基于机器学习方法对23太阳活动周质子事件的研究

在耀斑伴随日冕物质抛射(CME)事件编目数据的基础上,进行太阳质子事件(SPE)匹配,构建研究数据集.利用Apriori算法挖掘SPE与耀斑级别、耀斑发生日面位置以及CME角宽度和速度的关联关系.结果 表明:X级耀斑、全晕CME、高速(＞1000 km.s-1) CME和日面西半球耀斑是最可能伴随质子事件的4种特征,其...     

全晕CME与太阳质子事件的关系

日冕物质抛射（CME）是太阳质子事件的重要源头.CME的速度和源区位置是太阳质子事件产生的重要因素.通过统计最近5年全晕CME与太阳质子事件的关系发现,速度大且源区位置距离日面上连接地球磁力线足点近的全晕CME更易引发太阳质子事件,其中速度大于1200km…-1、角距离60°以内的样本引发太阳质子事件的概率最高.对3个未引发太阳质子事件的高速全晕CME进行了详细分析,发现CME的主体爆发方向和行星际磁场环境的变化也影响太阳质子事件的产生.因此,在太阳质子事件的实际预报中,综合CME爆发速度、源区位置、主体抛射方向和行星际环境等多个因素才能给出更准确的事件预报结果.      

1989年3月大宇宙线暴事件的特征

本文分析了1989年3月一系列大耀斑等离子体抛射引起的宇宙线强度变化的特征.除对中子成分分析外,还对μ介子成分及其各向异性特征作了分析,讨论了宇宙线强度变化与太阳耀斑特性和地磁扰动之间的关系。分析发现,宇宙线的Forbush下降不仅与太阳耀斑的级别、持续时间,以及在日面上的位置有关,而且还与光学耀斑是否伴有强的X射线暴、是否有强的射电爆发,以及是否引起强的地磁暴紧密有关.各向异性分析表明,3月大事件的各向异性明显小于宁静时的各向异性,这可能是因为受到太阳活动强烈调制之后,宇宙线各向异性趋于减小的原因.     

太阳质子事件与太阳耀斑的关系

通过对0°W-39°W,40°W-70°W,71°W-90°W经度范围内太阳质子事件与太阳耀斑的相关性计算分析,发现太阳质子事件与太阳耀斑的相关系数依赖于经度.太阳耀斑积分与地球磁链接区域（40°W-70°W）太阳质子事件强度的相关系数最大.相关系数的这种特点与耀斑加速粒子的最大流量只出现在磁链接区域的特征相吻合.计算结果表明,太阳耀斑对太阳质子事件具有贡献,即耀斑对E ≥ 10MeV的质子加速有贡献.耀斑和CME在磁链接区域对太阳质子事件的贡献相同,这说明太阳质子事件是混合型事件.      

X级以上耀斑与地磁效应关系研究

统计研究了2010年1月至2012年12月期间所有与耀斑爆发相伴生的日冕物质抛射(CME) 引发的地磁暴事件. 结果表明, 对于CME源区其主要分布在日面 45°E-45°W, 占总数的78.95%, 且西半球比东半球多, 即源区位于西半球的CME易产生地磁效应; X级耀斑与地磁效应的关联性更高, 60.0%的 X级耀斑在其爆发后的2～3天内观测到地磁暴, 而其他级别的耀斑与地磁效应的关联性低得多, 均不足10%; 通过对此期间日面爆发的所有X级耀斑研究分析后发现, 对于源区位于日面东经45°E-45°W 的X级耀斑, 若在其爆发过程中没有大尺度日面扰动, 则无伴生CME且后续产生地磁效应的可能性很低. 由此提出一种通过分析日面观测数据进行地磁暴预报的方法.      

米波Ⅲ型爆发与日冕物质抛射

对澳大利亚Culgoora天文台射电频谱仪在太阳活动第23周峰年期间记录到的米波Ⅲ型爆发(20～420 MHz),与日冕物质抛射(CME)、Hα耀斑及相关事件进行了统计分析,发现米波Ⅲ型爆发与CME的关系没有Ⅱ、Ⅳ型爆发与CME的关系密切;米波Ⅲ型爆发发生的时间在CME之前25～30 min最多;72%的CME事件伴随长寿命的Hα耀斑.从这些观测特征出发,对米波Ⅲ型爆发、CME和Hα耀斑进行了定性的解释.      

