1,721,017 research outputs found
Identification of the unknown pollution source in the Alsatian aquifer (France) through groundwater modelling and Artificial Neural Networks applications
No full textANN-based approach for the estimation aquifer pollutant source behaviour
No full textThe problem of identifying an unknown pollution source in polluted aquifers, based on known contaminant concentrations measurement, is part of the broader group of issues, called inverse problems. This paper investigates the feasibility of solving the groundwater pollution inverse problem by using artificial neural networks (ANNs). The approach consists first in training an ANN to solve the direct problem, where the pollutant concentration in a set of monitoring wells is calculated for a known pollutant source. Successively, the trained ANN is frozen and it is used to solve the inverse problem, where the pollutant source is calculated which corresponds to a set of concentrations in the monitoring wells. The approach has been applied for a real case which deals with the contamination of the Rhine aquifer by carbon tetrachloride (CCl4) due to a tanker accident. The obtained results are compared with the solution obtained with a different approach retrieved from literature. The results show the suitability of ANNs-based methods for solving inverse non-linear problems
Statistical Description of Calcite Surface Roughness Resulting from Dissolution at Close-to-Equilibrium Conditions
Full text linkLinking the evolution of the surface area (as quantified, e.g., through its spatial roughness) of minerals to their dissolution rate is a key aspect of mineral reactivity. Unraveling the nature of their main features requires relying on approaches yielding a quantitative characterization of the temporal evolution of surface topography/roughness. Here, a mechanically polished {104} calcite surface was dissolved at room temperature and at close-to-equilibrium conditions (ω = 0.6) with an alkaline solution (pH = 8) across a temporal window of 8 days. Surface topography images were acquired daily using vertical scanning interferometry, the ensuing topography data being then embedded within a statistical analysis framework aimed at describing comprehensively the surface roughness evolution. The strongest system variations were observed after 1 day: the probability density function of surface roughness was observed to transition from being approximately Gaussian to being left-skewed and leptokurtic, exhibiting a dramatic increase in the variance and a significant change in the semi-variogram structure. After a relaxation time of approximately 2 days, the reacting surface appeared to attain a steady-state configuration, being characterized by the values of the statistical moments characterizing surface roughness that become virtually independent of time. Attempting to unravel the underlying dissolution mechanism, an original numerical model able to reproduce satisfactorily the statistical behavior observed experimentally was developed and tested. Our results suggest that under the investigated conditions, dissolution may be characterized as a spatially correlated random process, with the areas most exposed to the flowing fluid being prone to preferential dissolution. The numerical model was also used to obtain insights into the influences of the initial surface roughness and of the fluid composition on the steady-state statistical characterization of the surface roughness. Our results suggest that the influence of the initial surface roughness is limited. The present study suggests that potential empirical relations linking the surface roughness of the reacted crystals to the saturation state at which they dissolved may be developed, which would allow to back-estimate the reacting conditions only based on topography data
Solute transport in random composite media with uncertain dispersivities
Full text linkCharacterization of dissolved chemical migration in porous media requires knowledge of the fluid velocity field and parameters governing solute dispersion within the diverse geomaterials constituting the internal architecture of the system. Several studies have been focused on the assessment of the impact on solute concentrations of an incomplete knowledge of the fluid velocity field, typically a result of the effects of uncertain hydraulic properties of the hosting media (e.g., permeability). Limited attention has been devoted to analyze propagation of the uncertainty associated with spatial distributions of local dispersivity values to solute concentration fields. Here, we address this issue by focusing on a random composite medium, where the location of the boundary between two distinct geomaterials is uncertain as well as their associated dispersivity values. We derive and solve the equations satisfied by the (ensemble) mean and variance of solute concentration and investigate the relative impact on these moments of the two sources of uncertainty considered. Our results suggest that, in the investigated set-up, the temporal and spatial evolution of ensemble moments of the solute concentration depends on (i) the overall dispersive length scales encompassed by the solute during its migration and (ii) the actual sequence of the materials traversed by the solute
Polluted aquifer inverse problem solution using artificial neural networks
No full textThe problem of identifying an unknown pollution source
in polluted aquifers, based on known contaminant concentrations
measurement in the studied areas, is part of the broader group of issues,
called inverse problems. This paper investigates the feasibility
of using Artificial Neural Networks (ANNs) for solving the inverse
problem of locating in time and space the source of a contamination
event in a homogeneous and isotropic two dimensional domain.
ANNs are trained in order to implement an input-output relationship
which associates the position. Once the output of the system is
known, the input is reconstructed by inverting the trained ANNs.
The approach is applied for studying a theoretical test case where the
inverse problem is solved on the basis of measurements of contaminant
concentrations in monitoring wells located in the studied area.
Groundwater pollution sources are characterized by varying spatial
location and duration of activity. To identify these unknown pollution
sources, concentration measurements data of monitoring wells are used. If concentration observations are missing over a length of
time after an unknown source has become active, it is more difficult
to correctly identify the unknown pollution source. In this work, a
missing data scenario has been taken into consideration. In particular,
a case where only one measurement has been made after the pollutant
source interrupted its activity has been considered
Ann based approach to solve groundwater pollution inverse problem
No full textThis paper investigates the feasibility of solving the groundwater pollution inverse problem by using Artificial Neural Networks (ANNs). Different ANNs have been trained to solve the direct problem with the objective of associating the input patterns with the output patterns. In order to solve the inverse problem and to identify the unknown pollution source and their characteristics, the trained ANN is inverted. By fixing the output pattern of the ANN, the proposed procedure is able to reconstruct the corresponding input. The approach has been applied for a real case which deals with the contamination of the Rhine aquifer by carbon tetrachloride (CCl4) due to a tanker accident. This case is well adapted to the problem since numerous concentrations have been measured at different piezometers and at different time. The location of the source and the beginning of the contamination are known. The ANNs are used to identify the contamination source and the results are compared with the solution obtained with a different approach
Grid convergence for numerical solutions of stochastic moment equations of groundwater flow
Full text linkWe provide qualitative and quantitative assessment of the results of a grid convergence study in terms of (a) the rate/order of convergence and (b) the grid convergence index, GCI, associated with the numerical solutions of moment equations (MEs) of steady-state groundwater flow. The latter are approximated at second order (in terms of the standard deviation of the natural logarithm, Y, of hydraulic conductivity). We consider (1) the analytical solutions of Riva et al. (Transp Porous Med 45(1):139–193, 2001) for steady-state radial flow in a randomly heterogeneous conductivity field, which we take as references; and (2) the numerical solutions of the MEs satisfied by the (ensemble) mean and (co)variance of hydraulic head and fluxes. Based on 45 numerical grids associated with differing degrees of discretization, we find a supra-linear rate of convergence for the mean and (co)variance of hydraulic head and for the variance of the transverse component of fluxes, the variance of radial fluxes being characterized by a sub-linear convergence rate. Our estimated values of GCI suggest that an accurate computation of mean and (co)variance of head and fluxes requires a space discretization comprising at least 8 grid elements per correlation length of Y, an even finer discretization being required for an accurate representation of the second-order component of mean heads
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
