自动行星际激波检测及其在Wind观测中的应用-CSDN博客

本文链接：https://blog.csdn.net/u014546828/article/details/146552797

We present an automated two-step detection algorithm for identification of interplanetary (IP) shocks regardless their type in a real-time data stream. This algorithm is aimed for implementation on board the future Solar Orbiter mission for triggering the transmission of the high-resolution data to the Earth. The first step of the algorithm is based on a determination of a quality factor, Q indicating abrupt changes of plasma parameters (proton density and bulk velocity) and magnetic field strength. We test two sets of weighting coefficients for Q determination and propose the second step consisting of three additional constraints that increase the effectiveness of the algorithm. We checked the algorithm using Wind (at 1 AU) and Helios (at distances from 0.29 to 1 AU) data and compared obtained results with already existing lists of IP shocks. The efficiency of the presented algorithm for the Wind shock lists varies from 60% to 84% for two Q thresholds. The final shock candidate list provided by the presented algorithm contains the real IP shocks, as well as different discontinuities. The detection rate of the IP shocks equals to 64% and 29% for two Q thresholds. The algorithm detected all IP shocks associated with the solar wind transient structures triggering intense (Dst<−100 nT) geomagnetic storms.

本文提出了一种自动化的两步检测算法，用于在实时数据流中识别行星际 (IP) 激波，而无需考虑其类型。该算法旨在未来太阳轨道器任务 (Solar Orbiter mission) 中实施，以触发高分辨率数据向地球的传输。算法的第一步基于一个质量因子 Q 的确定，Q 用于指示等离子体参数（质子密度和整体速度）以及磁场强度的突变 。本文测试了两组用于确定 Q 的加权系数，并提出了第二步，其中包括三个额外的约束条件，以提高算法的有效性。本文使用 Wind（位于 1 AU） 和 Helios（距离范围为 0.29 至 1 AU） 数据对算法进行了验证，并将结果与现有的 IP 激波列表进行了比较。对于两个 Q 阈值，本算法针对 Wind 激波列表的效率分别为 60% 至 84%。由本算法生成的最终激波候选列表包含真实的 IP 激波以及其他类型的不连续性。对于两个 Q 阈值，IP 激波的检测率分别为 64% 和 29%。此外，该算法成功检测到了所有与引发强烈（Dst < -100 nT）地磁暴的太阳风瞬态结构相关的 IP 激波。

Introduction

Interplanetary (IP) shocks are formed in the solar wind as precursors of the arrival of large transient structures of solar origin due to nonlinear effects [Tsurutani et al., 1988; Tsurutani et al., 2011]. They are generated by the interaction of fast and slow solar wind streams (specifically at the boundaries of corotating interaction regions, CIRs) and by the passage of transient phenomena such as Coronal Mass Ejections (CMEs) propagating through the interplanetary medium [e.g., Luhmann, 1997] as Interplanetary Coronal Mass Ejections (ICMEs).

Solar Orbiter was selected as one of the missions within the European Space Agency Cosmic Vision 2015–2025 programme [Müller et al., 2012]. Following the launch scheduled for January 2017, the Solar Orbiter spacecraft will orbit the Sun at distances reaching 0.28 AU by the end of the mission. The spacecraft will carry a payload including the remote sensing instruments tasked to monitor the dynamics of the Sun and its surface layers and the in situ instruments, which will study the particles, fields, and waves in the solar wind immediately above the remotely observed source regions. The payload is suitable to register IP shocks together with their drivers; therefore, it may significantly contribute to our further understanding of shock formations, their fine structure, and a connection with a particular driver [March et al., 2005], but these tasks require high-time resolution measurements. The present experimental techniques can provide the data with a sufficient time resolution, but data transmission rates allow only a limited sample of such data to be returned to the Earth.

