Particle‐size evidence of barrier estuary regime as a new proxy for ENSO climate variability

We investigate a new proxy for ENSO climate variability based on particle‐size data from long‐term, coastal sediment records preserved in a barrier estuary setting. Corresponding ~4–8 year periodicities identified from Wavelet analysis of particle‐size data from Pescadero Marsh in Central Coast California and rainfall data from San Francisco reflect established ENSO periodicity, as further evidenced in the Multivariate ENSO Index (MEI), and thus confirms an important ENSO control on both precipitation and barrier regime variability. Despite the fact that barrier estuary mean particle size is influenced by coastal erosion, precipitation and streamflow, balanced against barrier morphology and volume, it is encouraging that considerable correspondence can also be observed in the time series of MEI, regional rainfall and site‐based mean particle size over the period 1871–2008. This correspondence is, however, weakened after c.1970 by temporal variation in sedimentation rate and event‐based deposition. These confounding effects are more likely when: (i) accommodation space may be a limiting factor; and (ii) particularly strong El Niños, e.g. 1982/1983 and 1997/1998, deposit discrete >cm‐thick units during winter storms. The efficacy of the sediment record of climate variability appears not to be compromised by location within the back‐barrier setting, but it is limited to those El Niños that lead to barrier breakdown. For wider application of this particle size index of ENSO variability, it is important to establish a well‐resolved chronology and to sample the record at the appropriate interval to characterize deposition at a sub‐annual scale. Further, the sample site must be selected to limit the influence of decreasing accommodation space through time (infilling) and event‐based deposition. It is concluded that particle‐size data from back‐barrier sediment records have proven potential for preserving evidence of sub‐decadal climate variability, allowing researchers to explore temporal and spatial patterns in phenomena such as ENSO. Copyright © 2017 John Wiley & Sons, Ltd.     

沂沭断裂带重力场及地壳结构特征

沂沭断裂带为郯庐断裂带山东段,新构造运动显著,是华北地区的强震活动带之一。文中收集了该地区的布格重力数据,利用小波多尺度分析方法对重力场进行有效分离,研究区域地壳结构特征及断裂空间展布,并应用Parker变密度模型对区域莫霍面进行反演分析,得到以下几点结论:1)重力区域场显示,沂沭断裂带形成了NNE走向的大型重力梯度带,分隔了鲁西、鲁东地块,成为区域内重要的地球物理分界线。2)重力局部场显示,中上地壳结构复杂,沂沭带内部呈现两堑一垒的重力异常格局,5条主干断裂形成线性梯度带分布于东、西地堑内,鲁西块体的多条NW向活动断裂交切于沂沭断裂带,多数断裂只交切于西地堑,而蒙山山前断裂和苍尼断裂横穿沂沭断裂带;下地壳结构相对简单,发生明显的褶曲构造,表现出大规模高、低密度异常相间排列的典型特征。3)区域莫霍面形态东高西低,沂沭断裂带形成了莫霍面陡变带,造成了东西分异格局,潍坊东—莒县—临沂一线出现莫霍面上隆区,具有强震发生的深部孕震环境。4)区域内地震多发于高、低重力异常转化带之间,特别是活动断裂对应的重力梯度条带之上,地震的发生与断裂活动有着密切的关系,沂沭断裂带地震活动性最强,且东地堑强于西地堑。     

地电场信号去噪小波函数的选取

介绍了小波函数的基本性质,并根据密山地电场数据的特征选取适合密山地电场数据去噪的的小波族函数,最后综合去噪结果和误差分析结果选出适合密山地电场数据去噪的最优小波函数和最佳分解层数。     

辽河西部凹陷欢喜岭地区沙河街组层序界面的识别

对辽河西部凹陷欢喜岭地区层序地层的研究,总结出4种层序界面的识别方法:①钻井、测井资料;②泥砂比曲线;③小波变换;④地震时频分析。方法①表现为在沙河街组各三级层序界面处,岩性、相序、曲线形态和幅值有不同程度的突变;方法②表现为在一个完整的三级层序中,泥砂比曲线由小变大再变小,其中泥砂比值最大处为最大湖泛面(MFS)位置;方法③表现为在一个完整的三级层序中,谱系图从低频到高频再到低频的变化,在系数曲线上表现为一个或近似一个完整的振荡旋回,其中谱系图中高频部分对应着最大湖泛面位置;方法④在剖面上层序界面处为低频部分,高频部分是最大湖泛面位置。通过测井与地震信息相结合,综合运用4种方法识别出7个层序界面和6个最大湖泛面,由此建立了研究区的高精度层序地层格架。     

