| 摘要: |
| 使用球坐标下1.5 层约化重力浅水模式模拟海洋风生双环流, 结果显示双环流射流存在拉伸模态和收缩模态间的年际变化。以双环流从拉伸模态向收缩模态的转变过程为背景场, 利用条件非线性最优扰动(CNOP)方法, 考察初始误差对双环流变异可预报性的影响, 得到两类初始误差: 全局CNOP型和局部CNOP(LCNOP)型, 两类初始误差对双环流变异的影响几乎相反。通过考察误差发展, 发现在射流从拉伸模态向收缩模态转变过程中, CNOP 型初始误差使射流弯曲程度变大, 并在预报时刻导致涡脱落; 而LCNOP 型初始误差则使射流弯曲程度变小。相比LCNOP, CNOP 型初始误差引起更大预报误差, 导致双环流变异的预报技巧下降更多。两类误差得到较大发展的区域可能存在正压不稳定, 使误差能够不断从背景场吸收能量进而得到快速发展。给出了两类使双环流变异预报技巧下降最大的初始误差, 在实际的数值预报中减少这两种类型的误差, 将有助于提高双环流变异的预报技巧。 |
| 关键词: 双环流变异 条件非线性最优扰动(CNOP) 可预报性 |
| DOI：10.11759/hykx20130304001 |
| 分类号: |
| 基金项目:国家自然科学基金项目(41230420); 中国科学院知识创新工程重要方向项目(KZCX2-EW-201); 青岛市基础研究计划项目(11-1-4-95-jch) |
|
| The impact of initial error on predictability of Double-gyre variability |
|
|
| Abstract: |
| A 1.5-layer reduced-gravity shallow-water model was used to simulate wind-driven double-gyre circulation. The modeling results show that the eastward jet has interannual variability between the contracted path and the elongated path. Taking the process when the double-gyre was changing from the elongated path into the contracted path as the background field, we investigated the impact of initial error on predictability of double-gyre variability by using conditional nonlinear optimal initial perturbation (CNOP) method. We got two kinds of initial errors: CNOP and LCNOP. Results show that CNOP and LCNOP affect the predictability of double-gyre variability oppositely. Through observing the development of the optimal initial error, we found that when the jet was changing from the elongated path into the contracted path, the CNOP strengthened the meander and at last an eddy dropped; however, the LCNOP weakened the meander. Moreover, the prediction error caused by CNOP is greater than that caused by LCNOP. We also found the areas where the optimal initial error got quick evolution generally have large velocity shear which probably results in barotropic instability. As a result, the initial error could continuously get energy from the background field and then develop into big eddies. |
| Key words: double-gyre variability conditional nonlinear optimal perturbation(CNOP) 1.5 layer shallow-water model predictability |