行星际 (IP) 激波 是由于非线性效应在太阳风中形成的，作为太阳起源的大尺度瞬态结构到达的前兆。它们由快速和慢速太阳风流的相互作用（特别是在共转交互区域 (CIRs) 的边界处）以及通过行星际介质传播的瞬态现象（如日冕物质抛射 (CMEs)）产生，这些现象表现为 行星际日冕物质抛射 (ICMEs) 。

太阳轨道器 (Solar Orbiter) 被选为欧洲航天局“宇宙愿景 2015-2025”计划中的任务之一。按计划于 2017 年 1 月发射后，太阳轨道器将围绕太阳运行，并在任务结束时达到 0.28 AU 的距离。该航天器将携带一组有效载荷，包括用于监测太阳及其表层动态的遥感仪器，以及用于研究太阳风中粒子、场和波动的 原位仪器 ，这些仪器位于遥感观测源区的上方。有效载荷能够同时记录 IP 激波及其驱动源；因此，它可能显著促进本文对激波形成、其精细结构以及与特定驱动源的关联性的进一步理解，但这些任务需要高时间分辨率的测量。目前的实验技术可以提供足够时间分辨率的数据，但数据传输速率仅允许有限的样本返回地球。

One of the possible solutions is a trigger system operating onboard Solar Orbiter that would analyze data, select predetermined events according to the specified criteria, and transmit an identified sample of the data with the highest time resolution to the Earth.

This idea is not new; the first algorithm for IP shock detection based on magnetic field measurements was applied on Helios-1 and Helios-2 [Musmann et al., 1979]. The Helios automated event detector continuously computed the quality index A from a relative change of the squared magnetic field magnitude and compared short-time mean values with long-time mean values. Due to the memory shortage, this algorithm stored high-time resolution data for the event with the highest quality index between two telemetry sessions. However, the described algorithm was extremely sensitive to short-time fluctuations of the magnetic field that led to a great number of false IP shock detections or to registration of other solar wind discontinuities. Similar IP shock detection systems were developed later for the Intershock [Galeev et al., 1986] and ISEE [Joselyn et al., 1981] spacecraft.

一种可能的解决方案是开发一个在太阳轨道器上运行的触发系统，该系统将分析数据，根据指定标准选择预定事件，并以最高时间分辨率向地球传输选定的数据样本。

这个想法并不新鲜；第一个基于磁场测量的 IP 激波检测算法曾在 Helios-1 和 Helios-2 上应用。Helios 自动事件探测器通过计算磁场强度平方的相对变化连续计算质量指数 A，并将短时间均值与长时间均值进行比较。由于存储空间不足，该算法仅存储两次遥测会话之间具有最高质量指数事件的高时间分辨率数据。然而，所描述的算法对磁场的短期波动极为敏感，导致大量错误的 IP 激波检测或记录其他太阳风不连续性。类似的 IP 激波检测系统后来在 Intershock 和 ISEE 航天器上开发。

Recent space missions, Wind and ACE operating in the solar wind near the L1 point, are not equipped with IP shock detection systems because the telemetry rate allows to transmit a whole data set with sufficient time resolutions. A complex burst mode trigger is working onboard the STEREO spacecraft. It combines eight individual criteria from several instruments with different weighted components [Luhmann et al., 2008]. At the beginning of the spacecraft operation, the algorithm involved the changes of the following parameters: the magnetic field vector, electron, superthermal electron and proton density fluxes, and electric field fluctuation power in several frequency bands. Since the spacecraft launch, the trigger was continuously modified in order to optimize its criteria for maximum algorithm effectiveness; some of the components were disabled. The success rate of the algorithm changed from 30% in 2007 to 69% in 2011 [Jian et al., 2013]. However, this burst mode trigger is rather complicated and cannot be easily implemented into other missions.

Other IP shock “searching” algorithms based on different identification approaches, e.g., MHD approach, wavelet analysis, or generalized minimum variance analysis were proposed by Vandev et al. [1986] and Kartalev et al. [2002]. However, such algorithms are only suitable for analyses on the ground.

