中文摘要：

第一部分的结果显示，利用WRF模式及其循环3DVAR同化方案（WRF循环3DVAR）能有效地提高黄海海雾WRF数值模拟的初始场质量。在此基础上，进一步提出利用WRF循环3DVAR形成的初始场驱动RAMS模式的思路，并以2006年3月6～8日的1次大范围黄海海雾事件为研究对象，设计并实施了一系列RAMS数值模拟对比试验。对试验结果的仔细分析表明，WRF循环3DVAR提供的初始场明显优于RAMS模式自身等熵面客观分析方法生成的初始场，它在动力与物理上非常协调且对模拟结果的改善相当显著。这说明WRF循环3DVAR可以为RAMS模式改进其初始场提供一条切实可行的途径。

英文摘要：

The previous work（Part？？？s indicated that a cycling 3DVAR data assimilation scheme based on WRF and its 3DVAR module（WRF cycling 3DVAR）can obviously improve numerical modeling initial conditions of sea fog over the Yellow Sea.In this part,aproposal that RAMS model is forced to run by the initial conditions generated from this scheme is explored,by using several RAMS numerical experiments of a sea fog case over the Yellow Sea during 6？？8March,2006.The modeling results and their analyses show that the initial conditions provided by WRF cycling 3DVAR is much better than that created by the RAMS built-in isentropic analysis,because the former keeps the dynamic and physic balance and results in great improvement of modeling results.It suggests that WRF cycling 3DVAR can be a possible and useful approach to provide initial conditions for RAMS model.