利用闪电宽带干涉仪系统对地闪先导-回击过程的观测研究

利用闪电宽带干涉仪系统1999年在广东的观测资料, 对地闪预击穿、梯级先导及回击过程进行分析研究, 结果表明: 这种新型的辐射源定位系统能够再现闪电通道结构和电磁辐射源随时间演变的二维图像(方位角和仰角). 闪电辐射主要由负击穿过程产生, 地闪预击穿阶段闪电通道比较集中, 几乎没有分杈, 辐射源主要出现在通道头部; 而梯级先导后期辐射源定位比较离散, 有较多的分叉通道, 辐射源出现在具有一定长度通道段内, 该通道段高度随着先导向地面的发展逐渐降低. 预击穿过程和梯级先导期间均有一些快速负流光过程连续地从闪电起始区域或更远的位置开始, 沿已形成的通道发展, 为闪电通道进一步向前发展补充电荷, 其速度比闪电通道发展的平均速度约快一个量级.     

正地闪发展的时空结构特征与闪电双向先导

利用闪电VHF辐射源定位系统的观测资料, 对正地闪三维时空发展特征进行了分析, 结果表明正地闪过程大致可分为三个阶段, 在回击之前有较长的云内发展过程, 平均持续时间为370 ms, 传输速度为10⁵ m/s量级, 辐射强度与负地闪梯级先导辐射强度相当, 这一期间闪电是以负极性击穿过程在云中正电荷区发展, 通道沿水平方向延伸, 较少分叉;回击之后闪电在云内快速传输, 比回击前的发展速度快约2倍, 辐射点较少且比较弥散, 但辐射强度增强, 对应于连续电流过程, 并包含有多个正极性快脉冲;在闪电的最后阶段, 传输速度与回击前的传输速度相当, 辐射点主要集中在闪电通道的顶端. 与负地闪的时空发展特征存在明显的差异. 正地闪的辐射点均发生在云内正电荷区, 没有探测到回击之前的正先导过程所产生的辐射. 正地闪持续时间平均为730 ms, 持续时间在500~600 ms的闪电占总数的43%. 90%正地闪只有一次回击, 最多有4次回击, 回击电流峰值最大为70 kA, 最小为11.5 kA, 平均为36.5 kA, 大于40 kA的约有40%.     

三维地震动场数值模拟并行计算系统

并行计算系统是进行三维复杂场地地震动场数值模拟的基础。本文作者利用10台微机构建了一套并行计算集群(常称为Beowulf系统),节点微机上运行Windows2000操作系统,采用MPI标准的MPICH1．2．4版本作为并行计算的支撑环境,并以100M bps高速交换式以太网作为互连网络。利用编制的并行计算程序对该并行计算系统性能进行了测试。结果表明：该系统能够达到非常高的并行加速比和并行效率。该系统将用于三维地震动场数值模拟研究。     

长周期地震动三维有限元数值模拟方法

采用数值方法对长周期地震动进行模拟已成为当前强地面运动预测的研究方向之一。本文提出了一种有限元并行计算的高效地震动数值模拟方法,可以同时考虑震源、传播途径和场地条件对地震动的影响,并根据昆明地区未来可能发生的设定地震对昆明市的影响地震动进行了预测。这一方法可以应用于国内已探测的发震活断层进行地震危害性评价工作。     

地电台址构造条件的三维有限元数值模拟

本文应用粘弹性三维有限元程序,模拟了在定常速率加载边界条件下,断层对于层状介质应力、应变场的影响,并进而通过应力—应变场与地电阻率变化的关系,最终揭示出断层对地电观测的影响。计算结果表明,断裂所处位置、断裂的力学性质及类型对地电台址的选择有重要意义。破碎且较宽的平行断层对地电信息有一定的阻隔作用,而久不活动的被压实的断裂、垂直断层的中段、雁列式断裂端点对地电信息却有增强作用。另外本文还讨论了发震构造的类型对地电观测的影响,结果表明断错型地震的前兆信息强,而走滑型地震的前兆信息弱。     

