Subduction initiation in the Neo‐Tethys: constraints from counterclockwise P–T paths in amphibolite rocks of the Nagaland Ophiolite Complex,India

High‐P metamorphic rocks that are formed at the onset of oceanic subduction usually record a single cycle of subduction and exhumation along counterclockwise (CCW) P–T paths. Conceptual and thermo‐mechanical models, however, predict multiple burial–exhumation cycles, but direct observations of these from natural rocks are rare. In this study, we provide a new insight into this complexity of subduction channel dynamics from a fragment of Middle‐Late Jurassic Neo‐Tethys in the Nagaland Ophiolite Complex, northeastern India. Based on integrated textural, mineral compositional, metamorphic reaction history and geothermobarometric studies of a medium‐grade amphibolite tectonic unit within a serpentinite mélange, we establish two overprinting metamorphic cycles (M₁–M₂). These cycles with CCW P–T trajectories are part of a single tectonothermal event. We relate the M₁ metamorphic sequence to prograde burial and heating through greenschist and epidote blueschist facies to peak metamorphism, transitional between amphibolite and hornblende‐eclogite facies at 13.8 ± 2.6 kbar, 625 ± 45 °C (error 2σ values) and subsequent cooling and partial exhumation to greenschist facies. The M₂ metamorphic cycle reflects epidote blueschist facies prograde re‐burial of the partially exhumed M₁ cycle rocks to peak metamorphism at 14.4 ± 2 kbar, 540 ± 35 °C and their final exhumation to greenschist facies along a relatively cooler exhumation path. We interpret the M₁ metamorphism as the first evidence for initiation of subduction of the Neo‐Tethys from the eastern segment of the Indus‐Tsangpo suture zone. Reburial and final exhumation during M₂ are explained in terms of material transport in a large‐scale convective circulation system in the subduction channel as the latter evolves from a warm nascent to a cold and more mature stage of subduction. This Neo‐Tethys example suggests that multiple burial and exhumation cycles involving the first subducted oceanic crust may be more common than presently known.     

High‐P tectono‐metamorphic evolution of mylonites from the Variscan basement of the Northern Apennines,Italy

Strain localization within shear zones may partially erase the rock fabric and the metamorphic assemblage(s) that had developed before the mylonitic event. In poly‐deformed basements, the loss of information on pre‐kinematic phases of mylonites hinders large‐scale correlations based on tectono‐metamorphic data. In this study, devoted to a relict unit of Variscan basement reworked within the nappe stack of the Northern Apennines (Italy), we investigate the possibility to reconstruct a complete pressure (P)–temperature (T)–deformation (D) path of mylonitic micaschist and amphibolite by integrating microstructural analysis, mineral chemistry and thermodynamic modelling. The micaschist is characterized by a mylonitic fabric with fine‐grained K‐white mica and chlorite enveloping mica‐fishes, quartz, and garnet pseudomorphs. Potassic white mica shows Mg‐rich cores and Mg‐poor rims. The amphibolite contains green amphibole+plagioclase+garnet+quartz+ilmenite defining S₁ with a superposed mylonitic fabric localized in decimetre‐ to centimetre‐scale shear zones. Garnet is surrounded by an amphibole+plagioclase corona. Phase diagram calculations provide P–T constraints that are linked to the reconstructed metamorphic‐deformational stages. For the first time an early high‐P stage at >11 kbar and 510°C was constrained, followed by a temperature peak at 550–590°C and 9–10 kbar and a retrograde stage (<475°C, <7 kbar), during which ductile shear zones developed. The inferred clockwise P–T–D path was most likely related to crustal thickening by continent‐continent collision during the Variscan orogeny. A comparison of this P–T–D path with those of other Variscan basement occurrences in the Northern Apennines revealed significant differences. Conversely, a correlation between the tectono‐metamorphic evolution of the Variscan basement at Cerreto pass, NE Sardinia and Ligurian Alps was established.     

基于帧间差分的彩色视频背景提取

针对动态目标检测中的图像背景实时提取问题,采用红绿蓝三分量综合考虑的方法,进行了比较深入的算法研究。研究表明,如果图像序列时间段过短,背景中就会有目标空洞,如果图像序列时间段过长,则不能反映实时背景,在后续目标检测中就会引起较大误差。考虑合理的视频场景区域的动态目标密集程度因素,就可以较好地兼顾提取背景的实时性和背景中目标空洞的消除。     

复杂层状模型中多次波快速追踪——一种基于非规则网格的最短路径算法

使用不规则网格单元划分下的最短路径算法,结合分区多步计算技术实现了二维和三维复杂层状起伏介质中的多次透射、反射及转换波的追踪计算.其原理是将模型按速度界面分成若干个独立的计算区域(在速度界面和起伏地表处采用一种不规则网格单元划分),采用分步计算技术进行多次波的追踪计算.通过与有限差分下快速行进法的比较,表明无论是计算精度还是CPU时间,不规则最短路径算法均好于快速行进法算法.最后,实例模拟中给出了二维、三维复杂层状模型(包括Marmousi模型及含低速体的模型)中的多次透射、反射及转换波的追踪计算,验证了不规则最短路径算法的功能.     

