全球大洋中尺度涡致侧向混合系数各向异性分布规律及大尺度效应研究

The lateral mixing induced by meso-scale eddies in the ocean is an important factor in influencing the strength and structure of the large scale ocean circulations. It has long been assumed that the lateral mixing coefficient tensor is isotropic, i.e., it has the same mixing coefficients in two orthogonal directions (e.g., zonal and meridional), as commonly adopted in coarse resolution ocean general circulation models (OGCMs). Recent studies, however, found that the tensor is more anisotropic than isotropic. Yet, the 3D map of the anisotropic mixing coefficients of the global oceans, along with the formation mechanism and its effects on the large scale ocean states, is far from clear. The present program will apply a global ocean 4D-Var (adjoint) data assimilation system and use all the observed data in the recent 10 years (2005-2014) to estimate the optimal anisotropic mixing coefficients along and across the directions of background currents of the global oceans. The to-be-estimated anisotropic mixing coefficients consist of both isopycnal layer thickness diffusivities and along isopycnal diffusivities. With the results, the present program will further investigate the mechanical relationship of them to the structures of simulated circulations, temperature, density and the geography, water depths and so on; the program will also study the effects of them on the large scale ocean states at various time scales, and reveal the effect mechanisms. Finally, the program will propose a parameterization scheme to represent the optimized coefficients in order for coarse-resolution OGCMs to better simulate the oceans.

海洋中尺度涡致侧向混合是影响大尺度海洋环流和水团结构的重要因子。传统认识多认为中尺度涡致侧向混合系数是各向同性的，即在任意相互垂直的侧向方向上（如经向和纬向）具有相等的分量。近期研究发现中尺度涡在不同方向上混合效率是不同的，涡致混合系数存在各向异性。然而，人们对各向异性混合系数的分布规律、控制机理及其大尺度效应等的认识还不深入，也没有对其给出合适的参数化形式。本项目将利用一个全球大洋四维变分（伴随）同化系统，把各向异性的（此研究中选定为沿基本流方向和垂直基本流方向）等密度层厚度扩散系数和沿等密度面扩散系数作为控制变量，通过同化近10年（2005-2014）的海洋观测资料，反演获得上述各向异性扩散系数在全球大洋的三维分布特征，阐明其与大尺度环流、热盐结构及地理地形等因子的机制联系，揭示其对大尺度海洋状态的影响和作用过程，发展针对这些系数的参数化方案，以提高低分辨率海洋和海-气耦合模式的精度。

利用四维变分（伴随）数据同化系统GECCO同化了2007-2016年10年间海表面高度、温度、盐度和内部温盐观测资料，本项目得到了最优的中尺度涡致各向异性混合系数张量（包括等密度层厚度扩散系数和沿等密度面扩散系数）。 给出了各向异性混合系数的全球分布规律。结合全球和区域中尺度涡致输运特征分析，热带不稳定波（涡）不稳定性质分析，初步揭示了各向异性的产生机制。上述各向异性涡致混合系数可较大程度地降低大尺度环流模式的模拟偏差。