Data-driven modeling of the solar wind from 1 R-s to 1 AU
Data-driven modeling of the solar wind from 1 R-s to 1 AU
复制标题
从 1 R-s 到 1 AU 的太阳风数据驱动建模
DOI：
10.1002/2015ja021911
复制
发表时间：
2015
影响因子：
2.8
通讯作者：
Xiang Changqing
中科院分区：
文献类型：
--
作者：
Feng Xueshang;Ma Xiaopeng;Xiang Changqing
我们在这里提出了一个随时间变化的三维磁流体动力学（MHD）太阳风模拟从太阳表面到地球的轨道驱动的随时间变化的视线太阳磁场数据。模拟基于三维（3‐D）太阳行星际（SIP）自适应网格加密（AMR）时空守恒单元和求解单元（CESE）MHD（SIP‐AMR‐CESE MHD）模型。在这个模拟中，我们首先通过输入从天气光球磁场获得的位场，用时间松弛方法获得初始太阳风背景，然后通过用连续变化的光球磁场推进初始3-D太阳风背景来生成随时间变化的太阳风。模型采用投影法向特征线法修正内边界条件，输入经太阳自转差修正的高节奏光球磁场数据，限制光球逃逸的质量通量。利用全球振荡网络连续天气图驱动的模式研究了2008年7月1日至8月11日太阳风的演变。我们将数值结果与前人对太阳风的研究、太阳和日光层天文台极紫外成像望远镜板的日冕观测以及OMNI在1天文单位（Au）的观测结果进行了比较。比较结果表明，该模型的结果与大尺度日冕和行星际动力学结构，包括冕洞的大小和分布、流光带的位置和形状、Alfvénic表面的日心距离以及太阳风速度的转变等，具有较好的一致性。然而，该模型未能捕捉到小尺寸的赤道孔，并且在1 Au附近模拟的太阳风具有比观测到的更高的密度和更弱的磁场强度。也许对高节奏观测到的光球磁场（特别是三维全球测量）进行更好的预处理，结合等离子体测量和更高分辨率的网格，将使数据驱动模型能够更准确地捕捉环境太阳风的时间依赖性变化，以进一步改进。此外，在太阳活动高峰期使用该模型时，还可能需要其他措施。
We present here a time‐dependent three‐dimensional magnetohydrodynamic (MHD) solar wind simulation from the solar surface to the Earth's orbit driven by time‐varying line‐of‐sight solar magnetic field data. The simulation is based on the three‐dimensional (3‐D) solar‐interplanetary (SIP) adaptive mesh refinement (AMR) space‐time conservation element and solution element (CESE) MHD (SIP‐AMR‐CESE MHD) model. In this simulation, we first achieve the initial solar wind background with the time‐relaxation method by inputting a potential field obtained from the synoptic photospheric magnetic field and then generate the time‐evolving solar wind by advancing the initial 3‐D solar wind background with continuously varying photospheric magnetic field. The model updates the inner boundary conditions by using the projected normal characteristic method, inputting the high‐cadence photospheric magnetic field data corrected by solar differential rotation, and limiting the mass flux escaping from the solar photosphere. We investigate the solar wind evolution from 1 July to 11 August 2008 with the model driven by the consecutive synoptic maps from the Global Oscillation Network Group. We compare the numerical results with the previous studies on the solar wind, the solar coronal observations from the Extreme ultraviolet Imaging Telescope board on Solar and Heliospheric Observatory, and the measurements from OMNI at 1 astronomical unit (AU). Comparisons show that the present data‐driven MHD model's results have overall good agreement with the large‐scale dynamical coronal and interplanetary structures, including the sizes and distributions of the coronal holes, the positions and shapes of the streamer belts, the heliocentric distances of the Alfvénic surface, and the transitions of the solar wind speeds. However, the model fails to capture the small‐sized equatorial holes, and the modeled solar wind near 1 AU has a somewhat higher density and weaker magnetic field strength than observed. Perhaps better preprocessing of high‐cadence observed photospheric magnetic field (particularly 3‐D global measurements), combined with plasma measurements and higher resolution grids, will enable the data‐driven model to more accurately capture the time‐dependent changes of the ambient solar wind for further improvements. In addition, other measures may also be needed when the model is employed in the period of high solar activity.