The impact of using wind observations from the Aeolus satellite in a limited-area numerical weather prediction (NWP) system is being investigated using the limited-area NWP model Harmonie–Arome over the Nordic region. We assimilate the horizontal line-of-sight (HLOS) winds observed by Aeolus using 3D-Var data assimilation for two different periods, one in September–October 2018 when the satellite was recently launched and a later period in April–May 2020 to investigate the updated data processing of the HLOS winds. We find that the quality of the Aeolus observations has degraded between the first and second experiment period over our domain. However, observations from Aeolus, in particular the Mie winds, have a clear impact on the analysis of the NWP model for both periods, whereas the forecast impact is neutral when compared against radiosondes. Results from evaluation of observation minus background and observation minus analysis departures based on Desroziers diagnostics show that the observation error should be increased for Aeolus data in our experiments, but the impact of doing so is small. We also see that there is potential improvement in using 4D-Var data assimilation, which generates flow-dependent analysis increments, with the Aeolus data.

中文翻译：

---

评估 Aeolus 卫星观测在区域数值天气预报 (NWP) 模型 Harmonie-Arome 中的使用

正在使用北欧地区的有限区域 NWP 模型 Harmonie-Arome 研究在有限区域数值天气预报 (NWP) 系统中使用来自 Aeolus 卫星的风观测的影响。我们使用 3D-Var 数据同化在两个不同时期同化了 Aeolus 观测到的水平视线 (HLOS) 风，一个是在 2018 年 9 月至 10 月卫星最近发射时，另一个是在 2020 年 4 月至 5 月，以进行调查HLOS 风的更新数据处理。我们发现 Aeolus 观测的质量在我们域中的第一个和第二个实验期间有所下降。然而，来自 Aeolus 的观测，尤其是 Mie 风，对两个时期的 NWP 模型分析都有明显影响，而与无线电探空仪相比，预测影响是中性的。基于 Desroziers 诊断的观测减去背景和观测减去分析偏差的评估结果表明，在我们的实验中，Aeolus 数据的观测误差应该增加，但这样做的影响很小。我们还看到，在使用 4D-Var 数据同化时有潜在的改进，这会产生与 Aeolus 数据相关的流量分析增量。