid uptake of SfM, its reliability for surveying on uniform sand surfaces (e.g. fine-grained dissipative beaches) has not been investigated. In response, this thesis evaluates the efficacy of video monitoring, SfM and equilibrium modelling by applying the methods on a highly dissipative beach. Ngarunui Beach, Raglan, in Aotearoa/New Zealand was selected because of its low gradient, fine black sand, and underutilised archive of Cam-Era video imagery. A new shoreline detection method was applied to video imagery to produce the first multiyear shoreline record for the site. Findings revealed long-term erosion and substantial interannual variability over a six-year period, but various sources of error, compounded by the low gradient, ultimately prevented the measurement of change for timescales shorter than monthly. Repeat biweekly SfM surveys, however, successfully captured centimetrescale volumetric losses, supplementing the record of larger-scale change provided by video. Previous studies have raised concerns with applying SfM on smooth sand beaches. For the first time, results from this research revealed spurious surface model roughness in intertidal areas, indicating SfM is only reliable on the upper beach. Lastly, the efficacy of the ShoreFor equilibrium model was assessed. When fully calibrated, ShoreFor reproduced inter-annual variability and trends, revealing equilibrium behaviour at the southernmost of three locations. For parts of the beach that are proximal to a large ebb-tidal delta, ShoreFor failed to replicate shoreline variability. Observations of migrating sand waves adjacent to the delta imply that variability in sediment flux to the beach may be important. Analyses of shoreline response to wave forcing showed no relationships, due to observational error and, critically, because variability in sediment supply is not considered by simple forcing-response or equilibrium models. Overall, the findings suggest that when applied individually, these methods lack reliability at sites like Ngarunui. The thesis proposes a new multi-method framework for complex dissipative beaches in which SfM is combined with video and satellite imagery to monitor change at scales ranging from centimetres to kilometres, and days to decades.