基于广义回归神经网络的行星际/太阳风参数和地磁指数的紫外极光强度建模

极光卵极光强度的空间分布是太阳风-磁层-电离层能量耦合过程的重要表现,并且随着空间环境参数和地磁指数的变化而变化,是空间天气的重要指示器.建立合适的极光强度模型对亚暴的预测以及磁层动力学的研究具有重要意义.本文基于Polar卫星的紫外极光成像仪(Ultraviolet Imager,UVI)数据,采用两种不同的极光强度表征方法,即曲线拟合方法(从UVI图像数据中提取极光强度沿磁余纬方向上的曲线特征,Curve Feature along the Magnetic Co-latitude Direction of the Auroral Intensity,CFMCD_AI)和网格化方法(从UVI图像数据中提取极光强度的网格化特征,Gridding Feature of the Auroral Intensity,GF_AI),来构造极区极光强度特征数据库.然后,利用该数据库,采用广义回归神经网络(Generalized Regression Neural Network,GRNN)构建了以行星际/太阳风参数(行星际磁场三分量、太阳风速度和密度)和地磁指数(AE指数)为输入参数的两种极光强度预测模型(GRNN_CFMCD_AI模型和GRNN_GF_AI模型).利用图像质量评价指数结构相似度(structure similarity,SSIM)作为极光强度模型预测结果和对应的UVI图像的相似性评价标准(完全相似为1,不相似为0,一般认为SSIM大于0.5是具有较好的相似性),对两种极光强度模型进行了性能评价.结果显示,GRNN_GF_AI模型预测结果对应的SSIM值范围为0.36～0.77,均值为0.54,性能优于GRNN_CFMCD_AI模型的.     

极光电集流纬度变化特征初步分析

为研究极光电集流地磁纬度分布特征,利用北半球SME台站提供的极光电集流指数,通过时序叠加的方法,分析98-07年极光电集流中心地磁纬度随季节和世界时的分布特征.通过对磁扰程度的分级,分析极光电集流地磁纬度随地磁扰动程度的变化特征.结果表明:1)由于SME台站覆盖范围更广,更多地记录到最大的极光电集流强度,有利于研究极光电集流的变化特征;2)西向板光电集流纬度分布存在与强度相反的季节性变化特征,在春秋出现最低值,冬季、夏季出现最高值;3)在| SML|＜2000 nT时,西向极光电集流地磁纬度随着极光电集流强度的增强,近似以线性关系向低纬迁移.随后伴随SML的进一步增强,西向极光极光电集流中心地磁纬度仍有向低纬迁移的趋势,但主要是在磁纬62°N-63°N之间波动.     

极光亚暴期间的南极中山站地磁共轭点位置研究

通过对北极斯瓦尔巴特( Svalbard )岛Longyearbyen台站的 极光扫描光度计和地磁 观测数据在地磁亚暴膨胀相期间的对比分析,发现扫描光度计记录中的极光边缘的快速极向 运动和地磁数据x分量的陡峭负弯之间有着良好的对应关系,地磁数据可用来研究两极 高纬 地区极光亚暴的地磁共轭特征. 对南极中山站、挪威Troms Svalbard台链和东格陵兰岛 地 区共15个地磁台站在7个典型极光亚暴事件中的地磁数据进行对比分析后发现, 中山站的地 磁共轭点位置存在明显的漂移特征,漂移的范围在斯瓦尔巴特岛与东格陵兰岛之间,纬度值 与CGM模型值近似.     

地基观测的夜侧极光对行星际激波的响应

行星际激波与地球磁层相互作用通常会导致日侧极光活动增强,随后沿着极光卵的晨昏两侧向夜侧扩展的激波极光.行星际激波也可能直接导致夜侧扇区极光活动增强,甚至沉降粒子能通量的数量级可以与典型亚暴相比拟.本文首次利用我国南极中山站和北极黄河站连续多年积累的极光观测数据,对行星际激波与地球磁层相互作用期间地面台站在夜侧扇区(18—06MLT)观测的极光响应进行了分析.对18个极光观测事件的分析结果表明:行星际激波与磁层相互作用可以在夜侧触发极光爆发和极光微弱增强或静态无变化事件;太阳风-磁层能量耦合的效率以及磁层空间的稳定性决定着行星际激波能否触发极光爆发.     

太阳风参数对极光电集流中心纬度变化的影响

为了认识并了解磁层能量输入对极光电集流中心纬度变化的影响,本文使用1998-2006年的SuperMAG、OMNI以及EPI数据分别分析了西向和东向电集流中心纬度与行星际磁场Bz分量、太阳风速度、以及极光沉降粒子估算能量的变化特征.研究结果表明:(1)当Bz0时,磁场强度越大,西向和东西电集流中心纬度越低,西向电集流朝晨侧移动,东向电集流朝正午方向移动;(2)在Bz0时,电集流中心主要受太阳风速度控制,当太阳风速度较大时,电集流中心向低纬移动;(3)电集流中心纬度随其强度的变化特征与随EPI的变化特征一致.     

