DOI: 10.1175/JCLI_D_12_00521.1
Scopus记录号: 2-s2.0-84886303975
论文题名: Climate drift in the CMIP5 models
作者: Gupta A.S. ; Jourdain N.C. ; Brown J.N. ; Monselesan D.
刊名: Journal of Climate
ISSN: 8948755
出版年: 2013
卷: 26, 期: 21 起始页码: 8597
结束页码: 8615
语种: 英语
Scopus关键词: Biogeochemical models
; Coupled Model Intercomparison Project
; Drift-correction methods
; Internal variability
; Model comparison
; Model evaluation/performance
; Ocean biogeochemistry
; Ocean model
; Biogeochemistry
; Climate change
; Computer simulation
; Concurrency control
; Estimation
; Models
; Sea level
; Climate models
; biogeochemistry
; climate modeling
; computer simulation
; drift behavior
; historical perspective
; marine atmosphere
; performance assessment
英文摘要: Climate models often exhibit spurious long-term changes independent of either internal variability or changes to external forcing. Such changes, referred to as model "drift," may distort the estimate of forced change in transient climate simulations. The importance of drift is examined in comparison to historical trends over recent decades in the Coupled Model Intercomparison Project (CMIP). Comparison based on a selection of metrics suggests a significant overall reduction in the magnitude of drift from phase 3 of CMIP (CMIP3) to phase 5 of CMIP (CMIP5). The direction of both ocean and atmospheric drift is systematically biased in some models introducing statistically significant drift in globally averaged metrics. Nevertheless, for most models globally averaged drift remains weak compared to the associated forced trends and is often smaller than the difference between trends derived from different ensemble members or the error introduced by the aliasing of natural variability. An exception to this is metrics that include the deep ocean (e.g., steric sea level) where drift can dominate in forced simulations. In such circumstances drift must be corrected for using information from concurrent control experiments. Many CMIP5 models now include ocean biogeochemistry. Like physical models, biogeochemical models generally undergo long spinup integrations to minimize drift. Nevertheless, based on a limited subset of models, it is found that drift is an important consideration and must be accounted for. For properties or regions where drift is important, the drift correction method must be carefully considered. The use of a drift estimate based on the full control time series is recommended to minimize the contamination of the drift estimate by internal variability. © 2013 American Meteorological Society. Citation statistics:
资源类型: 期刊论文
标识符: http://119.78.100.158/handle/2HF3EXSE/51536
Appears in Collections: 气候变化事实与影响
There are no files associated with this item.
作者单位: ARC Centre of Excellence for Climate Systems Science, University of New South Wales, Sydney, NSW, Australia; Climate Change Research Centre, University of New South Wales, Sydney, NSW, Australia; Centre for Australian Weather and Climate Research, CSIRO Wealth from Oceans National Research Flagship, Hobart, TAS, Australia
Recommended Citation:
Gupta A.S.,Jourdain N.C.,Brown J.N.,et al. Climate drift in the CMIP5 models[J]. Journal of Climate,2013-01-01,26(21)