Constant Dispersion Coefficient Research Articles

Investigation of nonlinear Schrödinger equation for application to astronomical redshift

Cyclic soliton solutions of nonlinear Schrödinger equation could be found numerically. When they propagate through a medium with piecewise constant dispersion coefficient, the pulse width change in each segment is linearly proportional, but at different rate, to distance travelled. The overall pulse width change over a distance between two points is found to be the same as if the average dispersion coefficient is used. Applying these findings to light from cosmic sources, it can be concluded that electromagnetic wave theory based on nonlinear Schrödinger equation is the principal mechanism underlining the linear redshift-distance relation found is astronomy.

Modeling of the Saltwater Intrusion Using the Level Set Method. Application to Henry’s Problem

The salt intrusion phenomenon is caused by overexploitation of aquifers in coastal areas. This physical phenomenon has been the subject of numerous studies and numerous methods have been proposed, with the aim of protecting the quality of the water in these aquifers. This work proposes a two-dimensional saline intrusion model using the sharp interface approach and the level set method. It consists of a parabolic equation modeling the underground flow and a hyperbolic Equation (the level set equation) which makes it possible to track the evolution of the interface. High-order numerical schemes such as the space scheme WENO5 and the third-order time scheme TVD-RK were used for the numerical resolution of the hyperbolic equation. To limit the tightening of the contour curves of the level set function, the redistanciation or reinitialization algorithm proposed by Sussma et al. (1994) was used. To ensure the effectiveness and reliability of the proposed method, two tests relating to the standard Henry problem and the modified Henry problem were performed. Recall that Henry’s problem uses the variable density modeling approach in a confined and homogeneous aquifer. By comparing the results obtained by the level set method with reinitialization (LSMR) and those obtained by Henry (1964), and by Simpson and Clement (2004), we see in the two test cases that the level set method reproduces well the toe, the tip and the behaviour of the interface. These results correspond to the results obtained by Abarca for Henry’s problem with constant dispersion coefficients. The results obtained with LSMR, reproduced the interface with a slight spacing compared to those obtained by Henry. According to Abarca (2006), this spacing is due to the absence of the longitudinal and transversal dispersion coefficients in the model.

Numerical and experimental study on solute transport through physical aquifer model

Abstract Environmental concerns have drawn much research interest in solute transport through porous media. Thus, contaminants of groundwater permeate through pores in the ground, and adsorption attenuates the pollution concentration as the pollutants adhere to the solid surface. Mathematical models based on certain simplifying assumptions have been used to predict solute transport. The transport of solutes in porous media is governed by a partial differential equation known as the advection-dispersion equation. In this study, a two-dimensional numerical model has been developed for solute transport through porous media. Results of spatial moments have been predicted and analysed in the presence of both constant and time-dependent dispersion coefficients. Afterward, a numerical model is used to simulate experimentally observed breakthrough curves for both conservative and non-conservative solutes. Thus, transport parameters are estimated through numerical simulation of observed breakthrough curves. Finally, this model gives the best simulation of observed breakthrough curves, and it can also be used in the field scale.

Generalized Analytical Solutions of The Advection-Dispersion Equation with Variable Flow and Transport Coefficients

Demand has increased for analytical solutions to determine the velocities and dispersion coefficients that describe solute transport with spatial, temporal, or spatiotemporal variations encountered in the field. However, few analytical solutions have considered spatially, temporally, or spatiotemporally dependent dispersion coefficients and velocities. The proposed solutions consider eight cases of dispersion coefficients and velocities: both spatially dependent, both spatiotemporally dependent, both temporally dependent, spatiotemporally dependent dispersion coefficient with spatially dependent velocity, temporally dependent dispersion coefficient with constant velocity, both constant, spatially dependent dispersion coefficient with spatiotemporally dependent velocity, and constant dispersion coefficient with temporally dependent velocity. The spatial dependence is linear, while the temporal dependence may be exponential, asymptotical, or sinusoidal. An advection–dispersion equation with these variable coefficients was reduced to a non-homogeneous diffusion equation using the pertinent coordinate transform method. Then, solutions were obtained in an infinite medium using Green’s function. The proposed analytical solutions were validated against existing analytical solutions or against numerical solutions when analytical solutions were unavailable. In this study, we showed that the proposed analytical solutions could be applied for various spatiotemporal patterns of both velocity and the dispersion coefficient, shedding light on feasibility of the proposed solution under highly transient flow in heterogeneous porous medium.

