On the Evaluation of Data Quality in the OMNI Interplanetary Magnetic Field Database
On the Evaluation of Data Quality in the OMNI Interplanetary Magnetic Field Database
复制标题
OMNI行星际磁场数据库数据质量评价
DOI：
10.1029/2018sw002113
复制
发表时间：
2019
期刊：
影响因子：
3.7
通讯作者：
V. Sergeev
中科院分区：
文献类型：
--
作者：
M. Vokhmyanin;N. Stepanov;V. Sergeev
OMNI数据库是将在拉格朗日L1点附近测得的太阳风传播而成的，其结果可能与弓形激波鼻附近的实际太阳风不同。为了测试OMNI数据库的质量，我们将主要由ACE和WIND航天器提供的1分钟行星际磁场（IMF）数据的2小时间隔与弓形激波前的Geotail测量值（1997-2016年为10，409例）进行了交叉关联。我们使用了两个指标：皮尔逊相关系数（CC）和预测效率（PE）。重复以前的研究，我们发现实际IMF的预测质量随着OMNI航天器与日地线距离的增加而不断下降，对于RYZ ≥ 65 RE，差预测和好预测的数量几乎相等（它们占整个数据库的约12%）。在大约20%的分析数据中，低CC和PE值是低IMF变异性（低信噪比）的结果。剩余的数据集包括42%的非常好的数据（CC ≥ 0.8），33%的相对好的数据（0.5 ≤ CC < 0.8和PE ≥ 0），10%的数据具有正确的变异性但绝对值错误（0.5 ≤ CC < 0.8和PE < 0），以及15%的不良数据（CC < 0.5）。我们还发现，当地磁活动的PC指数与Kan和Lee（1979，https：//doi.org/10.1029/GL006i007p00577）提出的太阳风-磁层耦合因子良好相关时，OMNI数据通常质量良好。
The OMNI database is formed by propagating the solar wind measured at around Lagrange point L1, whose result may differ from the actual solar wind in the vicinity of the bow shock nose. To test the quality of the OMNI database, we cross‐correlate the 2‐hr intervals of 1‐min interplanetary magnetic field (IMF) data provided mostly by ACE and WIND spacecraft with Geotail measurements in front of the bow shock (10,409 cases in 1997–2016). We used two metrics: Pearson correlation coefficient (CC) and prediction efficiency (PE). Confirming previous studies, we found that the prediction quality of actual IMF degrades continuously with increasing distance of OMNI spacecraft from the Sun‐Earth line, with the amounts of poor and good predictions become nearly equal for RYZ ≥ 65 RE (they constitute ~12% of the entire database). In roughly 20% of the analyzed data, low CC and PE values were the consequence of low IMF variability (a low signal‐to‐noise ratio). The remaining data set includes 42% of very good data (CC ≥ 0.8), 33% of relatively good data (0.5 ≤ CC < 0.8 and PE ≥ 0), 10% of data having correct variability but wrong absolute values (0.5 ≤ CC < 0.8 and PE < 0), and 15% of poor data (CC < 0.5). We also discovered that the OMNI data are generally of a good quality when the PC index of geomagnetic activity correlates well with the solar wind‐magnetosphere coupling factor suggested by Kan and Lee (1979, https://doi.org/10.1029/GL006i007p00577).