利用无线电窄带干涉仪定位系统对地闪全过程的观测与研究

利用VHF无线电窄带干涉仪定位系统对一次包含有19次回击的地闪全过程产生的甚高频辐射源进行了定位研究,该系统基于双基线正交5天线阵列,采用直接高频放大检相,降低了采用变频电路时带来的系统误差,提高了整个系统的精度;系统采用交互式的图像分析程序来处理干涉仪本身存在的条纹模糊问题,以1μs的时间分辨率连续确定闪电辐射源发生的二维位置（仰角和方位角）,直观地再现了闪电放电随时空发生和发展的全过程．利用窄带干涉仪系统并结合同步观测的VHF辐射强度和快电场变化资料,对一次负极性地闪发生和发展的全过程研究发现,负地闪预击穿过程起始于负电荷区,先向下后向上发展;梯级先导辐射强且连续,且首次回击开始若干毫秒内辐射强度更大;直窜先导辐射相对离散且强度弱,由于通道电离程度的变化,梯级先导和直窜先导可以相互转化;初始梯级先导速度约为10^5m／s,直窜先导的平均速度约为4．1×10^6m／s,梯级一直窜先导的平均速度约为6．0×10^6m／s;M过程伴随有活跃的爆发式辐射,产生勾状电场变化,击穿发展最终沿着导电性良好的主放电通道迅速到地,其平均速度约为7×10^7m／s,大于直窜先导和梯级直窜先导平均速度;K变化和企图先导与先导本质上是一样的,只是没有到达地面而引发回击．     

地震地电阻率各向异性变化数值模拟

将APE模式中差应力与裂隙的演化理论与前期研究中建立的张量形式的各向异性电性模型相结合,推导出了裂隙演化参数(密度、孔压、纵横比、闭合角)与电导率关系的动态方程,用数值模拟方法研究了饱水岩石在水平向应力作用下电阻率的各向异性变化．并结合实验和观钡5结果,讨论了电阻率各向异性变化的可能机理。     

多辐射源地空瞬变电磁响应三维数值模拟研究

地空瞬变电磁法结合地面和航空电磁法的优点,可实现探测深度和工作效率的平衡.当前地空瞬变电磁法采用单一线源激发电磁辐射场,仅能从一个侧面与地质体耦合,难以获得地质体的全息影像.采用多辐射源是解决这一问题的途径.本文采用三维矢量有限元法对两个不同地质体多个辐射源情况下的地空瞬变电磁响应开展了模拟研究,分析了多辐射源在不同辐射方向、不同飞行高度电磁响应的分布特征.研究表明,由多辐射场源作为地空电磁法的发射源,通过分散布设的线源,可以在地下激发与地质体多方位耦合的电磁场,能够获得地下地质体多方位不同高度情况下的耦合信息.同时,多辐射场源能够增强源电磁场的辐射强度,减少单一线源体积效应影响,飞行高度较低时可获得较强的响应幅值,研究结果为多辐射地空瞬变电磁法深部精细探测提供理论依据.

     

沉陷区域埋地管线数值模拟分析

场地的不均匀沉陷是导致埋地管线破坏的重要原因之一。本文考虑了材料非线性、几何非线性以及管土接触非线性,将管线计算分析模型模拟为四节点薄壳单元结构,周围填覆土体采用八节点六面体单元划分。管土相互作用模拟为三维刚性与柔性的面面接触单元结构,并采用线性位移加载来模拟土体的沉陷作用,对三维薄壳有限元模型进行数值计算分析。通过比较不同参数,如沉陷长度、沉陷深度、埋深、管径、径厚比、土特性等对管线的反应影响,得出管线在沉陷情况下的应力和应变的关系,通过算例分析,说明了该方法能更好地模拟管线的破坏过程,该方法将为沉陷区域埋地管线数值模拟提供理论分析依据。     

大陆碰撞动力学的三维数值模拟

用三维有限单元模型模拟了印度—亚洲大陆碰撞在板内的力学效应。结果表明,印度板块北东方向的推挤力作用在亚洲大陆内造成逆断层和走滑断层型的应力状态,以及北西走向的平行的弧形应力等值线。岩石圈下伏岩层的塑性变形导致地壳变形范围扩大,垂直位移的水平梯度减小,应力传递的水平距离显著增加。在相对坚硬的地盾型构造区边缘,出现应力梯度较高的特征。对比了刚体水平挤入和低角度俯冲这两种大陆聚合模型,计算结果表明低角度俯冲是较合理的模型     

Three-dimensional EM computer simulation on sprite initiation above a horizontal lightning discharge

The purpose of this paper is to study the three-dimensional (3D) effects of the source current and the electromagnetic (EM) pulse on the distribution of upper atmospheric electric field and ionization created. A new lightning model has been employed; i.e., we have included a horizontal channel in addition to the conventional vertical channel. The effects of the horizontal lightning channel are summarized as follows: (1) the effect of a vertical channel plays a fundamental role in the whole view of sprite initiation; (2) the position of a sprite is shifted from the position of its parent vertical channel in response to the length of a horizontal channel; and (3) we observe very fine structures with local maxima and minima in the “reduced” electric field (the electric field divided by the neutral gas density). These theoretical inferences are discussed and compared with the observational facts so far reported (such as lateral shift of sprites, morphological difference of sprites, etc.).     