最近的任务（如 Wind 和 ACE）在靠近 L1 点的太阳风中运行，由于遥测速率允许传输具有足够时间分辨率的完整数据集，因此未配备 IP 激波检测系统。一个复杂的突发模式触发器正在 STEREO 航天器上运行，它结合了来自多个仪器的八个独立标准，并带有不同权重的组件。在航天器运行初期，该算法涉及以下参数的变化：磁场矢量、电子、超热电子和质子密度通量，以及多个频段的电场波动功率。自航天器发射以来，触发器不断被修改以优化其标准以实现算法的最大有效性；某些组件被禁用。该算法的成功率从 2007 年的 30% 提高到 2011 年的 69%。然而，这种突发模式触发器非常复杂，难以轻松应用于其他任务。

其他基于不同识别方法的 IP 激波“搜索”算法（例如 MHD 方法、小波分析或广义最小方差分析）由 Vandev et al. [1986] 和 Kartalev et al. [2002] 提出。然而，此类算法仅适用于地面分析。

Furthermore, there are several web applications with near real-time detectors that perform data analysis and select the possible shock candidates. IPS-SWS-ALERT (http://www.ips.gov.au) is an automated experimental product for the ACE data processing, and Shockspotter routines use data from the CELIAS/MTOF/PM sensor on the SOHO spacecraft (http://umtof.umd.edu/pm/). Vorotnikov et al. [2008, 2011] presented a fully automated code applied to the ACE data that selects upstream and downstream reference points, computes the shock normal, analyzes IP shocks using Rankine-Hugoniot jump conditions, and provides their solutions for real-time space weather applications. This shock-finder is able to find up to 40% of all manually identified shocks, and this rate increases to 67% with interactive solutions.

The development of such automated procedures for an identification of various structures could also contribute to a space weather warning system and space weather forecasting. It has been shown that fast forward IP shocks and the enhanced plasma densities downstream [Kennel et al., 1985] rapidly compress the magnetosphere after their impact [e.g., Echer et al., 2006; Tsurutani et al., 2011]. The structures associated with or driving the shocks can sometimes trigger intense geomagnetic storms, but the preconditioning of the magnetosphere is an important factor [Zhou and Tsurutani, 2001] for such process.

此外，还有一些具有近实时检测功能的网络应用程序，可执行数据分析并选择可能的激波候选者。IPS-SWS-ALERT (http://www.ips.gov.au ) 是一个用于处理 ACE 数据的自动化实验产品，而 Shockspotter 例程则使用 SOHO 航天器上的 CELIAS/MTOF/PM 传感器数据 (http://umtof.umd.edu/pm/)。Vorotnikov et al. [2008, 2011] 提出了一个完全自动化的代码，应用于 ACE 数据，选择上游和下游参考点，计算激波法向量，利用 Rankine-Hugoniot 跃变条件分析 IP 激波，并为实时空间天气应用提供解决方案。该激波查找器能够找到多达 40% 的所有手动识别的激波，而通过交互式解决方案，这一比例提高到 67%。

开发这样的自动化程序以识别各种结构，还可以为空间天气预警系统和空间天气预报做出贡献。研究表明，快速正向 IP 激波 和下游增强的等离子体密度在撞击后迅速压缩磁层。与激波相关或驱动激波的结构有时会引发强烈的地磁暴，但磁层的预条件是一个重要因素。

A strong association has been observed between ICMEs sheaths and IP shocks [Watari and Watanabe, 1998; Tsurutani et al., 1988; Lindsay et al., 1994], between IP shocks and magnetic clouds [Lepping et al., 2001; Luhmann, 1997], and between IP shocks and resulting geomagnetic disturbances [Gonzalez et al., 1999; Tsurutani et al., 1992; Tsurutani and Gonzalez, 1998].

One of important recent experimental results is that features of magnetic storms/substorms depend on the type of the interplanetary driver [Gonzalez et al., 1994; Tsurutani et al., 2006]. Therefore, the capability to forecast such events is critical to a successful prediction of space weather. IP shocks in front of potential storm drivers can be easily identified in interplanetary data by solar wind monitors at 1 AU, therefore, they serve as input data into numerical models that forecast the geomagnetic activity [Tóth et al., 2005; McKenna-Lawlor et al., 2006; 2012].

