无缝隙预报不仅是数值预报的前沿发展方向，也是国家防灾减灾的需求。而耦合资料同化是无缝隙预报的关键环节之一。国际国内已有的耦合同化系统主要采用弱耦合同化，极少使用跨圈层的背景误差协方差，以致单圈层的观测资料不能直接用于其他圈层的同化，这不仅导致观测资料得不到充分使用，而且产生的耦合初值在圈层间存在不协调。因此，发展强耦合同化是未来发展的必然趋势，其中跨圈层背景误差协方差矩阵的构建是关键。然而，跨圈层背景误差协方差的研究仍然存在时空分布不均匀以及由于不同圈层时空尺度差异较大导致的虚假关联等问题。因此，本项目拟以圈层界面为桥梁，研究圈层间误差的物理关联，缓减不同圈层时空尺度不一致给跨圈层背景误差协方差构建带来的困难，以此为基础，建立强耦合同化系统，并通过年代际预测试验，检验和评估跨圈层背景误差协方差对提高预测技巧的作用和贡献。项目预期为我国未来无缝隙预报系统的发展打下好的基础。

Seamless prediction is not only the frontier of numerical prediction, but also the need of prevention and reduction of natural disasters. Coupled data assimilation (CDA) is one of the key aspects of the seamless prediction. The available CDA systems home and abroad mainly use the weakly CDA (WCDA), rarely using cross sphere background error covariance (CSBEC). As a result, observations in a sphere cannot directly be assimilated into the variables of other spheres, leading to the insufficient use of them as well as the inconsistence of the initial conditions among / between different spheres. Therefore, developing strongly CDA (SCDA) is an inevitable trend of future development, where the construction of CSBEC is very important. However, there are still some problems in the CSBEC studies, such as the inhomogeneous spatiotemporal distributions, spurious relations between spheres due to quite different spatiotemporal scales in these spheres, and so on. For this reason, this project is going to study the physical error relations between spheres using the interfaces of these spheres as bridges, so as to alleviate the difficulty in construction of CSBEC caused by their quite different spatiotemporal scales. Then, the project will build a SCDA system, and evaluate the contribution of CSBEC to the increase of skill in decadal predictions. The project is expected to lay a good foundation for the future development of seamless forecasting system in China.