Mechanisms and predictability of the dominant atmospheric variability : An approach using a hierarchical modeling

我们开发了一组复杂程度不同的大气模型。它们被共同用于研究主导环流变化的机制，并阐明了例如夏季遥相关型的强迫过程。我们还开发了一个基于集合卡尔曼滤波的试验预报系统。通过扩展系统使模式的参数被同化和摄动，我们显示了1-mo预报技巧的改进，这是由于模式误差的减少。

We developed a group of atmospheric models having different levels of complexity. They are used cooperatively to examine the mechanisms of dominant circulation variability and clarified, for example, forcing processes for the summertime teleconnection patterns. We also developed an experimental forecast system based on the ensemble Kalman filter. By extending the system such that the model's parameters are assimilated and perturbed, we showed an improvement in the 1-mo forecast skill, which is due to the reduction of the model error.