The Critical Success Index as an Indicator of Warning Skill

Abstract

A form of the critical success index ( CSI) is used by the National Weather Service to indicate the value of warnings. This verification statistic assumes that the times wh�n an are of no consequence. It can be shown that the CSI is not an unbiased mdicator of fo� skill but 1s �ropomonal to the frequency of the event being forecast. This innate bias is demons�ted theoretically and :"'a example. . ev�nt was neither ex�ted nor o�rved . An unbiased verification statistic appropriate for forecasts of rare events 1s presented and applied to severe convective weather warnings. Comparisons of this score to the CSI show the extent of the penalty the CSI extracts from forecasters who work in areas that are not climatically prone to gjven events.