青藏高原一次地闪放电过程的分析

利用成像率为1000 幅/s的高速摄像系统和快、慢电场变化仪以及宽带干涉仪系统等探测仪器在青藏高原那曲地区所观测的地闪资料,对一次地闪回击及其之前的持续时间较长的云内放电过程进行了分析.结果表明：地闪先导前的云内放电过程发生于雷暴云下部正电荷区和中部负电荷区之间;云中部负电荷区距离地面的高度为2.8～4.5 km;闪电的起始放电发生区域距离地面的高度为1.0～1.7 km;初始流光在云外发展时具有很大的水平分量和较多的分支;梯级先导的速度为1×105 m/s,在向地面发展时出现较大的弯曲;首次回击放电过程与低海拔地区没有差异,通道中的峰值电流有24.1 kA;继后回击相对较弱.     

Spatiotemporal characteristics of positive cloud-to-ground lightning discharges and bidirectional leader of the lightning

The negative CG lightning discharges neutralizing negative charges in cloud usually dominate for most of thunderstorms. However, a lot of positive CG light-ning discharges often occur in the disappearing stage of thunderstorms, in the stratiform region of mesoscale convective systems and some supercells producing hail and tornado. Because the positive CG lightning discharges produce larger current of the return stroke and neutralize more charges due to the continuing currents with longer las…     

Development characteristics of cloud-to-ground lightning with multiple grounding points

Using the optical images of a cloud-to-ground lightning flash with multiple grounding points obtained by a highspeed video system in the Qinghai Province of China along with synchronous radiated electric field information, the propagation characteristic and the electric field change features of the leaders and the grounding behavior of discharge channels are analyzed.In addition, the two-dimensional velocity of the leader was estimated and its correlation with the time interval of the corresponding subsequent return stroke, and that with the peak current of return stroke are investigated. The results show that the average distance between the three obvious grounded points of the first return stroke channel is about 512.7 m, and the average time interval between the pulses of the corresponding electric field fast changes is 3.8 μs. Further, the average time interval between electric field pulses from the stepped leader is smaller than that of normal single grounding lightning. The observed lightning in our study has two main channels, namely the left and right channels. Based on our observations, it is clear that the dart leader comes close to the ground in case of the left channel after the first return stroke, but it fails to form a return stroke.However, the right channel exhibits a relatively rare phenomenon in that the subsequent return stroke R2 occurred about 2.1 ms after the dart leader arrived at the ground, which was unusually long; this phenomenon might be attributed to the strong discharge of the first return stroke and insufficient charge accumulation near the grounded point in a timely manner. The two-dimensional velocities for the stepped leader of the two main channels are about 1.23×105 and 1.16×105 m s-1, respectively. A sub-branch of stepped leader for the left channel fails to reach the ground and develops into an attempt leader eventually; this might be attributed to the fact that the main branch connects considerably many sub-branches, which leads to the instantaneous decline of the potential difference between the sub-branch and ground. Furthermore, it might also be because the propagation direction of this sub-branch is almost perpendicular to the atmospheric electric field direction, which is not conducive to charge transfer. The two-dimensional velocities for the dart leaders of five subsequent return strokes are all in the normal range, and they positively correlate with the peak current of the subsequent return stroke.     

Ten-year study of cloud-to-ground lightning activity in the Iberian Peninsula

The characteristics (annual and diurnal cycle, polarity, multiplicity and first stroke peak current) of ∼4.3×10⁶ cloud-to-ground flashes recorded in the Iberian Peninsula during the first decade of measurements of the lightning detection network installed in Spain are analyzed. The mean monthly variation shows maximum lightning activity between May and September, while minimum values are observed in January and February. The mean diurnal cycle shows maximum values at 1700 LT and minimum values at around 1000 LT. The average maximum flash density (not corrected for detection efficiency) is 2.1 flashes km⁻² year⁻¹. Maximum lightning activity is associated with mountainous areas. The effect of the Mediterranean Sea is also seen. The percentage of positive flashes is 9%, although this changes over the year from 6.5% in June to 22.6% in January. The average multiplicity is found to be 2.0 for the negative flashes and 1.1 for the positive flashes, and the percentages of single-stroke flashes are 53.6% and 89%, respectively. The monthly distribution of multiplicity for negative flashes peaks in the summer and minimum values are found in the winter. The multiplicity of the positive flashes does not seem to be function of the month. The median (mean) first stroke peak currents are found to be 23.5 kA (27.3 kA) for the negative flashes and 35.3 kA (47.1 kA) for the positive flashes. For both polarities, the peak current is higher in the summer than in the winter. The percentage of positive flashes and the mean peak currents for negative flashes are higher over the sea areas than over land.     