太阳高能粒子事件强度与相关双CME事件关系的两个事例

太阳高能粒子事件常伴随太阳耀斑和日冕物质抛射事件(Coronal Mass Ejections,CME)出现,由于太阳高能粒子事件的关键因素是双CME的相互作用,利用SOHO卫星观测的高能粒子强度、耀斑强度以及CME的相对高度与时间,通过高度与时间拟合得到的速度,分析了2001年4月15日和2005年1月20日的太阳高能粒子事件强度与相关双CME事件的关系,发现这两个太阳高能粒子事件中E ≥ 10MeV质子的强度与双CME事件无关.因此在这两次太阳高能粒子事件早期,E ≥ 10MeV质子的强度只与相关太阳耀斑和CME有关.      

第21～24太阳周4类空间天气事件爆发特征统计分析

对第21～24太阳周不同等级的太阳X射线耀斑事件、太阳质子事件、地磁暴事件及高能电子增强事件的爆发频次特征进行统计,结果表明:太阳周耀斑爆发的总数量与该太阳周的黑子数峰值呈正比,耀斑总数、X级耀斑事件数与峰值的相关系数分别为0.974,0.997;太阳质子事件主要发生在峰年前后1～2年,约占总发生次数的80%,峰值通量大于10pfu （1 pfu=1 cm-2·sr-1·s-1）的质子事件中,84%伴有耀斑爆发,并且主要伴随M或X级耀斑,少量伴随C级耀斑,峰值通量大于1000pfu的质子事件中,98%伴随M或X级耀斑,并且以X级耀斑为主;第21,22,23和24太阳周发生地磁暴最频繁的时间分别在1982,1991,2003年和2015年,分别滞后黑子数峰值时间3年、2年、2年和1年;72%的高能电子增强事件发生在太阳周下降期,24%的高能电子增强事件发生在太阳周上升期.      

用行星引潮力预报太阳耀斑的注记

太阳活动与空间坏境紧密相关,大耀斑会引起空间环境的剧烈扰动．太阳活动预报便成空间环境预报的基本依据．太阳预报水平长期以来提高缓慢,太阳物理学家皆有共识,寄希望于物理预报的进展,但举步维艰．近来,“太阳活动的行星潮汐效应”的研究取得了新进展1)[1,2],引潮力可以触发耀斑,从而,利用这类效应发展物理预报技术,呈现良好前景．“太阳耀斑发生率按行星引潮力的分布”已有几个具体结果,表面看来,其间似乎有出人需予澄清．1972年,董土仑和林柏森发现1958－1968年94个质子耀斑的发生率在其目面经度处（活动经度上）技引潮力…     

神经网络方法在太阳质子事件短期预报中的应用

在大量统计结果的基础上,深入研究了太阳质子事件预报机理．总结了质子事件爆发与太阳活动区面积、位置、McIntosh结构、磁结构以及前两天活动区爆发耀斑事件数目之间的关系．然后,在神经网络的基础上建立了太阳质子事件短期预报模型,并对2000年以后12个未参加训练的样本进行测试,结果对事件预报的准确率为83％．此外,我们还利用该模型对2002年1-4月发生的几次质子事件进行了预报试验,结果发现,这期间发生的6次事件都被预报．其中3次质子事件系统预报提前了3天,两次事件预报提前了2天,一次事件提前1天预报．     

A statistical study of post-flare-associated CME events

We present a statistical study of post-flare-associated CMEs (PFA-CMEs) during the period from 1996 to 2010. By investigating all CMEs and X-ray flares, respectively, in the LASCO and GOES archives, we found 15875 CMEs of which masses are well measured and 25112 X-ray flares of which positions are determined from their optical counterparts. Under certain temporal and spatial criteria of these CMEs and solar flare events, 291PFA-CMEs events have been selected. Linking the flare fluxes with CME speeds of these paired events, we found that there is a reasonable positive linear relation between the CME linear speed and associated flare flux. The results show also the CME width increases as the flux of its associated solar flare increases. Besides we found that there is a fine positive linear relation between the CME mass and its width. Matching the flare fluxes with CME masses of these paired events, we find the CME mass increases as the flux of its associated solar flare increases. Finally we find the PFA-CME events are in regular more decelerated than the other CMEs.     