电离层对流和极光区电集流的地磁链观测

本文采用31个高纬地磁台站资料考察1997年5月15日一次中等磁暴期间极光区电集流和电离层对流的空间分布和时间变化;其中20站处于纬度60°N～80°N之间的西半球,而另11站是偶极磁经度约为120°E的欧洲IMAGE地磁站链.对此纬度链和经度链上各站1 min精度地磁资料的综合分析结果表明,极光区电集流中心的相对强度及其纬度位置是随世界时和地方时区不断变化的.电集流中心所处位置的变化可能是其中心的南北移动造成的,也可能是中心带与磁纬圈间的相互倾斜所致.另一方面,电离层对流形态和晨昏对流圈的经向跨度及其两端的位置是基本不变的.有关结论得到同期的非相干散射雷达EISCAT观测的证实和补充.     

对流电场、场向电流和极光区电集流变化的地磁响应

对流电场、场向电流和极光区电集流是磁层一电离层耦合的主要物理过程．它们的演化发展时间分别为几分钟至半小时的量级．本文用１００°Ｅ和３００°Ｅ的两个地磁经度链附近各１１个台站的１ｍｉｎ均值地磁Ｈ和Ｚ分量资料,分析了１９９４年４月１６-１７日磁暴期间磁层耦合过程对极光区和中低纬区电离层扰动的地磁特征．强磁暴开始时,台站所处的地方时位置不同,则观测到的电离层和地磁响应也完全不同．这是磁层对流和一、二区场向电流共同作用的结果．一般说,扰时极光区的西向电集流变化更为强烈．随着耦合的发展,极光区范围会向南北扩展,电集流中心带则向低纬侧移动．在中低纬区,二区场向电流的建立能屏蔽一区场向电流所产生的扰动,并引起反向的电流及地磁变化．由此,中低纬区夜间有可能出现短时间的东向电场,又可通过ＥＸＢ的垂直向上漂移作用抬升Ｆ层等离子体,并发生同一经度链附近的多站电离层ｈ＇Ｆ同时突增现象．另一方面,磁赤道附近的台站则更多地受内磁层赤道环电流和电离层赤道电集流的影响．     

亚暴期间极光电集流带的变化

极光活动加剧和极光电集流增强是磁层-电离层能量耦合的两种重要表现形式,它们同为磁层带电粒子向电离层沉降的结果,但是变化规律却非常不同.本文用地基磁场资料,反演极区等效电流体系,研究地磁平静期和扰动期极光电集流带的运动特点.研究表明,Harang间断把极光电集流带分为两段：下午—黄昏段的东向电集流带较弱,而晨侧和子夜—凌晨段的西向电集流带较强.在亚暴膨胀相,随着AE指数增大,整个极光卵向赤道扩展,而极光电集流带却表现出分段差异的特点：下午—黄昏东向电集流带向低纬移动,晨侧西向电集流带也向赤道移动,而子夜—凌晨西向电集流带则向极移动.电动力学分析表明,在不同地方时段,控制电流的主要因素不同,因而,电流及其磁扰有不同的特点：下午—黄昏东向电集流和晨侧西向电集流组成了DP2电流体系,主要受控于磁层对流电场,反映了“驱动过程”的行为;而子夜—凌晨西向电集流是DP1电流体系的基本部分,主要受控于电导率,反映了“卸载过程”的特点.     

南极中山站电离层的极区特征

本文利用1996年的电离层数字测高仪DPS-4所测的f0F2、f0E以及美国NOAA和DMSP卫星观测估算的半球功率指数和午夜极光区赤道侧边界纬度等资料,考察中山站电离层的极区特征。结果表明,在太阳和地磁宁静环境下,冬季极夜磁正午中山站处于极隙区中心时,电离层内的电离密度达全天的最大值;上、下午各有数小时间隔位极光带内时,高能粒子的电离作用也很重要;夜间进入极差区后,电子密度则很低。夏季极昼时,太阳EUV辐射的电离效应使电离层电离密度比冬季值大许多,而且,日变化的最大值时间也提前了1~2h,强磁扰时,极隙区和极光带均向低纬侧移动;中山站上空的电子密度会大幅度下降。在中等扰动环境下情况要加复杂：磁正午前后极隙区内软粒子沉降的电离强度有所减小,而上、下午极光区的高能粒子电离则有较大增加。     