基于遗传算法优化的小波神经网络在地下水位预测中的应用

地下水位是衡量生态环境优劣和地下水资源的一个重要指标.地下水位下降,将引发地面沉降、地面塌陷和降落漏斗等.因此,地下水位预测对保护地质生态环境和实现地下水资源严格管理至关重要.由于BP算法存在极易收敛于局部极小点与过拟合等缺点,导致网络泛化能力不足,本文在构建小波神经网络基础上并引入遗传算法加以优化,以解决上述不足,并与BP和WNN对比预测了天津市深层承压水水位.预测结果表明,GA-WNN模型拟合精度较高,模型的预测能力有较大幅度提高.     

小波变换压制噪声在单道地震资料处理中的应用

海洋区域地质调查中,单道地震勘探以其高分辨率、方便快捷等优点成为揭示浅地层地质构造的重要手段之一.结合前人研究成果并出于保护信号的完整性,对低频部分的噪声采用带通滤波进行滤除,而对高频信号中的随机噪声采用小波阈值去噪方法进行压制处理.在小波阈值去噪过程中,通过比较选用最佳小波基函数sym5对地震剖面进行三层小波分解.结果表明:两种阈值去噪方法在带通滤波的基础上进一步提升了剖面的质量,使得反射层面变得连续可追踪.软阈值去噪滤掉了剖面上的较多信息,去噪效果较为彻底,硬阈值去噪则保留了较多的细节信息.     

Separation of obliquely incident and reflected irregular waves by the Morlet wavelet transform

An existing 2D time-domain method for separating irregular incident and reflected waves by wavelet transform [Ma et al., 2010. A new method for separation of 2D incident and reflected waves by the Morlet wavelet transform. Coastal Eng., 57(6):597–603] is extended to account for obliquely incident irregular waves propagating over sloping bottoms. The linear shoaling and refraction coefficients are adopted to determine the amplitude and phase changes of waves. The optimal central frequency of the Morlet wavelet is determined by the minimum Shannon wavelet entropy. Numerical tests show that the present method can accurately separate waves over horizontal depths. For waves at sloping bottoms, however, the separation errors increase as bottom slope increases and are significant for waves with incident angle larger than π/3.     

A novel technique in analyzing non-linear wave-wave interaction

During wave growth non-linear wave–wave interactions cause transfer of some wave energy from lower to higher wave periods as the spectrum grows. Wavelet bicoherence, which is a new technique in the analysis of wind–wave and wave–wave interactions, is used to analyze non-linear wave–wave interactions. A selected record of wind wave that contains the maximum wave height observed during 6 h of wave generation is divided into five segments and wavelet bicoherence is computed for the whole record, and for all divided segments. The study shows that the non-linear wave–wave interaction occurs at different bicoherence levels and these levels are different from one segment to another due to the non-stationarity feature of the examined data set.     

一种基于小波熵的时间尺度算法

结合小波分析和熵的基本理论,在分析国际原子时TAI的基础上,提出了基于小波熵的时间尺度算法．通过对原子钟信号进行多尺度分解,根据能量划分,综合所有尺度的小波系数得到小波熵,然后利用小波熵对原子钟进行加权．最后,通过某实验室的实测数据对算法进行了检验,结果表明:这种方法能够综合各原子钟信号在不同尺度上的频率波动特征,明显提高了时间尺度的稳定度．     

使用小波技术自动搜寻天体谱线

使用小波技术对包括恒星、近邻星系和ＡＧＮ等不同的天体光谱进行了自动处理．用小波滤波的方法将光谱中的连续谱与诸线分离;然后使用小波域隐含马尔可夫模型（ＨＭＭ）,对已去掉连续谱的光谱进行降噪,同时得到了噪声在每个光谱中的分布;在训练ＨＭＭ的过程中,使用改进的Ｔｙｉｎｇ方法增加训练数据以保证训练的可靠性;利用噪声分布确定出谱线信号的局部阈值,在已经降噪的光谱中找到吸收线和发射线;用高斯函数拟会出谱线的形状,标出线心的波长值,作为自动证认的基础．