Abstract
General Extreme Value (GEV) distribution models describe extreme climate events and are often used to generate quantitative metrics of the risk of these events. Probabilistic risk factors, generated using single covariate non-stationary GEVs, have effectively shown that the influence of El Niño Southern Oscillation (ENSO), North Atlantic Oscillation (NAO) and the Pacific Decadal Oscillation (PDO) on extreme winter across North America is significant. In this study, we fit a series of single and multiple covariate non-stationary GEVs to monthly Block Maxima (BM) series of daily accumulated precipitation (ACPCP) data from the North American Regional Reanalysis (NARR) model and the Community Climate System Model 4 (CCSM4), an ensemble member of the North American Multi-Mean Ensemble (NMME). Over the southwest and west coast regions of the US and Mexico, there was in increase in the risk of extreme precipitation event during La Niña events. During El Niño events, there was an increase in risk across south and southeast US as well as southern Canada. During the positive NAO phase, increased risk was found over western and central US. Additionally, we found that NAO exhibited greater influence than ENSO across the region. Risk factor analyses from the multiple covariate non-stationary GEVs suggested that the influence of NAO is more dominant than ENSO on the probabilistic risk of extreme winter precipitation events.