The behavior of the solute velocity and effective macrodispersivity of solute front in the fracture for the transport in a single fracture in the presence of rock matrix diffusion is analyzed using numerical modeling. The study is limited to a constant continuous solute source boundary condition in a single fracture with constant aperture. Analysis is made for linear and nonlinear sorption cases. Expressions are provided for solute velocity and effective macrodispersivity of the solute fronts during asymptotic and preasymptotic stages using spatial moment analyses. The effective retardation factor and dispersivity of solutes in the fracture are found to follow an exponential function with an exponent related to fracture-matrix parameters.

This paper presents a new hybrid methodology involving kriging and artificial neural network for ore grade estimation of two variables namely, Al₂O₃% and SiO₂%, in a bauxite deposit. The dataset was divided into three statistically similar subsets: training, calibration and validation sets using a genetic algorithm. The proposed hybrid ensemble model was formed using a kriging model and several neural network models. The outputs of these component models were combined using two methods to produce a unified prediction. The results indicated that the hybrid model was not a better estimator than the kriging model for the variable Al₂O₃%. However, it provides slightly better performance in comparison to any of the other component models in the ensemble for the variable SiO₂%.

In order to obtain meaningful predictions of contaminant transport, an accurate way of quantifying dispersivity needs to be developed. Results from the theoretical studies suggest that dispersion and the associated dispersivity is non-fickian near the source of contaminant and it grows with travel time and distance. In most tests of a limited duration it is quite probable that the asymptotic regime is not reached, and a proper interpretation of the test should be based on the time-dependent results due to the difficulty associated with the expensive experimental setups added to the marked scarcity of field data. An attempt has been made using spatial moment analysis to evaluate the time dependent dispersivity for a system of parallel fractures with matrix diffusion. The study is limited to non-reactive solutes, having a constant continuous source. An empirical relation to evaluate the dispersivity was developed by us based on the sensitivity analysis, when distinct parallel fractures have constant aperture width and is found to be functions of matrix porosity, matrix diffusion coefficient and injected fracture velocity at pre-asymptotic stage. The system becomes more complex when the aperture widths of the distinct parallel fractures are varied, as it appears that the initial development period of non-fickian behavior may be long due to the continuous lateral mixing of the solute body. It is found that dispersivity at pre-asymptotic regime increases with the coefficient of variation for distinct parallel fractures with varying aperture widths.