Aarhus Universitets segl

Analysis of tornado reports through replicated spatio-temporal point patterns

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review


  • Jonatan A. González, Department of Mathematics, Universitat Jaume I, Castellón, Spanien
  • Ute Hahn
  • Jorge Mateu, Department of Mathematics, Universitat Jaume I, Castellón, Danmark
Understanding the spatio-temporal distribution of tornado events is increasingly imperative, not only because of the natural phenomenon itself and its tremendous complexity but also because we can potentially reduce the risks that they entail. In particular, the U.S. regions are particularly susceptible to the occurrence of tornadoes and they are the focus and motivation of our statistical analysis. Tornado reports can be treated as spatio-temporal point
patterns, and we develop some methods for the analysis of replicated spatio-temporal patterns in order to identify significant structural differences between cold- and warm-seasons along the years.
We extend some existing spatial techniques to the spatio-temporal context to test the null hypothesis that two (or more) observed spatio-temporal point patterns with replications are realisations of point processes that have the same second-order descriptors. In particular, we develop a nonparametric test to approximate the null distribution of the test statistics. We present intensive simulation studies that demonstrate the validity and power of our test and apply our methods to the motivating problem of tornadoes.
TidsskriftJournal of the Royal Statistical Society, Series C (Applied Statistics)
Sider (fra-til)3-23
Antal sider21
StatusUdgivet - 1 jan. 2020

Se relationer på Aarhus Universitet Citationsformater

ID: 160411767