Detection and quantification of fraction of attributable risk of extremes at regional to global scale

Observed trends in the intensity of temperature extremes and heavy precipitation events are often not significant at local scales. However, using a spatially aggregated perspective, we demonstrate that the probability distribution of observed local trends across the globe for the period 1960–2010 is clearly different to what would be expected from internal variability. Using a simple and intuitive approach, we thereby detect a distinct intensification of heavy precipitation events and hot extremes, which is consistent with early studies using optimal fingerprinting methods.

We further demonstrate that spatial aggregation also allows for a robust quantification of changes in the Fraction of Attributable Risk (FAR). While FAR is primarily applied to individual observed extremes, we argue that the concept is equally meaningful for the probability of extremes at continental to global scale. We demonstrate that with a warming of 2°C above pre-industrial conditions, the probability of exceeding the 99^th percentile of daily precipitation increases by 25-75% globally. Thereby, in a 2°C warmer world roughly a third of the heavy precipitation events worldwide and more than 90% of the hot extremes are attributable to the temperature increase. We show that the changes in FAR per degree warming are highly sensitive to the thresholds used and the temporal scale of the events considered. However, at regional to global scale changes in FAR are robust across different GCMs.