Global analysis of cyclone-induced compound precipitation and wind extreme events

• Main Authors: Martina Messmer, Ian Simmonds
• Format: Article
• Language: English
• Published: Elsevier 2021-06-01
• Series: Weather and Climate Extremes
• Subjects: Compound extreme events; Extreme precipitation; Extreme wind; ERA5; Cyclone tracking and detection algorithms; Low-pressure systems
• Online Access: http://www.sciencedirect.com/science/article/pii/S2212094721000220

Compound extremes, which involve extremes in two or more meteorological parameters, have received little attention until now. We present a new global climatology of compound events that involve extreme precipitation and extreme wind and that are triggered by low-pressure systems. The analysis employs the 3-hourly ERA5 reanalysis (1979–2018) and two independent cyclone detection and tracking algorithms, both of which provide a quasi-Lagrangian perspective. In the analysis, firstly, spatially large and coherent individual extremes in the accumulated precipitation and maximum 10-m wind gust fields are identified. Then cyclone tracks that have both an extreme precipitation and extreme wind event are considered as compound extremes, and these account for 2–3% of the total number of cyclones. The main areas where compound extremes occur depend on the season, but can be summarised as the regions over North America, Japan, the Mediterranean, Australia and regions with high tropical cyclone occurrence. In most of the compound events, either both extremes occur at the same time during the cyclone's lifespan or the precipitation extreme sets in earlier than the wind extreme. There is little difference between the geographic distributions of compound and single precipitation extremes, and a similar comment applies to the wind extremes. However, it is striking that the lifetime of cyclones involved in single precipitation extremes are 48% or 60% and single wind extremes are 64% or 82% of the duration of cyclones associated with compound extremes, depending on the tracking algorithm. The duration of single precipitation extremes is depending on the algorithm 48% or 53% as long as compound extremes, while this number is further reduced to 48% or 51% in wind extremes. This shows that the single precipitation extremes are proportional to the shorter cyclone lifetime, while the duration of wind extremes are disproportionally reduced with respect to the lifetime in single extremes, indicating that especially wind extremes in compound events can profit from the simultaneous presence of precipitation extremes.