Rotary drums with sectional internals: Experimental investigation on the influence of section number and section length

The impact of sectional internals, as often used in industrial rotary drums, on the axial solid transport in rotary drums is investigated experimentally under variation of the operation parameters solid feed rate, drum inclination and rotational speed. Different section numbers and internal-lengths have been analyzed, applying the axial dispersion model, in a fully optically accessible lab scale rotary drum apparatus.The mean residence times are increasing with the section numbers and their length up to 4 times compared to bare drum results. Capacity limits have been detected, resulting in higher back-spillage ratios dependent on the set of operational parameters. The impacts of parameter changes on the axial dispersion are decreasing with section number and length, resulting in a more constant axial dispersion coefficient of about 1 × 10−6 m2/s. A qualitative similarity of the axial solid bed profile proposed by the Saeman-model for bare drums has been found for sectional internals.

Nonlocal transport models for capturing solute transport in one‐dimensional sand columns: Model review, applicability, limitations and improvement

AbstractModelling pollutant transport in water is one of the core tasks of computational hydrology, and various physical models including especially the widely used nonlocal transport models have been developed and applied in the last three decades. No studies, however, have been conducted to systematically assess the applicability, limitations and improvement of these nonlocal transport models. To fill this knowledge gap, this study reviewed, tested and improved the state‐of‐the‐art nonlocal transport models, including their physical background, mathematical formula and especially the capability to quantify conservative tracers moving in one‐dimensional sand columns, which represents perhaps the simplest real‐world application. Applications showed that, surprisingly, neither the popular time‐nonlocal transport models (including the multi‐rate mass transfer model, the continuous time random walk framework and the time fractional advection‐dispersion equation), nor the spatiotemporally nonlocal transport model (ST‐fADE) can accurately fit passive tracers moving through a 15‐m‐long heterogeneous sand column documented in literature, if a constant dispersion coefficient or dispersivity is used. This is because pollutant transport in heterogeneous media can be scale‐dependent (represented by a dispersion coefficient or dispersivity increasing with spatiotemporal scales), non‐Fickian (where plume variance increases nonlinearly in time) and/or pre‐asymptotic (with transition between non‐Fickian and Fickian transport). These different properties cannot be simultaneously and accurately modelled by any of the transport models reviewed by this study. To bypass this limitation, five possible corrections were proposed, and two of them were tested successfully, including a time fractional and space Hausdorff fractal model which minimizes the scale‐dependency of the dispersion coefficient in the non‐Euclidean space, and a two‐region time fractional advection‐dispersion equation which accounts for the spatial mixing of solute particles from different mobile domains. Therefore, more efforts are still needed to accurately model transport in non‐ideal porous media, and the five model corrections proposed by this study may shed light on these indispensable modelling efforts.

Solute transport model equation for mobile phase in semi-infinite porous media

An analytical solution of one-dimensional solute transport model equation for mobile phase is derived for semi-infinite heterogeneous porous media. The transport equation considers advection, longitudinal dispersion, linear isotherm, and porosity in its formulation. The dispersion is assumed to be Fickian, characterized by a constant dispersion coefficient. Temporally dependent velocity patterns are employed for the prediction of contaminant concentration. The analytical solution is compared with two reported solutions and can be used for benchmarking numerical codesand also as a preliminary predictive tool for groundwater management. Also, the obtained analytical solution of the proposed model may be helpful in the groundwater hydrology areas.

Assessment and Prediction of Pore‐Scale Reactive Mixing From Experimental Conservative Transport Data

AbstractWe investigate the impact of pore‐scale heterogeneity on reactive mixing using experimental data from inert fluid‐fluid displacement in a quasi 2‐D porous medium. We interpret the invading and defending fluids as the conservative components A+C and B+C of the instantaneous irreversible bimolecular chemical reaction A+B→C and determine the reaction product C. We find strong growth of the reaction rate at short and intermediate times due to deformation of the mixing interface and heterogeneity‐induced increase of the mixing area. This behavior is captured by a dispersive lamella approach that quantifies the mixing dynamics at the interface in terms of temporally evolving effective dispersion coefficients. The latter capture the dominant controls on the evolution of the mixing interface, namely, advective heterogeneity and transverse mixing. Reactive transport formulations based on constant hydrodynamic dispersion coefficients are not able to describe the observed behavior due to the lack of complete mixing on the support scale, which is required for these approaches to hold. These results shed some new light on the origins of heterogeneity‐induced mixing and reaction dynamics in porous media and their systematic upscaling.