Three-dimensional numerical simulation of glacial trough forming process

The glacial trough is a common glacier erosion landscape, which plays an important role in the study of glacier erosion processes. In a sharp contrast with the developing river, which is generally meandering, the developing glacial trough is usually wide and straight. Is the straightness of the glacial trough just the special phenomenon of some areas or a universal feature? What controls the straightness of the glacial trough? Until now, these issues have not been studied yet. In this paper, we conduct systematic numerical models of the glacier erosion and simulate the erosion evolution process of the glacial trough. Numerical simulations show that:(1) while the meandering glacier is eroding deeper to form the U-shaped cross section, the glacier is eroding laterally. The erosion rate of the ice-facing slope is bigger than that of the back-slope.(2) The smaller(bigger) the slope is, the smaller(bigger) the glacier erosion intensity is.(3) The smaller(bigger) the ice discharge is, the smaller(bigger) the glacier erosion intensity is. In the glacier erosion process, the erosion rate of the ice-facing slope is always greater than that of the back-slope. Therefore, the glacial trough always develops into more straight form. This paper comes to the conclusion that the shape evolution of the glacial trough is controlled mainly by the erosion mechanism of the glacier. Thereby, the glacial trough prefers straight geometry.     

随机性与电环境特征对地闪击地点影响的数值模拟

本文利用已有的随机放电参数化方案,结合四次探空资料,进行了12.5 m的高分辨率二维雷暴云数值模拟实验,得到了各种雷暴云电荷结构下的地闪个例,并就地闪击地点与空间电荷、电位分布之间的相互关系进行了分析.结果表明:(1)由空间电荷唯一确定的电位分布决定了先导的传播最大趋势,而闪电传播的随机性所带来的地闪击地点的不确定范围被限制在3 km之内,利用动态聚类法迭代得出的三个击地点位置之间的差为1 km左右.(2)负地闪的初始点与击地点的位置差主要分布在0~6 km范围内,且93%的负地闪分布在0~4 km范围内,正地闪的分布相对较广,0~3 km范围内占48%,3~6 km范围内占34%,6~10 km范围内占18%.(3)正、负地闪主要产生于离地面最近的一对电荷堆之间,其起始高度越高,初始点与击地点位置差分布越广;另外,产生于三级性雷暴云电荷结构下的正地闪,其起始于上部的主正电荷堆与中部主负电荷堆之间,由于下行正先导会绕过底部的次正电荷堆,因此其击地点与初始点的距离基本在6 km以上.     

The first electric field pulse of cloud and cloud-to-ground lightning discharges

In this study, the first electric field pulse of cloud and cloud-to-ground discharges were analyzed and compared with other pulses of cloud discharges. Thirty eight cloud discharges and 101 cloud-to-ground discharges have been studied in this analysis. Pulses in cloud discharges were classified as ‘small’, ‘medium’ and ‘large’, depending upon the value of their relative amplitude with respect to that of the average amplitude of the five largest pulses in the flash. We found that parameters, such as pulse duration, rise time, zero crossing time and full-width at half-maximum (FWHMs) of the first pulse of cloud and cloud-to-ground discharges are similar to small pulses that appear in the later stage of cloud discharges. Hence, we suggest that the mechanism of the first pulse of cloud and cloud-to-ground discharges and the mechanism of pulses at the later stage of cloud discharges could be the same.     

地闪对崇明地震台地球物理观测的干扰

通过对比崇明地区地闪及崇明地震台地球物理观测资料,发现受地闪引起的电磁场变化的直接作用或对仪器元件的间接作用,地磁、地电、电磁扰动和水位观测干扰比例较高,干扰幅度与地闪距离及形成的电流强度有关。具体干扰形态如下：①对电磁扰动干扰表现为单向突跳和测值的整体抬升;②对地磁测项干扰表现为正负方向的单点突跳;③对大地电场干扰表现为大幅震荡;④对水位观测干扰表现为大幅突跳。