This paper presents the variational diagnostic model and iterative procedure, which enables the wind field in subdomains to be adjusted. Diagnostic models are not time dependent. Consideration of more complex features of the thermodynamic structure requires models with high resolution, which require large calculation times. The model presented applies the variational approach and enables topographical complexity of the terrain to be considered. The problem of adjusting the wind field is solved in two steps. The first step adjusts the initial wind field by means of experimental measurements or a prognosis in the larger domain, which includes smaller domains. Then the results obtained are used as the initial wind field when the grid refinement in the smaller domain is performed. This allows more precise mapping of the terrain and its architecture. Nevertheless the algorithm proposed ensures a considerable reduction in calculation time. This approach also allows us to eliminate the problem of the lack of initial data when the number of meteorological stations in the smaller domain is insufficient. The algorithm is described and validated, and numerical simulations for pollutant dispersion for a chosen town are described, followed by discussion of the iterative procedure.

本文提出了变分诊断模型和迭代程序，可以调整子域中的风场。诊断模型不依赖于时间。考虑热力学结构的更复杂特征需要具有高分辨率的模型，这需要大量的计算时间。提出的模型应用了变分方法，并能够考虑地形的地形复杂性。调整风场的问题分两步解决。第一步通过实验测量或在更大域中的预测来调整初始风场，其中包括更小的域。然后将得到的结果作为在较小域进行网格细化时的初始风场。这允许更精确地映射地形及其架构。然而，所提出的算法确保了计算时间的显着减少。这种方法也可以让我们在较小域的气象台站数量不足时消除初始数据缺乏的问题。描述和验证了算法，描述了选定城镇污染物扩散的数值模拟，然后讨论了迭代过程。