High-dimensional statistical modelling of changes in wave climate and implications for maritime infrastructure

Future changes in ocean wind wave climate have broad implications for the physical infrastructure and environment in coastal, near- and offshore regions. A comprehensive assessment of the impact of climate change and the associated risks in these regions thus requires the consideration of changes in wave climate. However, the understanding of projected changes in wave climate is limited relative to many other climate variables such as temperature and precipitation. This results from the fact that commonly used climate models typically do not provide projections for future wave climate as a standard model output. To decrease this knowledge gap, the HDwave project has developed new statistical models for wave climate as well as tools for the analysis of wave climate projections. We have demonstrated that non-stationary, fuzzy time series approaches can be applied to generate projections of future wind and waves. Such models are able to make use of a variety of data sources such as hindcasts, measurements and satellite data. We have further developed a statistical prediction model for significant wave height that provides probabilistic predictions that are physically consistent in space and time. An open-source software for this model is available as an R-package. By using regional extreme value analysis, we have shown that regional estimates of extreme wave height quantiles are more robust and associated with a smaller uncertainty than those commonly obtained from local analyses. Generally, future wave climate projections are associated with a large uncertainty and an overall decreasing trend in the Northern Atlantic is observed. An effect study on ship responses indicates that both ship motions and bending moments may be lower in the future. However, these effects are associated with large uncertainties.

The project outcomes fall in two groups, statistical methodology and software, and assessment results. New statistical methodology for wave climate prediction with associated open-source software provides a computationally efficient alternative to obtain future wave climate information. Our assessment of future wave climate in the Northern Atlantic indicates a large degree of uncertainty regarding future changes, with some indications of an overall decreasing trend. As a consequence, ocean wave impact on ships and offshore structures may decrease. However, such results are highly dependent on geographical location and modelling choices in the generation and analysis of the wave data.

Future changes in ocean wind wave climate have broad implications for the physical infrastructure and environment in coastal, near- and offshore regions. A comprehensive assessment of the impact of climate change and the associated risks in these regions thus requires the consideration of changes in wave climate. However, the understanding of projected changes in wave climate is limited relative to many other climate variables such as temperature and precipitation. This results from the fact that coupled atmosphere-ocean general circulation models (GCMs) typically lack wind wave parameterizations so that wave parameters are not available as a standard model output. The HDwave project aims to decrease this knowledge gap by developing advanced statistical tools for the analysis of newly developed data sets of wave climate projections. The new framework includes methodology to handle multi-model ensembles, provides simultaneous confidence regions in space and time for events of interest and assesses the impact of future changes on maritime infrastructure.