时变行星际太阳风模拟及其结果评估

背景太阳风对于地球附近的空间环境有着重要的影响,三维磁流体力学太阳风模型是背景太阳风研究和预报的重要工具.通过太阳光球磁场数据驱动的边界条件,我们发展了一个时变的行星际三维磁流体力学太阳风模型.使用这个模型,我们模拟了2008年全年的行星际背景太阳风,分析了该年太阳风结构全球特征的演化和行星际局地观测与日冕结构间的联系.实现了一套太阳风连续参数和特征结构模拟质量的定量评估方法.对2008年模拟结果的评估表明,模型较好地重现了背景太阳风的大尺度特征.模拟与观测速度间的相关性系数达到了0.6以上,行星际磁场强度与观测吻合得较好,捕获了全部的行星际磁场极性反转和82.76%的流相互作用区,行星际磁场极性反转的误报率仅为6.67%,流相互作用区的误报率仅为11.11%,两种结构的到达时间误差在1天左右.同时,通过综合分析评估结果,我们明确了高速流结构、内边界磁场分布等模型在进一步改进中需要重点注意的问题.     

On auroral dynamics observed by HF radar: 1. Equatorward edge of the afternoon-evening diffuse luminosity belt

Observations and modelling are presented which illustrate the ability of the Finland CUTLASS HF radar to monitor the afternoon-evening equatorward auroral boundary during weak geomagnetic activity. The subsequent substorm growth phase development was also observed in the late evening sector as a natural continuation of the preceding auroral oval dynamics. Over an 8 h period the CUTLASS Finland radar observed a narrow (in range) and persistent region of auroral F- and (later) E-layer echoes which gradually moved equatorward, consistent with the auroral oval diurnal rotation. This echo region corresponds to the subvisual equatorward edge of the diffuse luminosity belt (SEEL) and the ionospheric footprint of the inner boundary of the electron plasma sheet. The capability of the Finland CUTLASS radar to monitor the E-layer SEEL-echoes is a consequence of the nearly zero E-layer rectilinear aspect angles in a region 5/10° poleward of the radar site. The F-layer echoes are probably the boundary blob echoes. The UHF EISCAT radar was in operation and observed a similar subvisual auroral arc and an F-layer electron density enhancement when it appeared in its antenna beam.     

Global geomagnetic and auroral response to variations in the solar wind dynamic pressure on April 1, 1997

The geomagnetic and auroral response to the variations in the solar wind dynamic pressure (Pd) are investigated in the periods of positive values of the IMF B _z component. It is shown that the growth of Pd results in the intensification of luminosity along the auroral oval and in the poleward expansion of the poleward boundary of luminosity (PBL) in the nightside part of the oval by ～7° in latitude at a velocity of ～0.5 km/s and is accompanied by an enhancement of the DP2-type current system. A decrease in Pd, accompanied by an abrupt reversal of the IMF B _y polarity from positive to negative, results in an enhancement of the westward electrojet and in a poleward shift of PBL and electrojet center. The conclusion has been made that the available three types of auroral response to Pd variations differ in the azimuthal velocity of the luminosity region or particle precipitation along the auroral oval: V ₁ ～ 30–40 km/s, V ₂ ～ 10, and V ₃ ～ 1 km/s.     

Interplanetary magnetic field By control of dayside auroras

The effect of the interplanetary magnetic field (IMF) By component on the dayside auroral oval from Viking UV measurements for March–November 1986 is studied. Observations of dayside auroras from Viking UV images for large positive (15 cases) and negative (22 cases) IMF By (∣By∣>4 nT), suggest that: (1) the intensity of dayside auroras tends to increase for negative IMF By and to decrease for positive By, so that negative IMF By conditions seem preferable for observations of dayside auroras; (2) for negative IMF By, the auroral oval tends to be narrow and continuous throughout the noon meridian without any noon gap or any strong undulation in the auroral distribution. For positive IMF By, a sharp decrease and spreading of auroral activity is frequently observed in the post-noon sector, a strong undulation in the poleward boundary of the auroral oval around noon, and the formation of auroral forms poleward of the oval; and (3) the observed features of dayside auroras are in reasonable agreement with the expected distribution of upward field-aligned currents associated with the IMF By in the noon sector.     

The association between giant pulsations (Pgs) and the auroral oval

Two features of giant pulsations (Pgs) which still require an explanation are firstly, why Pgs occur mainly in the early morning sector (i.e. 03:00-07:00 MLT) and not at other times of day, and secondly, why Pgs occur preferentially in a narrow latitudinal band (approximately 63○-68○ geomagnetic latitude). Using statistics from 34 Pg events observed by the EISCAT magnetometer cross, a comparison has been made between the location of the Pg resonant field lines and the equatorward edge of the auroral oval. The majority of these Pg events appear to occur just poleward of this boundary. Using these results, an explanation of the two features of Pgs as detailed above is made. This explanation involves the interaction of protons, which may be responsible for the Pg events, with the inner edge of the plasma sheet or with its ionospheric equivalent, the equatorward edge of the auroral oval.     

