Atmospheric Rivers and Cyclone Clustering from Reanalyses and High-Resolution Model Simulations

Serial clustering of extratropical cyclones is a phenomenon which has strong implications for the occurrence of extreme hydroclimate events, atmospheric moisture transport and land precipitation. We propose here a new approach for cyclone series identification. This approach is based upon Lagrangian concept, when the analysis is applied to all cyclones propagating over the domain, cyclone propagation velocities and cyclone-to-cyclone distances. This is different from the Eulearian approach based on application of Poisson distribution to the pressure minima waiting times. We apply new approach to several modern reanalyses (ERA-Interim, ERA5) starting from 1979 onward and to the high resolution North Atlantic atmospheric hindcast performed with WRF model for the same period. Applying the new Lagrangian approach we quantified cyclone clustering in different reanalyses and regional modeling products and established new metrics of clustering including frequency of serial clustering events, life time of cyclone clusters, as well as cumulative characteristics of individual cyclone clusters. For this purpose we extended cluster identification to the tracking of clusters, thus quantifying clusters’ space-time evolution. Also new algorithm was validated and compared to the Eulerian algorithm in order to quantify uncertainties of Poisson distribution – based approach. Further we associated cyclone serial clustering with atmospheric rivers and the intensity of moisture transports for the North Atlantic. For this purpose we performed the detection and tracking of atmospheric rivers in the same products using a hybrid approach based on the analysis of atmospheric vapor transport and involving both dynamical analysis and machine learning techniques. Specific consideration was given to the co-variability in cyclone clustering and atmospheric rivers over the last several decades and to climate variability in the occurrence of cyclone clusters and atmospheric rivers.