Effect of Velocity and Diffusion Functionality on Nonlinear Mass Transfer Mechanisms in Solvent Oil Recovery

Literature has indicated that, experimentally, solvent fronts in hybrid solvent recovery processes progress more rapidly than what can be predicted using current approximations and more rapidly than thermal processes alone. Research using finite differences to model the nonlinear advection, diffusion and dispersion (ADD) equation suggests that nonlinear mass transfer effects are important in predicting the rate of solvent advance. Nonlinearities can be ascribed to both diffusion and flow velocity functionality. Earlier work using linear concentration dependent diffusion and log-linear velocity behaviour confirmed the importance of nonlinear effects when compared to linear theory that uses constant diffusion, dispersion and velocity coefficients. The mathematical nature of the nonlinear ADD equation further suggests that the shape of concentration dependent diffusion and flow velocity will affect the shape of the solvent concentration profiles, and influence the rate of propagation of the solvent front. This research focuses on results obtained using finite differences to explore the effects of various diffusion and velocity functionalities that affect the solvent rate propagation using a nonlinear ADD equation. The results obtained from this analysis indicate that these functionalities determine the shape of the solvent concentration profile. The concentration dependent diffusion and velocity functions were chosen according to recent literature which proposes experimentally obtained functions to more accurately model solvent penetration in the media. Preliminary results from this study suggest that the velocity functionality has more influence on the process at both the lab and field scales for the parameters considered in this study. The shapes of the concentration profiles are affected by both diffusion functionality and velocity functionality.

Stable Lévy diffusion and related model fitting

A fractional advection-dispersion equation (fADE) has been advocated for heavy-tailed flows where the usual Brownian diffusion models fail. A stochastic differential equation (SDE) driven by a stable L\'{e}vy process gives a forward equation that matches the space-fractional advection-dispersion equation and thus gives the stochastic framework of particle tracking for heavy-tailed flows. For constant advection and dispersion coefficient functions, the solution to such SDE itself is a stable process and can be derived easily by least square parameter fitting from the observed flow concentration data. However, in a more generalized scenario, a closed form for the solution to a stable SDE may not exist. We propose a numerical method for solving/generating a stable SDE in a general set-up. The method incorporates a discretized finite volume scheme with the characteristic line to solve the fADE or the forward equation for the Markov process that solves the stable SDE. Then we use a numerical scheme to generate the solution to the governing SDE using the fADE solution. Also, often the functional form of the advection or dispersion coefficients are not known for a given plume concentration data to start with. We use a Levenberg--Marquardt (L-M) regularization method to estimate advection and dispersion coefficient function from the observed data (we present the case for a linear advection) and proceed with the SDE solution construction described above.

Computational Fluid Dynamics Simulations of Gas-Phase Radial Dispersion in Fixed Beds with Wall Effects

The effective medium approach to radial fixed bed dispersion models, in which radial dispersion of mass is superimposed on axial plug flow, is based on a constant effective dispersion coefficient, DT. For packed beds of a small tube-to-particle diameter ratio (N), the experimentally-observed decrease in this parameter near the tube wall is accounted for by a lumped resistance located at the tube wall, the wall mass transfer coefficient km. This work presents validated computational fluid dynamics (CFD) simulations to obtain detailed radial velocity and concentration profiles for eight different computer-generated packed tubes of spheres in the range 5.04 ≤ N ≤ 9.3 and over a range of flow rates 87 ≤ Re ≤ 870 where Re is based on superficial velocity and the particle diameter dp. Initial runs with pure air gave axial velocity profiles vz(r) averaged over the length of the packing. Then, simulations with the tube wall coated with methane yielded radial concentration profiles. A model with only DT could not describe the radial concentration profiles. The two-parameter model with DT and km agreed better with the bed-center concentration profiles, but not with the sharp decreases in concentration close to the tube wall. A three-parameter model based on classical two-layer mixing length theory, with a wall-function for the decrease in transverse radial convective transport in the near-wall region, showed greatly improved ability to reproduce the near-wall concentration profiles.

Can a Time Fractional-Derivative Model Capture Scale-Dependent Dispersion in Saturated Soils?

