【作者】 何兆海；

【导师】 刘振兴； 沈超；

【作者基本信息】 中国科学院研究生院（空间科学与应用研究中心） ， 空间物理学， 2008， 博士

【摘要】 等离子体注入现象是磁暴或磁层亚暴期间不同能量段粒子的通量突然急剧增加,几乎所有的磁层亚暴事件都伴随着等离子体注入过程。粒子注入是磁层亚暴的基本特征之一,也是确定磁层亚暴触发时间的标志之一,研究等离子体注入对深入了解磁层亚暴有着重要的意义。论文主要开展了以下三方面的工作:1.比较研究单一磁偶极子场作用下,以及Vollland-Stern电场和磁偶极子场共同作用下,粒子沿方位角方向和地球径向漂移的速度特征。通常粒子沿方位角方向的漂移运动主要是由磁场曲率和梯度漂移引起的,但是对于能量低于100keV的粒子,电场所引起的方位角方向的漂移速度不可忽视,通过计算发现能量越低的粒子,粒子的方位角方向漂移速度受电场的影响越大。粒子沿地球径向的漂移速度主要是电场作用的结果。从等离子体径向注入角度出发,利用粒子在磁偶极子场和Volland-Stern电场模型中的漂移模式结合单颗卫星观测到的色散注入现象,发展了一种返推等离子体无色散注入区径向距离和注入时间的方法。根据1998年6月14日单颗同轨道卫星LANL-97A观测到的等离子体色散注入现象,返推表明无色散注入边界的径向距离为7.1R_E,无色散注入的时间为04:17:20,返推得到的无色散注入时间与Polar卫星极紫外成像仪观测到的磁层亚暴触发时间相当吻合。利用漂移模式结合两颗卫星观测到无色散注入现象研究等离子体注入源区的径向位置。1998年3月11日1991-080和LANL-97A两颗卫星观测到的等离子体无色散注入事件表明注入前50～75keV能量段的粒子集中分布在磁尾8.1R_E,能量段75～105keV的粒子分布在磁尾7.9R_E附近,粒子在电磁场的作用下沿内磁层运动都需要约24分钟。2.研究磁尾(-18R_E<X<-10R_E)等离子体注入期间,质子、热离子的物理参数(数密度、温度和速度)和磁尾对流电场的变化特征。观测结果发现磁尾等离子体注入事件类似于同步轨道区注入事件,可以分成五类:(a)只有离子注入;(b)离子先于电子注入;(c)离子和电子同时注入;(d)电子先于离子注入;(e)只有电子注入。注入期间质子和热离子的数密度和温度显著增加。注入前质子持续沿尾向运动,速率维持在50km/s以内,注入后质子沿地向运动,平均运动速度约200km/s,注入后速度恢复到注入前的水平大概需要20分钟左右。注入前晨昏对流电场主要表现为在0～1mV/m之间作震荡变化,注入后电场有两种变化:(a)电场突然增大,其值为正,电场强度达到3.0mV/m;(b)电场突然反向,其值为负,电场强度降至-1.5mV/m。磁尾等离子体注入可能存在三种不同的物理机制:(a)电场为正时,晨昏对流电场所引起的电漂移可以解释磁尾粒子的注入过程;(b)电场反向时,这类事件的注入机制可能是粒子沿磁力线沉降导致大量粒子堆积;(c)注入期间β值的计算表明卫星穿越等离子体片边界层进入等离子体片,使得粒子的通量同时增加。3.考虑到对流电场与等离子体注入的紧密关系,本文利用赤道轨道卫星TC-1的数据来研究不同地磁活动条件下大尺度电场的特征。结果表明:不同地磁活动强度下电场的Ex分量随着壳指数L值的增大而单调递减,从0.7mV/m降至～0mV/m;K_p<4时电场分量Ex距离地球L>9R_E基本维持在～0mV/m附近,K_p>4时电场分量Ex在磁尾11R_E处近乎为零。晨昏向电场E_y在磁地方时18:00<MLT<06:00范围内始终指向昏侧。L<6～7时,电场E_y逐步增加;L> 6～7时,电场E_y开始减小;然而距离地球9～12R_E附近,电场变化不大,基本维持在0.4mV/m。比较观测的结果与Volland-Stern模型电场的计算结果发现:电场Ex分量主要是受共转电场的控制;晨昏向电场分量E_y与模型的结果有很大区别。模型电场随壳指数单调递增,而实际观测发现距离地球一定位置晨昏电场开始减小,这个位置大约距离地球6～7R_E,本文对Volland-Stern电场模型做了初步修改,讨论了修改后电场参数的物理意义。更多还原

