Beaches around the world continuously adjust to daily and seasonal changes in wave and tide conditions, which are themselves changing over longer time-scales. Different approaches to predict multi-year shoreline evolution have been implemented; however, robust and reliable predictions of shoreline evolution are still problematic even in short-term scenarios (shorter than decadal). Here we show results of a modelling competition, where 19 numerical models (a mix of established shoreline models and machine learning techniques) were tested using data collected for Tairua beach, New Zealand with 18 years of daily averaged alongshore shoreline position and beach rotation (orientation) data obtained from a camera system. In general, traditional shoreline models and machine learning techniques were able to reproduce shoreline changes during the calibration period (1999–2014) for normal conditions but some of the model struggled to predict extreme and fast oscillations. During the forecast period (unseen data, 2014–2017), both approaches showed a decrease in models’ capability to predict the shoreline position. This was more evident for some of the machine learning algorithms. A model ensemble performed better than individual models and enables assessment of uncertainties in model architecture. Research-coordinated approaches (e.g., modelling competitions) can fuel advances in predictive capabilities and provide a forum for the discussion about the advantages/disadvantages of available models.

@en

Understanding marine climate variability is important for coastal planning and marine operations. It is also particularly challenging for complicated settings (e.g., islands) and data-poor regions. The aim of this work is to establish a relationship between daily synoptic atmospheric patterns, and wave and storm surge conditions around New Zealand waters, based on instrumental and reanalysis data. The daily predictor we developed is able to represent sea and swell wave conditions as well as storm surge variability over different temporal scales. However, when climate variability is analysed on a longer temporal period, based on the 20th century reanalysis, large inhomogeneities are found. This highlights the dangers related to assessing climate variability, especially in data-poor regions (such as New Zealand), where inhomogeneities could be interpreted as actual changes.

@en

Rising seas coupled with ever increasing coastal populations present the potential for significant social and economic loss in the 21st century. Relatively short records of the full multidimensional space contributing to total water level coastal flooding events (astronomic tides, sea level anomalies, storm surges, wave run-up, etc.) result in historical observations of only a small fraction of the possible range of conditions that could produce severe flooding. The Time-varying Emulator for Short- and Long-Term analysis of coastal flood hazard potential is presented here as a methodology capable of producing new iterations of the sea-state parameters associated with the present-day Pacific Ocean climate to simulate many synthetic extreme compound events. The emulator utilizes weather typing of fundamental climate drivers (sea surface temperatures, sea level pressures, etc.) to reduce complexity and produces new daily synoptic weather chronologies with an auto-regressive logistic model accounting for conditional dependencies on the El Niño Southern Oscillation, the Madden-Julian Oscillation, seasonality, and the prior two days of weather progression. Joint probabilities of sea-state parameters unique to simulated weather patterns are used to create new time series of the hypothetical components contributing to synthetic total water levels (swells from multiple directions coupled with water levels due to wind setup, temperature anomalies, and tides). The Time-varying Emulator for Short- and Long-Term analysis of coastal flood hazard potential reveals the importance of considering the multivariate nature of extreme coastal flooding, while progressing the ability to incorporate large-scale climate variability into site specific studies assessing hazards within the context of predicted climate change in the 21st century.@en

Wave-induced flooding is a major coastal hazard on tropical islands fronted by coral reefs. The variability of shape, size, and physical characteristics of the reefs across the globe make it difficult to obtain a parameterization of wave run-up, which is needed for risk assessments. Therefore, we developed the HyCReWW metamodel to predict wave run-up under a wide range of reef morphometric and offshore forcing characteristics. Due to the complexity and high dimensionality of the problem, we assumed an idealized one-dimensional reef profile, characterized by seven primary parameters. XBeach Non-Hydrostatic was chosen to create the synthetic dataset, and Radial Basis Functions implemented in MATLAB® were chosen for interpolation. Results demonstrate the applicability of the metamodel to obtain fast and accurate results of wave run-up for a large range of intrinsic reef morphologic and extrinsic hydrodynamic forcing parameters, offering a useful tool for risk management and early warning systems. @en

The assessment of coastal flood risk on a reef-lined coastline presents several challenges. From the probabilistic side, we need to consider all possible events that could occur in the system, taking into account the different contributions of waves, storm surges, and tides that contribute to the total water level. To estimate reliable flood extents, we need to accurately model the complex wave processes that occur across the reef. To explore the multivariate nature of coastal flooding, we rely on a climate emulator that accounts for climate variability and simulates time series of all the variables involved. Due to the computational constraints to numerically simulate thousands of events, we explore the feasibility of using a recently developed tool, the HyCreWW (Hybrid Coral Reef Wave and Water level) meta-model to estimate wave run-up and flooding extents. Limitation of a 1D assessment are analyzed and results compared with 2D modeling.@en