列车引起的地基土应力状态变化的三维有限元分析

将列车移动荷载简化为多个移动轮轴荷载,基于列车-轨道-路基解析模型推求的列车运行时不同时刻、不同位置时作用于路基的振动荷载时程,采用多点输入方式实现列车荷载的移动施加方式,建立路(地)基-场地体系三维有限元动力分析模型,基于Abaqus软件的并行计算集群平台,对轨道交通振动荷载下路(地)基-场地体系的动力反应进行数值模拟,研究了列车荷载作用线正下方地基中的动应力特征及土单元应力状态变化,分析了列车轮轴荷载移动过程中不同深度处土单元的应力路径和主应力轴的旋转。     

上海地区基坑开挖卸荷土体回弹变形试验研究

通过对上海地区某基坑工程坑底不同深度土样进行一系列K0固结不排水三轴卸荷试验,对上海地区不同深度土层随基坑开挖卸荷变形性状进行研究。研究不同深度的不同土层在两种典型卸荷应力路径下的变形特性,分析了回弹率及回弹模量与卸荷比的变化关系。结果表明,软土的初始卸荷模量Ei及最大回弹率 不仅取决于土层性质,还与所经历的不同卸荷应力路径、预压最大荷载等有关,且不同卸荷路径下土体发生强回弹变形的极限卸荷比亦不同     

饱和地基上列车运行引起的地面振动隔振分析

利用推导出的频域内饱和土的以土骨架位移u和孔隙水压力p表示的u-p格式的2.5维有限元方程,分析列车低速和高速运行条件下不同隔振措施的隔振效果,包括空沟、填充沟（混凝土墙）和混凝土板（波阻板）。其中轨道被简化为饱和地基上的Euler梁,通过沿轨道方向的波数变换将三维空间问题降为平面应变问题,对求解的结果进行Fourier反变换,获得三维时域-空间域内的解。研究表明：弹性地基和饱和地基上设置空沟和填充沟隔振效果相差不大,而设置波阻板隔振效果存在较大差异。列车低速运行时,空沟基本无隔振效果,填充沟在一定范围内有隔振效果,波阻板隔振效果最好;列车高速运行时,空沟在较大沟宽和较深沟深条件下才能达到较好隔振效果,填充沟和波阻板只需较小尺寸就能达到理想的隔振效果     

基于PPP的对流层延迟估计方法及其影响因素分析

介绍利用精密单点定位（PPP）技术进行天顶对流层延迟（ZPD）估计的方法,从投影函数模型选取、卫星截止高度角设置、精密星历与精密钟差的使用3方面分析了各种因素对天顶对流层延迟估计精度的影响,确定了相对较优的模型和数据处理策略。大量的算例和分析表明：采用NMF与GMF均可获得较高精度的ZPD,二者差异甚小;采用5°~10°的截止高度角更利于得到较好的ZPD结果;采用快速精密星历和钟差、实时观测精密星历和快速精密钟差解算的ZPD结果与采用事后精密星历和钟差的精度是相当的,而采用外推超快精密星历和快速精密钟差解算测站ZPD值的结果精度稍有下降,但仍具较高的精度。     

中国城市经济网络结构空间特征及其复杂性分析

通过对当前城市体系研究进展和分析方法的阐述与总结,结合统计物理中复杂网络分析 工具,采用GIS、Matlab 和数据库等技术手段,构建城市联系网络,一定程度上突破了传统的等级 或位序城市关系研究,最后以2003 年和2007 年的Top1、Top5 和Top10 网络(城市联系强度的最大 值、前5、前10 名所连接的城市)为例证进行实证研究。研究表明：① 可将中国城市划分为北方城 市区、长江城市区、南部城市区３大城市区,形成了“三极多核”的空间格局。全国城市网络的联 系方向主要集中在环渤海地区、长三角地区和珠三角地区的主要城市(如北京、上海、广州)。但 是,从三大地区内部差异来看,环渤海地区城市等级性明显,长三角地区和珠三角地区呈现多核 化发展趋势,即城市网络化发展加强。Top10 网络反映各个城市前10 名的联系方向有所分散,即 呈现“最大联系极化、联系广度分散化”的发展趋势;② 从Top5、Top10 网络与随机网络、空间邻 近网络的对比来看,中国城市的实际网络既具有空间邻近网络这种规则网络的高聚类系数,又 具有随机网络所拥有的稀疏、长距离连接特性,表现为高聚类系数和短路径的双重特性,呈现 “小世界”网络特征;③ 从聚类系数的纵向变化看,2007 年比2003 年的城市网络连接更加分散, 区域的开放程度正在提高;④ 2007 年的平均路径长度稍有下降,说明全国城市网络中城市联系 更为紧密,连接通道增多。     

移动格林基函数样条二维插值算法研究

针对用于插值的已知点较多时,插值计算需要解算大规模矩阵、计算耗时长甚至无法解算的问题,引入移动曲面的思想,取插值点周边最邻近k个已知点进行格林基函数二维样条移动插值,实例计算结果表示,该方法的插值精度高于Shepard插值法与多项式拟合法的精度。插值范围大及测点数量众多时,该方法仍可用,无需数据分区与光滑接边,与整体插值相比可大大降低计算时间。