DOI: 10.1007/s00382-015-2517-1
Scopus记录号: 2-s2.0-84947489400
论文题名: Probabilistic tail dependence of intense precipitation on spatiotemporal scale in observations, reanalyses, and GCMs
作者: Cavanaugh N.R. ; Gershunov A.
刊名: Climate Dynamics
ISSN: 9307575
出版年: 2015
卷: 45, 期: 2017-11-12 起始页码: 2965
结束页码: 2975
语种: 英语
英文关键词: Daily intense precipitation
; Extreme value theory
; Extremes
; Heavy tails
; Scale dependence
英文摘要: Daily precipitation variability as observed from weather stations is heavy tailed at most locations around the world. It is thought that diversity in precipitation-causing weather events is fundamental in producing heavy-tailed distributions, and it arises from theory that at least one of the precipitation types contributing to a heavy-tailed climatological record must also be heavy-tailed. Precipitation is a multi-scale phenomenon with a rich spatial structure and short decorrelation length and timescales; the spatiotemporal scale at which precipitation is observed is thus an important factor when considering its statistics and extremes. In this study, we examine the spatiotemporal scaling behavior of intense precipitation from point-scale to large grid cells and from 1 day to 4 weeks over the entire globe. We go on to validate the current generation of historically-forced climate models and reanalyses against observational data at consistent spatial scales. Our results demonstrate that the prevalence and magnitude of heavy tails in observations decrease when moving to lower spatiotemporal resolutions, as is consistent with stochastic theory. Reanalyses and climate models generally reproduce large, synoptic scale distribution classifications, but struggle to reproduce the statistics in regions that are strongly affected by mesoscale phenomena. We discuss these results in relation to physically consistent atmospheric regimes. We conclude with a global view of precipitation distribution type at daily resolution as calculated from the best-performing reanalysis, the Climate Forecast System Reanalysis. © 2015, Springer-Verlag (outside the USA). 资助项目: NSF, National Science Foundation
; NSF, National Science Foundation
Citation statistics:
资源类型: 期刊论文
标识符: http://119.78.100.158/handle/2HF3EXSE/53918
Appears in Collections: 过去全球变化的重建
There are no files associated with this item.
作者单位: Earth Science Division, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Mail Stop 74R316C, Berkeley, CA, United States; Climate Atmospheric Science and Physical Oceanography Division, Scripps Institution of Oceanography, La Jolla, CA, United States
Recommended Citation:
Cavanaugh N.R.,Gershunov A.. Probabilistic tail dependence of intense precipitation on spatiotemporal scale in observations, reanalyses, and GCMs[J]. Climate Dynamics,2015-01-01,45(2017-11-12)