Uncertainties in estimating the effect of climate change on 100-year return value for significant wave height

The process of estimating the effect a changing climate on severity future ocean storms is plagued by large uncertainties; for safe design and operation offshore structures, it nevertheless important that best possible estimates effects made given available data. We explore variability in 100-year return value significant wave height (HS), changes over period time, output WAVEWATCH-III models from 7 representative Coupled Model Intercomparison Project (CMIP) Phase 5 General Circulation Models (GCMs), FIO-ESM v2.0 CMIP 6 GCM. Non-stationary extreme analysis peaks-over-threshold block maxima using Bayesian inference provide posterior values as function time; MATLAB software provided. Best overall values, 1979-2100, are calculated averaging individual GCMs. focus attention neighbourhoods locations east Madagascar south Australia where previous study CMIP5-derived reported decrease increase HS respectively, under Representative Concentration Pathway (RCP) scenarios RCP4.5 RCP8.5. There variation between different GCMs, with longitude latitude within each neighbourhood based samples corresponding to ≤ 165 years model output; these sources uncertainty tend be larger than those due typical modelling choices (such choice threshold peaks threshold, or length maxima). However, we also find careful critical Madagascar, because presence mixed population there. Nevertheless, there general evidence supporting trends others, but findings conditional 8 GCMs being evolution. use simple randomisation testing identify “significant” departures steady climate. long 700-year pre-industrial control (piControl) CMIP6 GCM offers an excellent opportunity quantify apparent inherent estimated subsample continuous time interval 20 160 length, no forcing present. approximately ±15% periods around 50 drawn piControl