Extreme events in a changing climate: Variability is more important than averages

Abstract

Extreme events act as a catalyst for concern about whether the climate is changing. Statistical theory for extremes is used to demonstrate that the frequency of such events is relatively more dependent on any changes in the variability (more generally, the scale parameter) than in the mean (more generally, the location parameter) of climate. Moreover, this sensitivity is relatively greater the more extreme the event. These results provide additional support for the conclusions that experiments using climate models need to be designed to detect changes in climate variability, and that policy analysis should not rely on scenarios of future climate involving only changes in means.

Spatial and Temporal Variations of Precipitation Extremes and Seasonality over China from 1961~2013

“…Katz et al [2] stated that changes in extreme climate events are relatively more sensitive to the variability of climate than to its average over many regions in the context of significant global warming. Moreover, the impact of changes of extreme climate events on society, the economy and the environment is far greater than the impact of changes of average climate conditions.…”

Extreme Value Statistics of the Total Energy in an Intermediate-Complexity Model of the Midlatitude Atmospheric Jet. Part I: Stationary Case

“…For example, the role of time-dependence in the statistics of extreme weather events has been at the heart of discussions about climate change since the work by Katz and Brown (1992). In particular, the detection of trends in the frequency of intense precipitation has been the object of much research, particularly at regional level, see e.g.…”

Recent Changes in Frost Days and the Frost-Free Season in the United States

“…For example, Mearns et al (1984) suggest that with an approximate 2°C warming in Iowa, the probability of summertime heat waves would triple; and Katz and Brown (1992) show that, statistically speaking, a change in variance is more important to the frequency of extreme events than a change in the mean. Therefore, if "extreme" is defined with the current temperature distribution, then shifting the entire distribution by raising the mean value with no change in variance will greatly impact the probability of occurrence of the extreme (Meehl et al 2000), as does an increase in the variance with or without an increase in the mean (Katz and Brown 1992). With this in mind, the purpose of this study is to examine changes in frost days, as defined by days when the minimum temperature is below 0°C, and changes in the frost-free season for the continental United States for the latter half of the twentieth century.…”