非线性动力学方法在气候突变检测中的应用

快速、准确的检测气候突变, 对于我们认识气候系统的变化和对未来气候系统演变趋势的预测有着重要的现实意义和社会经济价值. 本文主要回顾了近年来非线性突变检测技术的主要研究进展及其在实际观测资料中的应用, 其中包括基于气候系统长程相关性的检测方法, 如滑动去趋势波动分析方法、滑动移除去趋势波动分析方法、滑动移除重标极差方法和指纹法等; 以及基于时间序列复杂性的检测方法, 如近似熵方法, Fisher信息和小波Fisher信息等. 此外, 本文还指出发展针对空间场的突变检测技术是未来一个可能的发展方向. 由于空间场所包含的气候系统的演变信息远高于单点时间序列, 空间场的突变检测技术将会使得对气候突变的检测时间大大缩短, 从而使得人们有足够的时间去采取行动, 以便为适应气候突变所带来的新挑战做好准备.

Application of nonlinear dynamical methods in abrupt climate change detection

The research of abrupt climate change is an important field in the climate change. The rapid and accurate detection of the abrupt climate change has important practical significance and major economic-social costs, which will help us understand climate change and forecast the future evolutionary trend of the climate system. The detection results of most traditional abrupt climate change depend on the selection of the time scale concerned, which may result in the fact that we cannot identify an abrupt climate change until the event has been past for a long time. Moreover, these detection methods cannot extract the dynamical changes from the observational data of the climate system. As the rapid development in nonlinear science, the abrupt climate change detection technology has also been improved gradually. This article briefly reviews several new progresses in abrupt dynamical detection methods developed on the basis of recent nonlinear technologies, and some applications in the real observational data. These new methods mainly contain the technologies based on the long-range correlation of climate systems, such as moving detrended fluctuation analysis, moving cut data-detrended fluctuation analysis, moving cut data-R/S analysis, degenerate fingerprinting, and red noise. Moreover, some abrupt dynamical detection methods developed by the complexity of the time series, namely, entropy, such as approximate entropy, moving cutting data-approximate entropy, Fisher information, and wavelet Fisher's information measure. Furthermore, there are some other abrupt dynamical detection methods based on the theory of phase space, such as the dynamics exponent Q. Climate system is a complex dynamical system with nonlinear and interactive nature, which has long-range persistence in spatio-temporal variation, thus the abrupt detection method on spatial field change is pointed out to be a promising direction for further research in future. Because the spatial field contains abundance of information about the evolution of climate system which is much more than that in a time series in single meteorological station, the detecting methods on spatial field will greatly help us detect an abrupt climate change as soon as possible. And then we will have enough time to take action and make preparations for the new challenges due to the abrupt climate change.