Time nonlocal transport models such as the time fractional advection-dispersion equation (t-fADE) were proposed to capture well-documented non-Fickian dynamics for conservative solutes transport in heterogeneous media, with the underlying assumption that the time nonlocality (which means that the current concentration change is affected by previous concentration load) embedded in the physical models can release the effective dispersion coefficient from scale dependency. This assumption, however, has never been systematically examined using real data. This study fills this historical knowledge gap by capturing non-Fickian transport (likely due to solute retention) documented in the literature (Huang et al. 1995) and observed in our laboratory from small to intermediate spatial scale using the promising, tempered t-fADE model. Fitting exercises show that the effective dispersion coefficient in the t-fADE, although differing subtly from the dispersion coefficient in the standard advection-dispersion equation, increases nonlinearly with the travel distance (varying from 0.5 to 12 m) for both heterogeneous and macroscopically homogeneous sand columns. Further analysis reveals that, while solute retention in relatively immobile zones can be efficiently captured by the time nonlocal parameters in the t-fADE, the motion-independent solute movement in the mobile zone is affected by the spatial evolution of local velocities in the host medium, resulting in a scale-dependent dispersion coefficient. The same result may be found for the other standard time nonlocal transport models that separate solute retention and jumps (i.e., displacement). Therefore, the t-fADE with a constant dispersion coefficient cannot capture scale-dependent dispersion in saturated porous media, challenging the application for stochastic hydrogeology methods in quantifying real-world, preasymptotic transport. Hence improvements on time nonlocal models using, for example, the novel subordination approach are necessary to incorporate the spatial evolution of local velocities without adding cumbersome parameters.

Influence of Various Dispersion Coefficients on Contaminant Migration in a Fracture-matrix System with Skin Formation

Subsurface contamination due to transient migration of chemicals through weathered fractures is a serious problem in hard rock terrains. In this paper, a numerical model has been used to analyse the effect of various dispersion coefficients on the movement of contaminants in a fracture-matrix syste m with skin formation. A constant continuous source of contaminants is considered at the inlet of a set of parallel fractures and an implicit finite difference technique has been used to solve the model. The transport of contaminants with constant dispersion coefficients were compared with those subjected to distance and time dependent dispersion. Sensitivity analysis has been conducted to study the effect of different fluid velocities, fracture-skin porosities, fracture-skin diffusion coefficients, fracture aperture, retardation factors, and dispersivity-distance ratio on solute transport mechanism within the system. Results suggest that the solute mass retained in the fracture is higher when time-dependent dispersion coefficients are used which may be due to seasonal conditioning of the rocks. When constant dispersion coefficient is considered, the reduction in the solute mass with increment in the fracture-skin porosity is gradual unlike other cases.The solute mass retained in the fracture increases with increase in dispersivity-distance ratio for time-dependent dispersion coefficients.

A comparison of Eulerian and Lagrangian transport and non-linear reaction algorithms

When laboratory-measured chemical reaction rates are used in simulations at the field-scale, the models typically overpredict the apparent reaction rates. The discrepancy is primarily due to poorer mixing of chemically distinct waters at the larger scale. As a result, realistic field-scale predictions require accurate simulation of the degree of mixing between fluids. The Lagrangian particle-tracking (PT) method is a now-standard way to simulate the transport of conservative or sorbing solutes. The method’s main advantage is the absence of numerical dispersion (and its artificial mixing) when simulating advection. New algorithms allow particles of different species to interact in nonlinear (e.g., bimolecular) reactions. Therefore, the PT methods hold a promise of more accurate field-scale simulation of reactive transport because they eliminate the masking effects of spurious mixing due to advection errors inherent in grid-based methods. A hypothetical field-scale reaction scenario is constructed and run in PT and Eulerian (finite-volume/finite-difference) simulators. Grid-based advection schemes considered here include 1st- to 3rd-order spatially accurate total-variation-diminishing flux-limiting schemes, both of which are widely used in current transport/reaction codes. A homogeneous velocity field in which the Courant number is everywhere unity, so that the chosen Eulerian methods incur no error when simulating advection, shows that both the Eulerian and PT methods can achieve convergence in the L1 (integrated concentration) norm, but neither shows stricter pointwise convergence. In this specific case with a constant dispersion coefficient and bimolecular reaction A+B→P, the correct total amount of product is 0.221MA0, where MA0 is the original mass of reactant A. When the Courant number drops, the grid-based simulations can show remarkable errors due to spurious over- and under-mixing. In a heterogeneous velocity field (keeping the same constant and isotropic dispersion), the PT simulations show an increased reaction total from 0.221MA0 to 0.372MA0 due to fluid deformation, while the 1st-order Eulerian simulations using ≈ 106 cells (with a classical grid Peclet number Δx/αL of 10) have total product of 0.53MA0, or approximately twice as much additional reaction due to advection error. The 3rd-order TVD algorithm fares better, with total product of 0.394MA0, or about 1.14 times the increased reaction total. A very strict requirement on grid Peclet numbers for Eulerian simulations will be required for realistic reactions because of their nonlinear nature. We analytically estimate the magnitude of the effect for the end-member cases of very fast and very slow reactions and show that in either case, the mass produced is proportional to 1/Pe, where Pe is the Peclet number. Therefore, extra mass is produced according to D, where the dispersion includes any numerical dispersion error. We test two PT methods, one that kills particles upon reaction and another that decrements a particle’s mass. For the bimolecular reaction studied here, the computational demands of the particle-killing methods are much smaller than, and the particle-number-preserving algorithm are on par with, the fastest Eulerian methods.

