Serial clustering of extratropical cyclones over Western Europe: dynamics and associated impacts

• Main Author: Priestley, Matthew D. K.
• Format: Thesis
• Language: English
• Published: 2019
• Subjects: North Atlantic
• Online Access: https://centaur.reading.ac.uk/88666/; https://centaur.reading.ac.uk/88666/1/19006939_Priestley_thesis.pdf; https://centaur.reading.ac.uk/88666/2/19006939_Priestley_form.PDF

Extratropical cyclones are a key process in the atmospheric variability of the North Atlantic and western Europe. They are associated with heavy precipitation and extreme winds which can result in considerable socio-economic impacts. Throughout the winter season these intense extratropical cyclones have been shown to occur in groups (this is known as clustering), which is more prevalent over the eastern North Atlantic and western Europe. In this thesis the drivers of cyclone clustering are investigated, particularly focussing on the large-scale dynamical mechanisms and the role secondary cyclones. During clustering events the upper-troposphere over the North Atlantic is characterised by a strong and zonally extended jet which steers large numbers of cyclones toward western Europe. The extended jet is associated with anomalous Rossby wave breaking on one or both flanks, with the balance on each flank being associated with changes in the angle of the jet and the latitude at which the clustering occurs. Secondary cyclones are objectively identified and are shown to contribute approximately 50% to the increase in cyclone numbers during periods of intense clustering. This increase is mainly a result to the large-scale flow steering more secondary cyclones along a similar track toward western Europe with there also being a slight increase in genesis rate near western Europe. The climate model HiGEM is used to examine the impact of cyclone clustering on seasonal wind damage estimates across Europe. HiGEM is able to represent the large-scale dynamics associated with clustering and a wind speed proxy, the storm severity index, is used to estimate the associated losses. It is found that clustering acts to increase the losses that are experienced across Europe by 10-20% for high return period seasonal losses compared to a random series of cyclones in a season. These results demonstrate the importance of correctly representing clustering in modelling studies and loss estimations.