渤海湾盆地歧口凹陷地层压力结构特征

深化含油气沉积盆地的压力结构研究,厘清异常压力的空间展布,对划分含油气系统、评价有利输导体系与明确勘探甜点区带具有重要的理论和实践意义。为深化渤海湾盆地富油凹陷的油气二次勘探,本文以歧口凹陷为研究对象,对其压力结构进行重点刻画。在实测地层压力的校正下,综合单井、连井和二维地震地层压力结构分析,厘清了歧口凹陷的压力结构特征,识别出4类纵向压力结构：①单超压带结构;②双超压带结构;③多超压带结构;④静水压力结构。纵向上,歧口地区存在3类纵向压力系统样式——单超压系统、双超压系统、静水常压系统。双超压系统是歧口凹陷的主要压力系统样式,广泛发育于主凹和各大次凹;从凹陷中心向盆地边缘,双超压系统逐渐向单超压系统、静水常压系统过渡。单超压系统主要分布于盆地边缘的斜坡和潜山区,如歧北高斜坡、羊三木-扣村潜山等。静水常压系统则主要分布在离深凹区更远的沈青庄潜山和埕北斜坡区域。上部超压系统和下部超压系统的顶板分别位于东营组和沙三段内部,侧向上受盆地边缘和深大断裂控制。上部超压系统的形成主要受欠压实作用控制,以歧口主凹为中心呈环带分布;而下部超压系统的形成主要受生烃作用控制,以主凹和几大次凹为中心分布。未来,下部超压系统中保存的天然气将成为歧口地区超深层天然气勘探的重点对象。     

Bedrock erosion due to hoeing as tillage technique in a hilly agricultural landscape,southwest China

Tillage on hillslopes may not only induce severe soil erosion, but may also cause bedrock erosion under certain conditions. Yet, little is known about bedrock erosion by tillage in a hilly agricultural landscape, southwest China. The aim of this study is to quantify the translocation of rock fragments derived from bedrock fragmentation by hoeing under different conditions, including slope gradient, hoeing depth and soil-covered thickness using a gravel tracing method. The reliability of the gravel tracing method was confirmed by the bedrock dyeing tracing method. Hoeing depth is a significant factor affecting the translocation rate of rock fragments (Q_r ). Meanwhile, under the condition of overlying soil layers (0.06−0.10 m thick), the values of Q_r were significantly smaller with a reduction of 20.7−25.6%, compared with rock fragmentation by hoeing for bare bedrock. However, slope gradient was found to have insignificant effects on Q_r . Fractured bedrock moved as individual small fragments, which was mainly controlled by the hitting force of the hoe, while soil moved in the shape of lumps, which was dominated by both drag force of the hoe and gravity. This study suggests that hoeing into soil-covered bedrock can diminish bedrock erosion while providing soil matrix for shallow soil layers. Our work presents a quantitative assessment of bedrock erosion by hoeing and an underlying insight into characteristics of bedrock erosion by tillage operations in hilly agricultural regions with mudstone and shale, southwest China. © 2020 John Wiley & Sons, Ltd.     

短期重现性地磁日变化感应电流集中分布与地震关系初步研究

通过分析台站之间地磁垂直分量日变化相关性,研究中国大陆南北带地区90°E～110°E,2009～2018年地磁日变化感应电流线状集中分布现象与M_S≥6.0地震关系。统计发现,我国每月平均出现约2次感应电流集中分布现象,且10天左右内有原地重复出现现象——感应电流集中分布重叠异常,原地重现的集中分布感应电流日期间隔一般在10天左右以内,原地重现日期一般不连续,重现次数一般仅有2次。研究发现,我国南北带2009～2018年共出现重叠异常23次,20次重叠异常出现后18个月内重叠段发生M_S≥6.0地震,对应率高达87%,地震基本发生在重叠段端部250km以内,重叠段发生2次M_S≥6.0地震的比例较低,约35%;重叠段2个端部均发生M_S≥6.0地震的比例极小,仅约9%。部分重叠段走向与已经发现的上地幔和地壳高导带埋深走向基本一致,推测重叠异常发生在上地幔和地壳高导带附近,是来自上地幔和地壳内高导带附近的地震异常信息。目前已知,中国大陆强地震主要发生在上地幔高导层的隆起区一带及其壳内高导层发育地区,深部电性测深结果表明,大震易于发生在电导率急剧变化的梯度带,但这些地区何时发生地震还未得知,本文发现的重叠异常解释了这些地区1～2年内是否发生地震这一问题。     

基于深度卷积神经网络的剪切波分裂质量检测

剪切波分裂是分析地震各向异性的一种重要手段,常规方法是利用网格搜索获取分裂参数,再通过不同方法的测量结果对比测量结果进行质量检测,这一过程会耗费大量计算时间。本文针对这一问题提出了一种利用深度卷积神经网络对剪切波分裂进行质量检测的新方法,对使用了Resnet残差结构的深度神经网络进行训练,直接对二分量剪切波波形数据的质量进行分类。整个过程为:神经网络通过卷积层提取波形特征,计算损失函数后反向传播训练模型参数,完成迭代训练后的模型对输入波形数据正向计算自动输出类型。本文利用川西台站接收到的实际数据以及随机生成的合成数据分别对该网络进行训练,均可以获得准确的分类结果。相比于通过多种剪切波分裂方法对比测量结果的质量检测方法,基于神经网络的方法可以省略网格搜索的计算过程直接判断质量类型,在运算速度上的优势明显,并可继续通过训练提高模型的精度,为提升剪切波分裂方法在数据处理过程中的操作效率提供帮助。     