Position of projections of the nightside auroral oval equatorward and poleward edges in the magnetosphere equatorial plane

The position of the auroral oval poleward and equatorward boundary projections on the equatorial plane in the nightside MLT sector during magnetically quiet periods (|AL| < 200 nT, |Dst| < 10 nT) has been determined. The oval boundary positions were determined according to the precipitation model developed at Polar Geophysical Institute (http://apm.pgia.ru/). The isotropy of the averaged plasma pressure and the experimentally confirmed balance of pressures during the nighttime have been taken into account. The morphological mapping method has been used to map the oval poleward and equatorward edges without the use of any magnetic field model on the assumption that the condition of magnetostatic equilibrium is valid. Ion pressures at ionospheric altitudes and in the equatorial plane have been compared. It has been shown that the auroral oval equatorward boundary in the midnight sector is localized at geocentric distances of ~7 R_E, which is in good agreement with the position of the energetic particle injection boundary in the equatorial plane. The oval poleward edge is localized at the ~10 R_E geocentric distance, which is in good agreement with the position of the equatorward boundary of the region with a high turbulence level in the Earth’s magnetosphere plasma sheet.     

Relation between sudden increases in the solar wind dynamic pressure,auroral proton flashes,and geomagnetic pulsations in the Pc1 range

The interrelation between sudden increases in the solar wind dynamic pressure, auroral proton flashes on the dayside equatorward of the oval, and geomagnetic pulsations in the Pc1 range is considered on the basis of simultaneous observations of the solar wind plasma parameters, proton auroras on the IMAGE satellite, and geomagnetic pulsations at the Lovozero Observatory. It is indicated that proton luminosity flashes were observed in 70% of cases equatorward of the auroral oval during sudden changes in the solar wind pressure. In this case, flashes of proton auroras were observed in 85% of cases during sudden changes in the pressure, which were related to interplanetary shocks. Increases in pressure during tangential discontinuities were accompanied by flashes of proton auroras only in 45% of cases. When the ground station was conjugate to the region occupied by a proton aurora flash, the appearance or intensification of existent pulsations in the Pc1 range was observed in 96% of cases. When the ground station was not conjugate to the region of a proton luminosity flash, the response in geomagnetic pulsations was observed in 32% of events. When a sudden change in the solar wind pressure was not accompanied by a proton luminosity flash, the response in pulsations in the Pc1 range was hardly observed.     

Auroral intensification structure and dynamics in the double oval: Substorm of December 26, 2000

Based on results of the simultaneous TV observations at Barentsburg high-latitude observatory and Lovozero auroral observatory and using the IMAGE auroral luminosity images, the auroral fine structure and dynamics has been studied during the substorm of December 26, 2000, when the auroral luminosity distribution represented a double oval. It has been indicated that the interaction between the processes proceeding in different magnetospheric regions, the projections of which are the poleward and equatorward edges of the double oval, is observed in auroras in the process of substorm development.     

Position of proton precipitation during auroral intensifications

The unique spectrographic observations of auroras on the Kola Peninsula, simultaneously performed in 1970 at Loparskaya and Kem stations using C-180-S cameras, have been analyzed by up-to-date digital data processing. The position and dynamics of proton precipitation relative to other manifestations of auroral and substorm activity (auroral arcs and electrojets) under moderately and weakly disturbed conditions have been analyzed. Several previously known regularities in the morphology of proton auroras have been confirmed. It has been indicated that the direction of motion of the proton band equatorward boundary in the evening sector changes at a sign reversal of the IMF Z component. Weak breakups affect the poleward boundary of the proton band but do not influence the position of the equatorward boundary of this band, which results in the expansion of the proton emission region. When a disturbance is stronger, the proton emission disappears near an active electron arc and subsequently appears poleward of its position before intensification. Short-term proton precipitation is also observed in the region of active electron precipitation during an intense breakup in the form of N–S structures.     

Relationship between auroral oval poleward boundary intensifications and magnetic field variations in the solar wind

As a rule, bright auroral arcs evolve near the poleward boundary of the auroral oval at the growth phase of a substorm, a phenomenon that is known to occur near the poleward edge of the auroral oval. The closeness of these arcs to the projection of the magnetic separatrix on the night side suggests that their generation is related to magnetic reconnection in the magnetospheric tail in a particular way. In this study this suggestion is confirmed by the fact that integral brightness of the auroral oval at the poleward edge correlates with magnetic field structures in the solar wind that are observed by ACE and Wind satellites at distances of 50–300 R_E upstream and are shifted towards the magnetospheric tail with time delays of ~ 10–80 min, consistent with measurements of the solar wind velocity. About 50 examples of this correlation have been found. The possible physical mechanisms of the effect observed are discussed.