Analysis of non-stationary climate processes and extremes across scales

Numerous studies show that climatic extremes have increased substantially in the second half of the 20th century. For this 
reason, analysis of extremes under a non-stationary assumption has received a great deal of attention. Extreme value 
analysis (EVA) is the branch of statistics concerned with making inferences on events with very low probabilities of 
occurrence. A review of the univariate setting using Bayesian inference under non-stationary assumption will be given, 
where tools are readily available. Beyond the univariate setting, when estimating the probability of extreme meteorological 
events, it is often of interest to characterize spatial dependencies. It offers potential predictability, both on short and 
long time scales and may also be useful to reduce the considerable level of uncertainty in modeling of extremes with local 
models; interpolate values to unobserved locations, or to utilize information from similar locations in order to obtain 
more powerful results. The study of extremes in space is an active area of research. This talk will give a brief overview 
of some proposed methods, including a new conditional EVA approach.