已经观察到 ICME 前鞘 与 IP 激波之间的强烈关联 ，IP 激波与磁云之间的关联 ，以及 IP 激波与由此产生的地磁扰动之间的关联。

一项重要的近期实验结果表明，磁暴/亚暴的特征取决于行星际驱动源的类型。因此，预测此类事件的能力对于成功预测空间天气至关重要。潜在风暴驱动源前方的 IP 激波可以通过 1 AU 处的太阳风监测器在行星际数据中轻松识别，因此它们作为空间天气预报模型的输入数据。

In the paper, we present a simple and flexible algorithm that indicates the possible IP shock arrival. This onboard algorithm is designed for application in the Radio and Plasma Waves (RPW) instrument [Maksimovic et al., 2007] for Solar Orbiter. It is based on interplanetary magnetic field and plasma measurements. The algorithm would allow us to register possible shocks in the solar wind from a beginning phase of the mission as well as at smaller distances from the Sun. Since a whole mission is long, the algorithm should reflect the variations of solar activity. The main task of the suggested algorithm is to identify all types of IP shocks (fast/slow and forward/reverse) regardless of their drivers, but it should exclude the events corresponding to abrupt large increases/decreases of plasma densities that are not associated with IP shocks [Zastenker et al., 2006].

The algorithm development is based on Wind measurements covering the half of the solar cycle 23, and the results are tested against IP shocks identified by other methods. We applied the suggested algorithm on a list of IP shocks observed by Wind that was compiled by J. C. Kasper. The list is available on web (CfA Interplanetary Shock Database), and we call it as “Kasper's list” throughout the paper. The test of the algorithm effectiveness in different distances from the Sun is based on the list of IP shocks observed by Helios between 0.29 and 1 AU [de Lucas et al., 2011].

本文提出了一种简单且灵活的算法，用于指示可能的 IP 激波到达。该机载算法专为太阳轨道器上的无线电与等离子体波 (RPW) 仪器设计。它基于行星际磁场和等离子体测量。该算法将使本文能够从任务的初始阶段以及更接近太阳的距离记录可能的激波。由于整个任务周期较长，算法应反映太阳活动的变化。建议算法的主要任务是识别所有类型的 IP 激波（快速/慢速和正向/反向），而不考虑其驱动源，但它应排除那些不与 IP 激波相关的等离子体密度急剧大幅增加/减少事件。

算法开发基于覆盖第 23 太阳周期一半的 Wind 测量数据，结果通过其他方法识别的 IP 激波进行测试。本文将提出的算法应用于 J. C. Kasper 编制的 Wind 观测 IP 激波列表。该列表可在网络上获取（CfA 行星际激波数据库），在本文中称其为“Kasper 列表”。算法在不同太阳距离下的有效性测试基于 Helios 在 0.29 至 1 AU 范围内观测的 IP 激波列表。

2 Shock Detection Algorithm

IP shocks are observed as abrupt changes of plasma parameters (solar wind speed, temperature, and density) and the magnetic field strength. The sense of such jumps (positive and negative) differs according to the IP shock type. The properties of the different types of shocks are referred e.g., in Kennel et al. [1985] and Tsurutani et al. [2011].

The first step of our automated identification of the IP shocks is based on detection of simultaneous jumps of the density, velocity, and magnetic field strength. Since we do not focus on a particular type of IP shocks, we use the magnitude of parameter jumps in the first step. The positive sense of the velocity jump that is obligatory across IP shocks will be incorporated into the second step of the algorithm. The sign of variations of the magnetic field and density could be used for the additional analysis of the shock type, but this is not a part of the present algorithm. Slow and reverse shocks [Ho et al., 1998; Lin et al., 2009] are rather rare in the solar wind (see e.g., the Kasper's list). It means that these shocks would not significantly increase the volume of the transmitted data. On the other hand, these shocks are understood in a much lesser extent than the fast forward shocks and the high-resolution data are desirable for their investigation. It should be noted that the details of shocks such as their normal angles relative to the upstream magnetic field and the Mach number are not included into the algorithm. This analysis will be done on the ground using more sophisticated fitting techniques, but the plasma density and magnetic field jump conditions will be given.

