随机强迫对集合预报效果的影响研究

Study of the drift of ensemble forecast effects caused by stochastic forcing

中文关键词:  集合预报  随机强迫  集合平均  Lorenz96模式

英文关键词:Ensemble forecasting  Stochastic forcing  Ensemble mean  Lorenz96 model

中文摘要:

      以Lorenz96模式为动力框架,建立了考虑模式随机强迫不确定性的集合预报系统,并选择模式气候态和集合平均预报效果为研究对象,研究随机强迫对集合预报效果的影响.结果表明,在数值模式积分过程中引入恰当的随机强迫构成的新计算范式,较非随机强迫更接近真值的气候平均与气候标准差,对刻画数值模式的气候态也有正效果;且随机强迫的正效果主要体现在长时效阶段.集合平均预报方面,绝大部分白噪声随机强迫对应的集合预报效果优于非随机强迫集合预报,集合预报效果也随白噪声强迫增大非单调变化,并且非线性系统不同,相同比率的白噪声随机强迫产生的效果也不同.同时,绝大部分红噪声随机强迫对应的集合预报效果也优于非随机强迫集合预报,但仅部分φ(表示所引入外强迫的随机性部分和确定性部分相互耦合的一个度量)值对应的红噪声强迫集合预报优于白噪声随机强迫集合预报;而且红噪声随机强迫集合预报改善效果随系数的正负分布非对称且非单调变化.此外,相关系数φ的选择也依赖于模型.

英文摘要:

      Based on the dynamic framework of the Lorenz 96 model,the ensemble prediction system(FPS) containing stochastic forcing has been developed.In this system,the model climate state and ensemble mean forecasts are chosen to study the scientific problems of noise-induced drift of FPS effects.The results show that the proper stochastic forcing being introduced into numerical model integration process new computing paradigm by means of the is closer to the true values of the climatic mean and standard deviation than the non-stochastic models.In another word,numerical model integration process with stochastic forcing has positive effect on model climate state, and the effect is found to be positive mainly for the longer lead times.Meanwhile,with respect to ensemble forecast effect yielded by white noise stochastic forcing, most of the results are better than those provided by no-stochastic forcing and improvements pertaining to white noise stochastic forcing vary non-monotonically with the increasing size of white noise.Moreover, the effects made by the identical white noise stochastic forcing also vary from nonlinear system to nonlinear system.Also, With respect to FPS effect yielded by red noise stochastic forcing, most of the results are better than those provided by no-stochastic forcing, but only a part of ensemble forecast influenced by red noise are superior to results influenced by white noise.Finally, improvements pertaining to red noise stochastic forcing vary non-symmetrically and non-monotonically with the distribution of coefficient φ.And the selection of coefficient φ is also dependent on nonlinear models.