【Abstract】 The plasma dispersionless injection is sudden increases of particle ?uxes invarious energy channels during the magnetospheric storm and substorm. Almostall of substorms are accompanied by the plasma injections, which are the mostcommon features of substorms and, more specifically, among the most reliableindicators of substorm onsets. Plasma injections play an important role in un-derstanding the substorm process.The following three aspects are studied in the thesis.1. Two components of drift velocity have been studied in dipole magneticfield, comparing with the results calculated in a dipole magnetic field and theVolland-Stern electric field. Generally, the azimuth component of velocity arecaused by the magnetic gradient and curvature e?ect. If the particle’s energy islower than 100keV, the velocity along the azimuth direction caused by the electricfield can not be neglected. The lower energy the particle has, the more a?ectioncaused by electric field. The radial component of velocity mainly depends on theelectric field.Based on the radial injection, a method has been developed which allowsremote sensing of the plasma injection time and the radial distance of injectionboundaries by using measured energy dispersion and modeling particle driftswithin the Volland-Stern electric field and dipole magnetic field model. The radialdistance of the injection boundary deduced from a dispersion event observed byLANL-97A satellite on June. 14, 1998 is 7.1R_E,and the injection time agreeswell with the substorm onset time identified by the Polar Ultraviolet Imager.The method has been applied to an event happened at 22.9 UT on March. 11,1998, when both satellites (1991-080 and LANL-97A) observed the dispersionlesscharacter. The results indicate that the radial distance of injection source for50～75keV particles locate at 8.1R_E and 75～105keV particles locate at 7.9R_E atmagnetotail, they move earthward from magnetotail into inner magnetosphere at 22.5UT.2. The properties of proton (0eV<E<40keV) and hot ion (5eV<E<32keV)and the convection electric field in the plasma sheet are examined during themagnetotail (-18R_E<X<-10R_E) plasma injection. The results show that fiveclasses of magnetotail injection events are found to be similar with the geosyn-chronous observation: (a) pure ion injections; (b) ion injections followed by anelectron injection; (c) simultaneous ion and electron injections; (d) electron in-jections followed by an ion injection; (e) pure electron injections. Proton and hotion show a significant increase in temperature and density at the onset. Beforeplasma injection, the proton moves tailward with a velocity about 50km/s; afterinjection, proton injects earthward with an average velocity more than 200km/s.It takes 20 minutes for the velocity to recovery to the pre-injection one. Con-vection electric field ?uctuates between 0 and 1 mv/m before injection. Afterinjection, two di?erent electric field configurations are found: (a) electric fieldincreases suddenly at the onset and the value is positive and the magnitude isabout 3.0mV/m; (b) electric field changes the direction at the onset, and turnsinto a negative value with the magnitude around -1.5mV/m.Three possible mechanisms for magnetotail plasma injection are discussed:(a)electric drift caused by dawn-dusk convection electric field is one of the mecha-nisms of the particles injected earthward in magnetotail; (b) particles precipitatealong the magnetic field line during the reversal convection electric field; (c) Cal-culatingβvalue shows that satellite moves from plasmapause into plasma sheetalso can produce the ?ux enhancement.3. Considering the relationship between convection electric field and plasmainjection, the large scale magnetospheric electric field changes with the geomag-netism activity is discussed by using TC-1 satellite data. The X component ofelectric field decreases monotonically from 0.7 to～0mV/m with the increasingshell index L. For K_p<4, the curls of Ex electric field component maintain about～0mV/m at a distance beyond 9R_E. As to K_p>4, Ex turns into zero at 11R_E.The dawn-dusk component of electric field is almost always duskward over themagnetic local time range from 18:00 to 06:00. E_y enhances for L< 6～7 and be- comes week gradually for L> 6～7. The convection electric field ?uctuates around0.4mV/m between L=9 and L=12.Comparing the observations and the results calculated from Volland-Sternelectric field model, Ex component dominates by the corotation electric fieldand the observation convection component does not agree well with the modelresults. The most important is that E_y increases monotonically with the shellvalue L, while the observations has been found that E_y should decrease at acertain distance, which locates at 6～7R_E identified in this work. According tothe di?erences between the observation and model test, we modify the Volland-Stern electric field, and discuss the physics means for the parameters in themodified expression.更多还原