State Key Laboratory of Numerical Modeling for Atmospheric Sciences and
Geophysical Fluid Dynamics (LASG)
Institute of Atmospheric Physics, Chinese Academy of Sciences
Vol. 4/No. 26 December 2017
[Climate Predictability] The predictability of atmospheric and oceanic motions: Retrospect and prospects
With the development of atmospheric and oceanic science and the deepening understanding of the weather and climate forecasts (predictions), scientists have a great insight into the predictability. The connotation of predictability is more reasonable and is more suitable for the demands of not only weather but also climate study. However, there still remain many challenging issues that require further study.
Recently, Profs. Mu Mu (Fudan University), Duan Wansuo (Institute of Atmospheric Physics), and Tang Youmin (Second Institute of Oceanology) review discussions on the concept of predictability, and extend the concept of the traditional predictability of weather and climate to that of the general physical variables and events of coupled atmosphere and ocean systems. Based on this new definition of predictability, they further summarizes the challenging and urgent to be solved problems of the predictability of weather and climate phenomenon of different spatial and temporal scales, which include the intrinsic predictability limit (IPL) problems, initial errors problems, model error problems and the targeted-observation problems, as well as the predictability problems of the probabilistic forecasting. These issues presented in this paper could provide a useful reference for readers who wish to examine weather and climate predictability.
Figure 1 The ratio of root mean square of the forecast errors from the different initial errors and different magnitudes to the root mean square of the climate variance, i.e., the relative error. The smaller the initial error is, the faster the growth of the initial error will be; All the relative errors exceed 0.5 at around two weeks, no matter what kind of initial errors there are. This is when the weather signal is exceeded by the forecast errors, and the predictability is lost. This suggests that the IPL of the numerical weather forecast is about two weeks (the figure is from Kalnay (2011))
This study is published in Science China Earth Sciences (doi: 10.1007/s11430-016-9101-x).
Reference: Mu Mu, Wansuo Duan, and Youmin Tang, 2017: The predictability of atmospheric and oceanic motions: Retrospect and prospects. Science China Earth Sciences, Vol.60 No.11:2001–2012. doi: 10.1007/s11430-016-9101-x.
Contact: Mu Mu, firstname.lastname@example.org
Editors:Wangchuanyi(email@example.com); Zhouwenling(firstname.lastname@example.org); Lisiying(email@example.com)