Abstract We propose a new ‘Bi-Reduced Space’ approach to solving 3D Variational Data Assimilation using Convolutional Autoencoders. We prove that our approach has the same solution as previous methods but has significantly lower computational complexity; in other words, we reduce the computational cost without affecting the data assimilation accuracy. We tested our proposal with data from a real-world application: a pollution model of a site in Elephant and Castle (London, UK) and found that we could (1) reduce the size of the background covariance matrix representation by O ( 1 0 3 ) , and (2) increase our data assimilation accuracy with respect to existing reduced space methods.

中文翻译：

---

用于 3D 变分数据同化的基于注意力的卷积自编码器

摘要 我们提出了一种新的“双缩减空间”方法来使用卷积自动编码器解决 3D 变分数据同化。我们证明了我们的方法与以前的方法具有相同的解决方案，但计算复杂度显着降低；换句话说，我们在不影响数据同化精度的情况下降低了计算成本。我们使用来自现实世界应用程序的数据测试了我们的建议：Elephant and Castle（英国伦敦）的一个站点的污染模型，发现我们可以 (1) 将背景协方差矩阵表示的大小减小 O ( 1 0 3 ) 和 (2) 相对于现有的缩减空间方法提高了我们的数据同化精度。