Quantifying the rarity of extreme multi-decadal trends: how unusual was the late twentieth century trend in the North Atlantic Oscillation?

Abstract Climate trends over multiple decades are important drivers of regional climate change that need to be considered for resilience. Of particular importance extreme society may not expecting and is well adapted to. This study investigates approaches assess the likelihood maximum moving window in historical records indices by making use simulations from models stochastic time series with short- long-range dependence. These applied unusualness large positive trend occurred North Atlantic Oscillation (NAO) index between 1960s 1990s. By considering models, we show chance determined variance process, which generally increases when there more serial correlation series. We find Coupled Model Intercomparison Project (CMIP5 + 6) have very rarely (around 1 200 chance) simulated greater than observed maximum. Consistent this, NAO CMIP were found resemble white noise, almost no correlation, contrast exhibits year-to-year correlation. Stochastic model best fits suggest an unlikely 20) 31-year as since 1860. suggests current do fully represent aspects mechanism low frequency variability NAO.