Rock crevices determine woody and herbaceous plant cover in the karst critical zone

The study of the critical zones(CZs) of the Earth link the composition and function of aboveground vegetation with the characteristics of the rock layers, providing a new way to study how the unique rock and soil conditions in karst regions affect the aboveground vegetation. Based on survey results of the rocks, soils and vegetation in the dolomite and limestone distribution areas in the karst area of central Guizhou, it was found that woody plant cover increases linearly with the number of cracks with a width of more than 1 mm, while the cover of herbaceous plants shows the opposite trend(p0.01). The dolomite distribution area is characterized by undeveloped crevices, and the thickness of the soil layer is generally less than 20 cm, which is suitable for the distribution of herbaceous plants with shallow roots. Due to the development of crevices in the limestone distribution area, the soil is deeply distributed through the crevices for the deep roots of trees, which leads to a diversified species composition and a complicated structure in the aboveground vegetation. Based on moderate resolution imaging spectroradiometer(MODIS) remote sensing data from 2001 to 2010, the normalized differentiated vegetation index(NDVI) and annual net primary productivity(NPP) results for each phase of a 16-day interval further indicate that the NDVI of the limestone distribution area is significantly higher than that in the dolomite distribution area, but the average annual NPP is the opposite. The results of this paper indicate that in karst CZs, the lithology determines the structure and distribution of the soil, which further determines the cover of woody and herbaceous plants in the aboveground vegetation. Although the amount of soil in the limestone area may be less than that in the dolomite area, the developed crevice structure is more suitable for the growth of trees with deep roots, and the vegetation activity is strong. At present, the treatment of rocky desertification in karst regions needs to fully consider the rock-soilvegetation-air interactions in karst CZs and propose vegetation restoration measures suitable for different lithologies.     

浙江省测震观测台网井下地震计方位角检测与校正

选取浙江省测震台网景宁、北仑、南麂岛地震台井下地震计观测数据,采用地震数据波形相关性分析法,与地表地震计同步观测数据进行相关性分析,检测并校正井下地震计方位角,结果发现,3个地震台井下地震计方位角实际偏差分别为82.2°、-22.8°、32.2°。分析认为,井下地震计安装易受线缆应力影响,导致方位角与实际偏差过大,应在应力释放完成后固定地震计,并架设地表地震计进行数据相关性分析,以确定方位角可靠性。     

强震动记录H/V谱比法计算处理的若干关键环节

强震动记录H/V谱比法自提出以来,已广泛应用于地震工程各研究领域中。但对数据处理的2个关键环节——Taper预处理及傅氏谱平滑未有详细研究。因此,本文针对强震动记录H/V谱比法计算涉及的重要数据处理过程,对四川地区19个强震动台站在汶川地震余震中获取的642组强震动记录开展研究。研究实例表明:S波H/V谱比振幅在周期 < 1s时高于全时程,论证计算H/V谱比时截取S波窗口的必要性;S波窗口的截断会引起傅里叶振幅谱的边瓣效应,若不加以处理,将显著影响低频部分H/V谱比结果,Taper预处理对于消除这种截断误差具有良好效果;在兼顾平滑效果及卓越周期识别准确度的基础上,通过不同带宽的Parzen窗试算,认为0.5Hz带宽宜用于傅里叶H/V谱比曲线平滑。     

面向地震巨灾保险的建筑特性快速提取方法

房屋建筑分类是抗震设计和地震风险分析的基础,是巨灾保险的纽带环节,也是结构易损性准确、完备分析的前驱保障,快速获取建筑特性参数非常关键。基于影像数据获取结构特性相比传统手段具有显著优势,然而其准确性具有一定挑战性,从影像数据得到实时的、较准确的结构特性成为地震保险数据获取技术的关注焦点。本文采用深度学习方法开展从影像数据中提取面向地震保险需求的建筑特性数据,构建基于深度学习方法的建筑高度识别模型和基于机器视觉的建筑高度识别方法,运用基于Xception神经网络深度学习和机器视觉的模型,对北京地区的建筑高度进行模型测试,该方法可为地震保险分析提供重要的基础数据支持。     

Novel uniaxial concrete constitutive model considering bond-slip effect

A uniaxial concrete constitutive model considering the bond-slip effect is proposed and its finite element analysis (FEA) implementation on a fiber section     

Post-collisional mafic magmatism: Record of lithospheric mantle evolution in continental orogenic belt

Science China Earth Sciences - In order to better understand the role of post-collisional mafic magmatism at convergent plate boundaries in revealing the earth’s evolution, this paper has...