A Linear Programming Approach to Determine the Normative Composition of Sedimentary Rocks

The analysis of modal and normative composition of sedimentary rocks is widely used for studying their sources and tectonic settings. The normative calculation of the mineral composition of rocks in this study is formulated as a linear programming problem and is solved by means of the simplex method. This enables both simultaneous and successive subtraction of a set of basic minerals from a rock sample represented by its chemical composition {SiO₂...LOI}. Such an approach provides a more exact calculation of the contribution of basic minerals in the rock. This mathematical approach is used to study two representative sets of sandstones and fine-grained rocks from a Meso- to Neoproterozoic marginal basin of southeastern Siberia (Uchur–Maya region, Yakutia) and a Pennsylvanian-Lower Permian uplifted continental block in Colorado, USA. The calculated normative mineral compositions of the Siberian sandstones are consistent with the observed modal compositions. These sandstones vary from K- Feldspar rich arkoses at the base of the sequence (the Uchur Group, lower Riphean) to quartz arenites or lithic sandstones and wacke in transgressive successions of the middle-upper Riphean. Arkoses and quartz arenites are dominant in Meso- to Neoproterozoic Siberia. These samples represent craton interior uplifted basement and quartzose, recycled orogen provenance of a stable craton in Rodinia. There are higher but consistent discrepancies between the calculated and observed compositions for the Pennsylvanian to Lower-Permian arkoses and quartz arenites (Sangre de Cristo, Belden, and Maroon Formations). The differences between the predicted and observed mineralogy may be due to uncertainties in the modes in the matrix and cement of the sandstones. This normative program should supplement modal calculations and provide better genetic constructions, especially in case of matrix-rich sandstones.     

Smoothing and Change Point Detection for Gamma Ray Count Data

Gamma ray detectors are used to measure the natural radioactivity of rocks. For a number of boreholes drilled at a site the gamma ray detector is lowered into each borehole and counts of gamma ray emissions at different depths are recorded as the instrument is gradually raised to ground level. The profile of gamma counts can be informative about the geology at each location. The raw count data are highly variable, and in this paper we describe the use of adaptive smoothing techniques and change point models in order to identify changes in the geology based on the gamma logs. We formulate all our models for the data in the framework of the class of generalized linear models, and describe computational methods for Bayesian inference and model selection for generalized linear models that improve on existing techniques. Application is made to gamma ray data from the Castelreagh Waste Management Centre which served as a hazardous waste disposal facility for the Sydney region between March 1974 and August 1998. Understanding the geological structure of this site is important for further modelling the transport of pollutants beneath the waste disposal area.     

Layer parallel shortening in salt-detached folds: constraint on cross-section restoration

Cleavage-fissility perpendicular to bedding is a common feature in the external part of fold-and-thrust belts. Three techniques were used to determine the internal distortion in the frontal Southern Pyrenees: the analysis of strain markers such as burrows and rain drops, the measurement of fissility, and the measurement of anisotropy of magnetic susceptibility (AMS). The comparison of the three techniques showed a good fit although they differ in sensitivity to penetrative strain variations in the range of deformation values explored in the study case. On the regional scale, the values of layer parallel shortening (LPS) derived from the markers analysis are very constant and account for 16–23% of shortening. These values are two to three times larger than the shortening values calculated from the restoration of the macroscopic scale structures and indicate a good decoupling above the Cardona salt Formation. This study permitted an accurate restoration of the low-amplitude el Guix detached anticline.     

声波测试中振子衰减的线列阵模型

从声波在岩体中传播产生衰减的基本原理出发，建立了声波衰减振子的线列阵模型。该模型是将传播声波路径的岩体分为若干个振子，从波动概念出发，以振动原理作依据建立波动方程．为探讨岩体声波衰减机制提供了一种新的研究方法，把振动和波动直观地联系起来．单一方向上有限点的声波测试技术有可能被多方向的声测技术所代替．该模型的建立，使模态分析和参数识别方法应用到岩体声波测试技术中成为可能。     

非线性编辑与线性编辑在气象电视节目中的使用

通过对非线性编辑与线性编辑两者的概念及其特点、优势的介绍,指出在气象电视节目的后期制作中,怎样恰当地选择使用非线性编辑、线性编辑或二者同时使用。     

n维非线性波动方程对应的线性方程Cauchy问题解的一个估计式

对于地震波在地球介质中传播的传播非线性而言，在求解n维非线性波动方程Canchy问题时，需要求解其对应的线性波动方程Cauchy问题和建立一些解的估计式。我们已经求得了其对应的线性波动方程Canchy问题解的表达式。在此基础上，本文应用函数空间L^1(R^n)、L^∞（R）的范数，建立了n维非线性波动方程对应的线性波动方程Cancny问题解的一个估计式，为求解n维非线性波动方程莫定了基础。     

非线性模型线性近似的容许曲率

由于不同的非线性模型具有不同的非线性强度，使得一些非线性模型可以线性近似，而另一些则不能。本文介绍度量非线性强度的方法，提出判断非线性模型能否线性近似的数值标准——容许曲率     

线性综合模型用于崩滑体变形预报的研究

综合分析和比较了自回归模型和回归模型的特点，提出了采用线性综合模型来预报崩滑体变形的思想，以弥补自回归模型或回归模型预报的不足。实测资料的处理结果表明，综合线性模型具有特别的适用性。     

储液罐液—固—土耦合非线性翘离振动分析

本文考虑液固耦联及底板几何非线性与土壤弹性，建立了液－固－土耦合的储液罐翘离分析力学模型。算例表明，恢复力矩与翘离角之间的关系可以表达为分段线性，翘离将使地震响应增大。     

线性地质统计学在地质工作计算机化中的作用及意义

线性地质统计学的发展与地质工作逐渐由定性向定量、由人工劳动逐渐被计算机取代的过程密切相关的。对样品品位赋予一定的权值进行滑动加权平均，来估计中心块段平均品位的方法，使地质问题的计算能在计算机上实现开辟了新的途径。