IP 激波 表现为等离子体参数（太阳风速度、温度和密度）以及磁场强度的突变。这些跳跃的方向（正向或负向）因 IP 激波的类型而异。不同类型激波的特性可参考 Kennel et al. [1985] 和 Tsurutani et al. [2011]。

本文自动识别 IP 激波的第一步基于检测密度、速度和磁场强度的同时跳跃。由于本文不专注于特定类型的 IP 激波，因此在第一步中使用参数跳跃的幅度。跨越 IP 激波时速度跳跃的正向变化（这一特征是强制性的）将被纳入算法的第二步。磁场和密度变化的符号可用于进一步分析激波类型，但这不是当前算法的一部分。慢速激波和反向激波在太阳风中较为罕见（例如，见“Kasper 列表”）。这意味着这些激波不会显著增加传输数据的体积。另一方面，这些激波比快速正向激波的研究程度低得多，因此高分辨率数据对其研究非常宝贵。需要指出的是，激波的细节（例如其相对于上游磁场的法向角度和马赫数）并未包含在算法中。这些分析将在地面通过更复杂的拟合技术完成，但等离子体密度和磁场跳跃条件会被提供。

The detection algorithm continuously evaluates a quality factor Q that is based on changes of moving averages of several solar wind parameters over a time interval corresponding to a typical IP shock scale. If we expect a sliding window with a width ΔT centered at the time t, the quality factor can be defined as:

where the magnetic field magnitude B, proton density n, and proton flow velocity V with 1 and 2 subscripts standing for mean values calculated on , t] and time intervals, respectively. The time interval of averaging can be considered as a free parameter of the detection algorithm. Coefficients α, β, and γ denote weights of the parameters ΔB, Δn, and ΔV. The proton temperature could be also included as an additional input parameter since it should exhibit a significant jump at the shock. However, the proton temperature is the most uncertain plasma parameter because a determination of the temperature requires careful processing of the full 3-D distribution. Simplified versions of onboard algorithms for computation of plasma moments do not provide reliable temperature data in highly disturbed plasma upstream and downstream of a shock.

检测算法持续评估一个质量因子 Q，该因子基于多个太阳风参数在一段时间内的移动平均值的变化，这段时间对应于典型的 IP 激波尺度。如果本文期望一个宽度为 ΔT 的滑动窗口以时间 t 为中心，则质量因子可以定义为：

其中，磁场强度 B、质子密度 n 和质子流速 V 的下标 1 和 2 分别表示在 , t] 和时间间隔内计算的平均值。平均化的时间间隔 可被视为检测算法的一个自由参数。系数 α、β 和 γ 表示参数 ΔB、Δn 和 ΔV 的权重。质子温度也可以作为附加输入参数，因为它在激波处应表现出显著的跳跃。然而，质子温度是最不确定的等离子体参数，因为确定温度需要对完整的三维分布进行仔细处理。机载简化版等离子体矩计算算法在激波上下游高度扰动的等离子体中无法提供可靠的温度数据。

An example of the quasi-perpendicular shock (i.e., the θBn angle between the shock normal and upstream magnetic field is >45°) detected on 19 July 2000 at 1530 UT by the Wind spacecraft together with the continuously computed quality parameter Q is presented in Figure 1.

Normalization coefficients α, β, and γ are equal to one third in this particular case. Each data point of Q, ΔB, Δn, and ΔV is calculated according to equations 1–5, and we use min throughout the paper. The quality factor Q shows a significant enhancement at the shock front (vertical red line) and such shock is recorded by the detection algorithm.

An example of the quasi-perpendicular shock observed by Wind on 19 July 2000. The panels represent the magnetic field magnitude B, the proton density n, proton flow velocity V, ΔB, Δn, ΔV, and quality factor Q with coefficients . The vertical red line denotes to the shock mark made by automated shock detection algorithm.

图 1 展示了一个准垂直激波（即激波法线与上游磁场之间的 θBn 角 >45°）的示例，该激波由 Wind 航天器于 2000 年 7 月 19 日 15:30 UT 检测到，并附有连续计算的质量参数 Q。在此特定情况下，归一化系数 α、β 和 γ 均为三分之一 。每个 Q、ΔB、Δn 和 ΔV 数据点根据公式 1–5 计算，本文在本文中始终使用固定的时间间隔 。质量因子 Q 在激波前沿（红色竖线）显示出显著增强，这样的激波被检测算法记录下来。