Using the Fractional Advection Dispersion Equation to Improve the Scale Effect in the Sampling Process of Column Tests with Lateritic Soils

Breakthrough curves (BTCs) obtained from column tests in heterogeneous soils are not satisfactorily simulated with the advection-dispersion equation (ADE) for some heavy tailed cases. Furthermore, the dispersion coefficient calculated with the ADE for heavy tailed BTCs are scale dependent when simulating columns of soil larger than the original test depth. In this paper we compare the usage of a fractional ADE (FADE) and the classical ADE to fit column tests BTCs made with Brazilian lateritic soils, discussing both contaminant transport theories and underlying stochastic models. The FADE can more accurately simulate heavy tailed BTCs, and when applying the adjusted FADE parameters to longer depths of soil, the FADE also predicts a more realistic scenario of contaminant transport through heterogeneous soil. The addition of fractional calculus in the advection-dispersion equation proves to improve contaminant transport predictions based on column tests over the classical ADE, with the use of a constant fractional dispersion coefficient that is scale independent.

Effect of distance-dependent and time-dependent dispersion on non-linearly sorbed multispecies contaminants in porous media

In this study, the effect of distance-dependent and time-dependent dispersion coefficient on the transport of multispecies contaminants subjected to Freundlich non-linear adsorption in porous media has been analyzed. An implicit finite difference has been used to develop the numerical model. A constant continuous source of first species contaminant has been assumed at the inlet of the porous domain. The results obtained from distance-dependent and time-dependent dispersion coefficient has been compared with that obtained using constant dispersion coefficient for different fluid velocities and dispersivity–distance ratios. The first species of the multispecies transport is subjected to Freundlich non-linear sorption. Sensitivity analysis has been conducted to study the effect of different simulation time, fluid velocities, sorption exponent, and dispersivity–distance ratio on solute transport mechanism within the porous media. Results suggest that the effect of distance-dependent dispersion coefficient on multispecies contaminant movement is negligible when the fluid velocity is very low. The fluid velocity has a significant impact on the multispecies solute movement when non-linear sorption and distance-dependent dispersion coefficient are considered. The effect of variation of the Freundlich sorption exponent on contaminant concentration is negligible when distance-dependent and time-dependent dispersion coefficients are considered.

Kinematics of age mixing in advection-dispersion models

This paper investigates age mixing processes arising in advection-dispersion models, where large-scale travel and residence time distributions can be explicitly calculated based on the underlying velocity field. In particular, we analyze spatially integrated age mixing dynamics by comparing the age distributions of the storage and of the outflow(s). The relevance of the work lies in the impact of age mixing dynamics on the shape of travel time distributions (TTDs), which ultimately control the long-term memory of catchment transport processes. We set up a theoretical framework that bridges previous Lagrangian and Eulerian water age theories in heterogeneous media. The framework allows for the analysis of the dynamical connection between water age distributions in large-scale volume and flux samples. Theoretical advances are then illustrated through the application to a finite one-dimensional domain with constant advection and dispersion coefficient. Therein, we analyze the type of mixing emerging for different Peclet numbers and diverse spatiotemporal patterns of solute input. We find that in spite of the enhanced nonstationarity of TTDs, the type of mixing is markedly invariant. Moreover, for relatively low Peclet numbers the different ages available in the control volume are systematically removed from the domain at a rate nearly proportional to their relative abundance (random sampling). Emerging large-scale patterns of age mixing yield theoretical and practical implications for watershed hydrology, where TTDs can be used to infer general patterns of catchment response across scales.