The impact of large solar events on the total electron content of the ionosphere at mid latitudes

Ionospheric disturbances associated with solar activity may occur via two basic mechanisms. The first is related to the direct impact on the ionosphere of EUV photons from a flare, and the second by prompt electric field penetration into the magnetosphere during geomagnetic storms. In this paper we examine the possibility that these two mechanisms may have an impact at mid latitudes by calculating the total electron content (TEC) from GPS stations in Mexico during several large X-ray flares. We have found that indeed large, complex flares, which are well located, may affect the mid latitude ionosphere. In fact, in the solar events of July 14, 2000 and April 2001 storms, ionospheric disturbances were observed to increase up to 138 and 150 TECu, respectively, due to the influence of EUV photons. Also, during the solar events of July 2000, April 2001, Halloween 2003, January 2005 and December 2006, there are large ionospheric disturbances (up to 393 TECu in the Halloween Storms), due to prompt penetration electric field, associated with CME producing geomagnetic storm.     

A statistical study of CME-associated flare during the solar cycle 24

We investigate on the relationship between flares and coronal mass ejections (CMEs) in which a flare started before and after the CME events which differ in their physical properties, indicating potentially different initiation mechanisms. The physical properties of two types flare-correlated CME remain an interesting and important question in space weather. We study the relationship between flares and CMEs using a different approach requiring both temporal and spatial constraints during the period from December 1, 2008 to April 30, 2017 in which the CMEs data were acquired by SOHO/LASCO (Solar and Heliospheric Observatory/Large Angle Spectrometric Coronagraph) over the solar cycle 24. The soft X-ray flare flux data, such as flare class, location, onset time and integrated flux, are collected from Geostationary Environmental satellite (GOES) and XRT Flare catalogs. We selected 307 CMEs-flares pairs applying simultaneously temporal and spatial constraints in all events for the distinguish between two associated CME-flare types. We study the correlated properties of coincident flares and CMEs during this period, specifically separating the sample into two types: flares that precede a CME and flares that follow a CME. We found an opposite correlation relationship between the acceleration and velocity of CMEs in the After- and Before-CMEs events. We found a log-log relation between the width and mass of CMEs in the two associated types. The CMEs and flares properties show that there were significant differences in all physical parameters such as (mass, angular width, kinetic energy, speed and acceleration) between two flare-associated CME types.     

Properties and relationship between solar eruptive flares and Coronal Mass Ejections during rising phase of Solar Cycles 23 and 24

Statistical relationship between major flares and the associated CMEs during rising phases of Solar Cycles 23 and 24 are studied. Totally more than 6000 and 10,000 CMEs were observed by SOHO/LASCO (Solar and Heliospheric Observatory/Large Angle Spectrometric Coronagraph) during 23rd [May 1996–June 2002] and 24th [December 2008–December 2014] solar cycles, respectively. In particular, we studied the relationship between properties of flares and CMEs using the limb events (longitude 70–85°) to avoid projection effects of CMEs and partial occultation of flares that occurred near 90°. After selecting a sample of limb flares, we used certain spatial and temporal constraints to find the flare-CME pairs. Using these constraints, we compiled 129 events in Solar Cycle 23 and 92 events in Solar Cycle 24. We compared the flare-CME relationship in the two solar cycles and no significant differences are found between the two cycles. We only found out that the CME mean width was slightly larger and the CME mean acceleration was slightly higher in cycle 24, and that there was somewhat a better relation between flare flux and CME deceleration in cycle 24 than in cycle 23.     

CME–flare association during the 23rd solar cycle

The relation between coronal mass ejections (CMEs) and solar flares are statistically studied. More than 10,000 CME events observed by SOHO/LASCO during the period 1996–2005 have been analyzed. The soft X-ray flux measurements provided by the Geostationary Operational Environmental Satellite (GOES), recorded more than 20,000 flares in the same time period. The data is filtered under certain temporal and spatial conditions to select the CME–flare associated events. The results show that CME–flare associated events are triggered with a lift-off time within the range 0.4–1.0 h. We list a set of 41 CME–flare associated events satisfying the temporal and spatial conditions. The listed events show a good correlation between the CME energy and the X-ray flux of the CME–flare associated events with correlation coefficient of 0.76.     

