Multivariate real-time assessment of droughts via copula-based multi-site Hazard Trajectories and Fans

- An original multivariate multi-site analysis of drought dynamics is presented.
- The most recent multivariate copula-based procedures are adopted.
- The multivariate tool of Dynamic Return Period is used to assess the drought state.
- Multivariate previsional tools named “Hazard Trajectories & Fans” are introduced.
- Valuable indications for a multi-site real-time assessment of droughts are provided.

SummaryDroughts, like floods, represent the most dangerous, and costly, water cycle expressions, with huge impacts on society and built environment. Droughts are events occurring over a certain region, lasting several weeks or months, and involving multiple variables: thus, a multivariate, multi-site, approach is most appropriate for their statistical characterization. In this methodological work, hydrological droughts are considered, and a multivariate approach is proposed, by regarding as relevant variables the duration and the average intensity. A multivariate, multi-site, frequency analysis is presented, based on the Theory of Copulas and the joint Survival Kendall's Return Periods, by investigating the historical drought episodes occurred at five main river sections of the Po river (Northern Italy), the most important Italian basin. The tool of Dynamic Return Period is used, and the new concepts of Hazard Trajectories and Fans are introduced, in order to provide useful indications for a valuable multi-site real-time assessment of droughts.