PAPER SUMMARIES

Paper Summaries

An important task of modern Geostatistical modeling to the uncertainty of the geological systems requires quantification. Various methods or models are presented in this research article. It is integral to note that in order to quantify, “Geostatistical Model” requires lots of input or input parameters. The input univariate distribution histogram is perhaps the most important with this respect. Another method for estimating the uncertainty in the histogram, including the uncertainty in the mean is known as the “Contingent Finite Domain” (DFP). This requires the size of the local data, and air conditioning. Apart from this, there is a “Stochastic Model” based on the multivariate Gaussian distribution.

CFD is another statistical approach to parameter uncertainty that relates directly with climate data and the size of the area. This method is the first known direct approach, which includes two important factors. The CFD is a stochastic method based on multivariate Gaussian model, with the restriction of the (expected) uncertainty in the statistics of interest to determine. The proposed method converges, independent design and parameterization of all languages.

The CFD is shown as the convergent, independent design and parameterization of languages. Performance CFD approach is illustrated by the example focuses on the effect on the amount of data, and the limit of the rank correlation of uncertainty in the parameters. Within this paper, “Spatial Load Method” and “CFD” approach is compared. The number of data increases, the uncertainty in the sample mean decreases in both the spatial and CFD download. Unlike the luggage, the uncertainty in the half of the samples in the CFD approach decreases with increasing distance correlations. This is a direct result of fitting the data which can be associated with the place, not selected in the final result. Furthermore, the uncertainty about the variogram limits and offset limits are also discussed at length.

The bootstrap re-sampling procedure is widely used to quantify the uncertainty in the statistical parameters. The two most important assumptions include the initial distribution of the data which is representative of the total population and the data that are independent. This short note discussed here presents a boot program, which decreases with the correlation, and that relaxes the second assumption of independence. An LU simulation is used to model the values ??of the multivariate Gaussian. The program works very fast and is suitable for almost all data sets. The parameters of programs are described in depth, as well as a number of synthetic and real examples depict the location of the program. The program writes two output files which includes the simulated values and the bootstrap statistics.

The LU simulation approach is very fast when there are 100s of data. The bootstrap is generally considered to apply to homogeneous geological rocks that create uncertainty in the parameters such as mean, variance and distribution of the form. Hence the uncertainty in rock types and their sizes must be considered separately. It would be easy to simulate various values ??related to drawing sets with probability values that ...