| 摘要: |
| 深层叶绿素最大值（Deep Chlorophyll Maximum，DCM）现象的数值模拟是研究海洋表层生态系统和全球碳循环的重要组成部分之一。但是由于自身的复杂性和观测的局限性，数值模式中物理参数的不确定性给模拟结果带来了一定程度的误差。其中，垂向湍流扩散系数（vertical turbulence diffusion）是模式所包含的物理参数中很难直接通过观测来确定的参数，它在模式中的来源和取值往往具有很大的不确定性。本文通过条件非线性最优（参数）扰动（Conditional nonlinear optimal perturbation related to parameter，CNOP-P）方法，研究了垂向湍流扩散系数的不确定性对模式模拟结果的影响。我们发现，垂向湍流扩散系数对DCM模拟产生最大影响的CNOP型扰动位于生产力层的上半部分。并且，去掉生产力层内湍流扩散系数的误差，模式模拟的改进程度最高达到了80%。可见，垂向湍流扩散对生态系统的发展和保持起着极其重要的作用，改进垂向湍流扩散系数的不确定性，对DCM的数值模拟有着重要意义。 |
| 关键词: 物理参数 湍流扩散系数 不确定性 条件非线性最优扰动(CNOP) |
| DOI：10.11759/hykx20180911001 |
| 分类号:P731.26 |
| 基金项目:国家自然科学基金委员会-山东省人民政府海洋科学研究中心联合资助项目（U1606402）；国家自然科学基金委创新群体项目（41421005） |
|
| A study of uncertainty related to the coefficient of vertical turbulence diffusion in ocean ecosystem model |
|
GAO Yong-li1,2,3
|
|
1.CAS Key Laboratory of Ocean Circulation and Waves, Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, China;2.University of Chinese Academy of Sciences, Beijing 100049, China;3.China University of Petroleum(East China), Qingdao 266580, China
|
| Abstract: |
| The simulation of deep chlorophyll maximum (DCM) is one of the most important parts in the study of the ocean surface ecosystem and global carbon cycle. However, the complexity and limited observation of the physical parameters in the numerical model usually cause various errors in the output results. Among such parameters, the coefficient of vertical turbulence diffusion is difficult to be directly determined by observation; therefore, its uncertainty in the model is very large. This study investigates the influence of turbulence diffusion on the numerical model output results using the conditional nonlinear optimal perturbation method (related to parameters). We determined that the strongest perturbation of the vertical turbulence diffusion occurred in the productivity layer, and proved that eliminating this perturbation resulted in significant improvement of the model output. This illustrates that the physical condition is very important for the development and maintenance of the ocean ecosystem model. |
| Key words: physical parameters vertical turbulence diffusion uncertainty CNOP |