A study on radio-loud interacting/non-interacting CMEs-associated SEPs and solar flares

We have established a data set of 58 major hybrid SEP events associated with meter-to-decahectometer wavelength (m-to-DH) type II bursts, solar flares, and radio-load CMEs during the period of 1997–2014. The main focus of our study is to address the following two questions: Does the interaction of CMEs play a role in the enhancement of SEP intensity? Is there any difference in the seed population, and parent eruptions in the SEP events with and without CME interactions? Hence, the sample of 58 events is classified into two sets: (i) 35 non-interacting-CME-associated SEP events; (ii) 23 interacting-CME-associated SEP events. All the characteristics of SEPs, their associated CMEs/flares and the relationships between them are statistically analyzed and compared. Some of the basic attributes and relative elemental abundances (Fe/O ratios) of the both the sets are also compared. The results indicate that the seed particles in non-interacting-CME-associated SEP events are mostly from solar wind/coronal materials. But in the case of interacting-CME-associated SEP events, it may be associated with both flare material from preceding flares and coronal materials from solar wind/preceding CMEs. The correlation studies reveal that there are clear correlations between logarithmic peak intensity of SEP events and properties of CMEs (space speed: cc?=?0.56) and solar flares (peak intensity: cc?=?0.40; integrated flux: cc?=?0.52) for non-interacting-CME-associated SEP events. But these correlations are absent for the interacting-CME-associated events. In addition, the results suggest that interaction of primary CMEs with their preceding CMEs plays an important role in the enhancement of peak intensity of SEPs at least for a set of m-to-DH type II bursts associated SEP events.     

X级质子耀斑的特征研究

为了更加准确地判断X级耀斑是否引发质子事件,对X级质子耀斑和非质子耀斑的耀斑积分通量、源区、CME速度、CME角宽度、背景太阳风速度及背景X射线通量的分布进行了统计研究.发现非质子耀斑和质子耀斑的积分通量、经度、CME速度和CME角宽度具有明显不同的分布.非质子耀斑大多集中在东部,耀斑积分通量小于0.3J·m-2,CME速度小于1300km·s-1的区域内;质子耀斑大多集中在中部或西部,耀斑积分通量大于0.3J·m-2,CME速度大于1300km·s-1的区域内.质子耀斑伴随的CME角宽度主要集中在360°,非质子耀斑的CME角宽度分布则相对分散.两类耀斑的背景太阳风速度和背景X射线通量分布差别不大.利用两类耀斑各个参量分布上的差异,有望提高X级耀斑预报的准确率.      

基于PCA方法对太阳活动区主要参量的分析

太阳耀斑与太阳质子事件的发生通常与太阳活动区存在非常密切的关系, 对这种关系的深入分析有助于太阳耀斑和太阳质子事件预报模型的建立. 本文利用主成分分析(Principal Component Analysis, PCA)方法对1997-2010年太阳质子事件所在活动区的主要参量进行分析, 选取的参量包括黑子磁分类、 McIntosh分类、太阳黑子群面积、10.7 cm射电流量、耀斑指数、质子耀斑位置和软X射线耀斑强度. 结果得到81个太阳活动主成分得分值排序(得分值代表每个事件的强弱), 与太阳质子事件峰值流量、太阳黑子年均值以及10.7 cm射电流量年均值的对比显示相似度非常高, 表明主成分得分值一定程度上可以反映太阳活动的强弱规律.      

CME productivity associated with Solar Flare peak X-ray emission flux

It is often noticed that the occurrence rate of Coronal Mass Ejections (CMEs) increases with increase in flare duration where peak flux too increase. However, there is no complete association between the duration and peak flux. Distinct characteristics have been reported for active regions (ARs) where flares and CMEs occur in contrast to ARs where flares alone occur. It is observed that peak flux of flares is higher when associated with CMEs compared to peak flux of flares with which CMEs are not associated. In other words, it is likely that flare duration and peak flux are independently affected by distinct active region dynamics. Hence, we examine the relative ability of flare duration and peak flux in enhancing the CME productivity. We report that CME productivity is distinctly higher in association with the enhancement of flare peak flux in comparison to corresponding enhancement of flare duration.