Change of Support for Estimating Local Block Grade Distributions

An important aspect in mineral resource evaluation is the reduction of variance when post-processing the grade distributions defined on the support (volume) of the available data into distributions defined on the support of the proposed selective mining units. Although the volume-variance relationship is well understood for the estimation of global grade distributions, it is still an unsolved issue for local estimation studies based on non-parametric geostatistical methods, such as indicator kriging, for which the support correction is not inherent to the method. To clarify this relationship, the local change of support problem is examined in the scope of two parametric models (multi-Gaussian and discrete Gaussian models). It is shown that the variance reduction factor between point and block-support local distributions depends on the block being considered and is less than the global variance reduction factor. As a consequence, post-processing the local point-support grade distributions on the basis of the latter systematically understates the importance of the change of support at the local scale and makes selective mining appear more economically attractive than it really is. In the light of these results, a methodology is proposed to post-process the local point-support distributions obtained via non-parametric (indicator) methods into block-support distributions. An application to simulated data indicates that this methodology provides an accurate estimation at the block support when dealing with diffusion-type random fields.     

Efficient generation of conditional simulations by chebyshev matrix polynomial approximations to the symmetric square root of the covariance matrix

Consider the problem of generating a realization y₁ of a Gaussian random field on a dense grid of points ₁ conditioned on field observations y₂ collected on a sparse grid of points ₂. An approach to this is to generate first an unconditional realization y over the grid =_{1 2}, and then to produce y₁ by conditioning y on the data y₂. As standard methods for generating y, such as the turning bands, spectral or Cholesky approaches can have various limitations, it has been proposed by M. W. Davis to generate realizations from a matrix polynomial approximations to the square root of the covariance matrix. In this paper we describe how to generate a direct approximation to the conditional realization y₁, on ₁ using a variant of Davis' approach based on approximation by Chebyshev polynomials. The resulting algorithm is simple to implement, numerically stable, and bounds on the approximation error are readily available. Furthermore we show that the conditional realization y₁ can be generated directly with a lower order polynomial than the unconditional realization y, and that further reductions can be achieved by exploiting a nugget effect if one is present. A pseudocode version of the algorithm is provided that can be implemented using the fast Fourier transform if the field is stationary and the grid ₁ is rectangular. Finally, numerical illustrations are given of the algorithm's performance in generating various 2-D realizations of conditional processes on large sampling grids.     

图像平滑算法比较研究及改进策略

图像平滑作为去除图像中含有噪声的图像增强处理技术，是各种与图像有关软件中必不可少的功能模块。文中在分析噪声模型的基础上，对均值平滑、中值平滑、边缘保持平滑等技术从图像处理空间域中的算法原理、实现方法及效率等方面进行了详细比较。针对图像中含有的不同噪声模型，通过多次试验，得出了值得从事图像处理研究者参考的结论。这将有助于提高图像处理软件开发的效率。     

Conditional curvilinear stochastic simulation using pixel-based algorithms

In geology, structures displaying differing local directions of continuity are widespread, a typical example being a flusial depositional system. Conventional pixel-based geostatistical algorithms, may fail to reproduce such curvilinear structures. Conversely, object-based algorithms can reproduce curvilinear shapes but are difficult to condition to dense local data. Local depositional directions as obtained from dipmeter data. 3D seismic data, and geological interpretation represent critical information. An improved pixel-based geostatistical algorithm is proposed to account for such directional information. Case studies demonstrate the potential and limitations of the algorithm.     

SEDIMENT TRANSPORT MODEL OF A WATERSHED

ChinaisoneOfthecountrieswhichsufferfromthemostserioussoilerosionintheworld,andthehillyandgullyareasof~PlateauinthemiddlereachesOftheYellowRiverarethemostseriouserosionareainChina.Inthisarea,themainerosivefactorsarerainfallandrunoff.Sometimes,thesedimentyieldinasinglefloodeventcanbe40%oftheannualsedimentload.Therefore,itisveryimPOrtanttostudytheyieldandgraphOfSedimentOfsinglefindevent.Forthispn~,thispaperiswritten.AInongsedimentmodelsforsinglefloodevent,moreandmoreattentionwaspaidtOunitse…     

Transformation of Residuals to Avoid Artifacts in Geostatistical Modelling with a Trend

Trend modelling is an important part of natural resource characterization. A common approach to account for a variable with a trend is to decompose it into a relatively smoothly varying trend and a more variable residual component. Then, the residuals are stochastically modelled independent of the trend. This decomposition can result in values outside the plausible range of variability, such as grades below zero or ratios that exceed 1.0. We transform the residuals conditional to the trend component to explicitly remove these complex features prior to geostatistical modelling. Back transformation of the modelled residual values allows the complex relations to be reproduced. A petroleum-related application shows the robustness of the proposed transformation. Furthermore, a mining application shows that when this conditional transformation is applied to the original variable, instead of the residual, simulated values are assured to be nonnegative.     

用L-曲线法确定半参数模型中的平滑因子

提出了一种新的方法——L-曲线法确定平滑因子。通过确定合适的平滑因子，更好地控制了残差部分V^TPV与光滑度部分S^TRS之间的平衡，得到了更准确的参数估值。通过算例，将基于L-曲线法确定平滑因子的半参数模型解算方法和其他方法进行了比较。结果表明，用L-曲线法确定平滑因子后，提高了半参数模型计算结果的精度，可以更好地将观测值中的系统误差分离出来。     

Histogram and variogram inference in the multigaussian model

Several iterative algorithms are proposed to improve the histogram and variogram inference in the framework of the multigaussian model. The starting point is the variogram obtained after a traditional normal score transform. The subsequent step consists in simulating many sets of gaussian values with this variogram at the data locations, so that the ranking of the original values is honored. The expected gaussian transformation and the expected variogram are computed by an averaging operation over the simulated datasets. The variogram model is then updated and the procedure is repeated until convergence. Such an iterative algorithm can adapt to the case of tied data and despike the histogram. Two additional issues are also examined, referred to the modeling of the empirical transformation function and to the optimal pair weighting when computing the sample variogram.     

Lame surfaces as a generalisation of the triaxial ellipsoid

In modern geodesy the triaxial ellipsoid as a generalisation of the ellipsoid of revolution has a significant position in studying the figure of the Earth. Lame surfaces represent a generalisation of the triaxial ellipsoid. The following paragraphs are devoted to curvatures of the Lame surfaces.     

Correcting the Smoothing Effect of Ordinary Kriging Estimates

The smoothing effect of ordinary kriging is a well-known dangerous effect associated with this estimation technique. Consequently kriging estimates do not reproduce both histogram and semivariogram model of sample data. A four-step procedure for correcting the smoothing effect of ordinary kriging estimates is shown to be efficient for the reproduction of histogram and semivariogram without loss of local accuracy. Furthermore, this procedure provides a unique map sharing both local and global accuracies. Ordinary kriging with a proper correction for smoothing effect can be revitalized as a reliable estimation method that allows a better use of the available information.