A new fractional finite volume method for solving the fractional diffusion equation

The inherent heterogeneities of many geophysical systems often gives rise to fast and slow pathways to water and chemical movement. One approach to model solute transport through such media is by fractional diffusion equations with a space–time dependent variable coefficient. In this paper, a two-sided space fractional diffusion model with a space–time dependent variable coefficient and a nonlinear source term subject to zero Dirichlet boundary conditions is considered.Some finite volume methods to solve a fractional differential equation with a constant dispersion coefficient have been proposed. The spatial discretisation employs fractionally-shifted Grünwald formulas to discretise the Riemann–Liouville fractional derivatives at control volume faces in terms of function values at the nodes. However, these finite volume methods have not been extended to two-dimensional and three-dimensional problems in a natural manner. In this paper, a new weighted fractional finite volume method with a nonlocal operator (using nodal basis functions) for solving this two-sided space fractional diffusion equation is proposed. Some numerical results for the Crank–Nicholson fractional finite volume method are given to show the stability, consistency and convergence of our computational approach. This novel simulation technique provides excellent tools for practical problems even when a complex transition zone is involved. This technique can be extend to two-dimensional and three-dimensional problems with complex regions.

A joint velocity‐concentration PDF method for tracer flow in heterogeneous porous media

The probability density function (PDF) of the local concentration of a contaminant, or tracer, is an important component of risk assessment in applications that involve flow in heterogeneous subsurface formations. In this paper, a novel joint velocity‐concentration PDF method for tracer flow in highly heterogeneous porous media is introduced. The PDF formalism accounts for advective transport, pore‐scale dispersion (PSD), and molecular diffusion. Low‐order approximations (LOAs), which are usually obtained using a perturbation expansion, typically lead to Gaussian one‐point velocity PDFs. Moreover, LOAs provide reasonable approximations for small log conductivity variances (i.e., σY2 < 1). For large σY2, however, the one‐point velocity PDFs deviate significantly from the Gaussian distribution as demonstrated convincingly by several Monte Carlo (MC) simulation studies. Furthermore, the Lagrangian velocity statistics exhibit complex correlations that span a wide range of scales, including long‐range correlations due to the formation of preferential flow paths. Both non‐Gaussian PDFs and complex long‐range correlations are accurately represented using Markovian velocity processes (MVPs) in the proposed joint PDF method. LOA methods can be generalized to some extent by presuming a certain shape for the concentration PDF (e.g., a β PDF fully characterized by the concentration mean and variance). The joint velocity‐concentration PDF method proposed here does not require any closure assumptions on the shape of the marginal concentration PDF. The Eulerian joint PDF transport equation is solved numerically using a computationally efficient particle‐based approach. The PDF method is validated with high‐resolution MC reference data from Caroni and Fiorotto (2005) for saturated transport in velocity fields, which are stationary in space and time, for domains with σY2 = 0.05, 1, and 2 and Péclet numbers ranging from 100 to 10,000. PSD is modeled using constant anisotropic dispersion coefficients in both the reference MC computations and our PDF method.

The backward group preserving scheme for 1D backward in time advection‐dispersion equation

AbstractIn this article, we propose a backward group preserving scheme (BGPS) on the advection‐dispersion equation (ADE) for tackling of the contamination problems. The BGPS has been successfully applied on the backward heat conduction problems as well as the backward in time Burgers equation, but it has never been applied to solve the ADE. The BGPS is able to recover the spatial distribution of groundwater contaminant concentration in this work. Several numerical examples are worked out, and we show, based on those numerical examples, that the BGPS is applicable to the ADE and the method can also handle the ADE with piecewise constant dispersion coefficients. When a steep gradient is appeared in the solution, several steps of the BGPS can be used to retrieve the desired initial data and its result is better than the marching‐jury backward beam equation (MJBBE) method as far as our examples are concerned. © 2008 Wiley Periodicals, Inc. Numer Methods Partial Differential Eq, 2010