Projections of Global Multivariate Wave Climate for the End of The 21st Century: Robustness and Uncertainties

• Main Author: Morim Nascimento, Joao M
• Format: Text
• Language: English
• Published: Griffith University 2020
• Subjects: Ocean wind-waves; coastal sea-level; shoreline dynamics; coastal populations; marine ecosystems; New Zealand; Pacific; Southern Ocean; North Atlantic
• Online Access: https://dx.doi.org/10.25904/1912/4002; https://research-repository.griffith.edu.au/handle/10072/398879

Ocean wind-waves are dominant contributors to coastal sea-level and shoreline dynamics and can be primary disruptors of coastal populations, marine ecosystems and coastal and offshore infrastructures. Hence, understanding climate-driven changes in the multivariate global ocean wind-wave climate and its extremes is paramount to successful offshore and coastal climate adaptation planning. When projecting future ocean wave conditions under climate change, uncertainty is introduced from a wide range of different sources including future emission scenario, climate model-forcing, and wind-wave modelling methodology. However, the use of single-method wave downscaling ensembles and inherent variability arising from different methodologies has led to unquantified uncertainty amongst existing global wave climate projections. This thesis addresses this significant knowledge gap by investigating the current state of knowledge in projected ocean wind-wave climate. It then, for the first time, quantifies the relative significance of these major sources of uncertainty and establishes the robustness among existing global wave climate projections. This thesis addresses former limitations associated with independent studies by performing a unique analysis of a coordinated, multi-method ensemble of future global wave scenarios assembled from ten individual state-of-the-art studies to yield a 155-member ensemble. This thesis analyses significant wave height, mean wave period and mean wave direction and also explores the underlying large-scale drivers of projected future changes in storm ocean wave events which remain largely unknown. The results of a systematic literature review demonstrate qualitative consensus between existing global and regional-scale wave climate projection studies in terms of a projected increase in mean significant wave height within the Southern Ocean area, Baltic Sea, and tropical eastern Pacific, and a projected decrease in the North Atlantic and Mediterranean Sea. In contrast, current projections of mean ocean significant wave height lack scientific consensus across the remaining ocean regions, with existing projections of extreme wave climate lacking consensus almost everywhere. In addition, results demonstrate a notable lack of knowledge in terms of projected changes in wave period and/or wave direction which are of critical importance, particularly for the mitigation of coastal hazards/risks. Furthermore, the projection uncertainty surrounding wind-wave climate projections has been poorly sampled to date, therefore implying a need for a shift towards a systematic, community-based framework to foster concerted knowledge. Quantitative analysis of surface wind fields from CMIP5 models used to force dynamical and statistical wave models demonstrate that climate models are typically well capable of reliably reproducing large-scale spatial patterns of historical surface winds, albeit there is considerable uncertainty between models with strong latitudinal dependence. Inter-model uncertainty is ~2-4 times larger than uncertainty stemming from GCM internal variability. In addition, results show that CMIP5 climate models exhibit limitations in their ability to accurately represent inter-annual/seasonal variability particularly within tropical cyclone-affected areas. In further analysis, results that surface wind field bias from climate models are largely intrinsic to their atmospheric core components. Inconsistency between surface wind fields from atmospheric-only and coupled models are largely driven by sea surface temperature errors. To address the existing limitations surrounding future wave climate, a large 155-member ensemble has been collated from ten contributing organisations as part of the COWCLIP (herein Coordinated Ocean Wave Climate Project) community. The ensemble comprises general and extreme statistics of significant wave height, mean wave period, and mean wave direction computed over historical (1979-2004) and future (2081-2100) time slices. This ensemble comprises data originated from different wave downscaling approaches, multiple climate-model forcing and future emissions scenarios. Results of wave ensemble model skill against satellite-database measurements of ocean significant wave height and multivariate wave reanalysis for the present-day time-slice demonstrate that models are able to successfully represent annual and seasonal wave variable statistics. Quantitative results derived from this new coordinated multi-method ensemble show, that under a high-emissions scenario, widespread ocean areas exhibit robust changes in annual mean significant wave height and mean wave period of 5-15% and pattern shifts in mean wave direction of 5-15°. Results show that approximately 50% of the world’s coastline is at risk from wave climate change, with ~40% revealing projected robust changes across at least two wave climate variables. Furthermore, results demonstrate that uncertainty in current projections is dominated by climate model-driven uncertainty and also that single-method downscaling studies are unable to resolve up to ~50% of the total uncertainty. In further analysis, hemispheric-scale changes in extreme ocean wave events are analysed using a set of proposed indices that describe the frequency, magnitude and/or persistency of such events. The results demonstrate changes in high-frequency (sub-annual) extreme wave events of the order of 50 to 100% for global warming exceeding 2° C; thus, nearly doubling the projected future changes expected for global warming stabilizing below 2° C (when globally integrated). These changes exhibit strong inter-hemispheric asymmetry, with an increase over the tropics and high latitudes of the Southern Hemisphere, and a widespread decrease across most of the Northern Hemisphere. Lastly, statistical comparison between projected future changes and historical large-scale patterns of change demonstrate that projected changes in extreme wave events across the tropics, high latitudes of the Southern Hemisphere region and across most of the Northern Hemisphere are consistent with their historical response to the positive and negative phase modes of the Southern Annular Mode (SAM) and El Nino-Southern-Oscillation (ENSO), respectively. These results highlight that many nations with low-adaptive capacity located within the Southern Hemisphere region (e.g. West Africa, Australia, New Zealand) will likely face increasing exposure to much more frequent extreme wave events in the future. The results presented in this thesis, and the dataset developed, provide a new perspective on the knowledge on global wave climate projections, providing key insights and support for broad-scale coastal risk and vulnerability assessments and climate adaptation analysis around the world. Whilst the results presented in this thesis have far-reaching implications from many different perspectives, they only address meteorologically-driven changes in ocean wave characteristics. Concentrated community effort is urgently needed to quantify morphologically-driven wave climate change as a contributor to global coastal water level changes, as we look towards improved coastal vulnerability assessments from the climate community.