New Math Model Tracks Toxic Metal Leaks From Landfills With Unprecedented Precision

Groundwater contamination from heavy metals is one of the most stubborn environmental problems of our time. Once lead, cadmium, or arsenic seeps into an aquifer, it can travel for decades, quietly poisoning wells and ecosystems far from the original source. Now, researchers Yuanning Li and Ruishi Liang have unveiled a new closed-form mathematical solution, published in Results in Engineering, that promises to make predicting this hidden migration faster, cheaper, and more physically faithful than ever before. Their work tackles a long-standing blind spot in contaminant modeling: how pollutants behave when they enter the ground as a controlled flux rather than as a fixed concentration, and how adsorption and desorption in the soil reshape the plume along the way.

Most existing models of pollutant transport rely on what mathematicians call Dirichlet boundary conditions. In plain terms, these assume that the concentration of contaminant at the source is known and imposed, as if a valve were holding the chemical at a constant level. That assumption is convenient for solving equations, but it rarely matches reality. Landfill liners leak at rates governed by the mass of waste and the permeability of the barrier. Industrial discharge zones release pollutants according to source strength and release kinetics, not according to a fixed concentration dial. When the real world controls the flux of contaminant entering the soil, a model that prescribes concentration instead can badly distort predictions, particularly near the source where regulators and engineers care most.

The new study addresses this gap by formulating the problem with a Neumann boundary condition, which specifies the rate at which contaminant mass flows into the soil rather than the concentration itself. Li and Liang derived an exact analytical solution for two-dimensional reactive transport in a homogeneous, isotropic porous medium, incorporating advection, hydrodynamic dispersion, linear equilibrium adsorption, first-order decay, and a finite-width line source. According to the authors, no previous closed-form solution had simultaneously combined a Neumann flux boundary, a two-dimensional line source, linear adsorption, and first-order decay. The derivation proceeds by transforming the two-dimensional advection-dispersion equation into an equivalent one-dimensional form and then applying Laplace transforms in both time and space, ultimately yielding a general solution expressed in terms of complementary error functions and source integrals.

What sets the formulation apart is its treatment of desorption. Many classical models treat adsorption as a simple equilibrium process with first-order decay, or handle desorption only as a function of time. The new model introduces directional desorption terms along both the longitudinal and transverse axes, capturing the fact that in real soils, the release of adsorbed metals back into pore water is coupled to concentration gradients and exhibits spatial directionality. Physically, this means that as a plume moves and its concentration field evolves, metals previously locked onto soil particles can be remobilized preferentially along the main transport directions, accelerating migration in ways that simpler models miss entirely.

The researchers tested two release scenarios relevant to landfill leakage. The first is an instantaneous release, modeled with a Dirac delta impulse in time, representing a sudden one-time injection of pollutant along a line within the soil. The second is an exponentially decaying release, in which source strength diminishes over time according to a decay coefficient, a more realistic picture of long-term leakage from a degrading liner. For both cases, the closed-form solution was verified against a finite-element numerical solver, with relative errors below 0.5 percent, confirming that the analytical expression correctly satisfies the governing equations and boundary conditions. Because the solution requires no numerical Laplace inversion or discretization of the boundary condition, it runs quickly and serves as a rigorous benchmark for testing more complex numerical codes.

The simulations, run with parameters typical of sandy aquifers, reveal how strongly chemistry controls the fate of a plume. The single most influential parameter is the retardation factor, which measures how much the soil matrix slows contaminant movement relative to groundwater. A sensitivity analysis showed that a 20 percent increase in retardation reduces the peak concentration by roughly 19.6 percent, a sensitivity index of negative 0.979, by far the dominant effect among the parameters tested. The adsorption coefficient also reduces peak concentrations, though more modestly, while the width of the contaminating line source has only a weak positive influence on peak magnitude and essentially no effect on where the peak occurs. Effective groundwater velocity, meanwhile, mainly shifts the location of the peak rather than changing its height.

Perhaps the most striking results come from the direct comparison between Neumann and Dirichlet boundary conditions. At early times, when the plume has not yet reached the boundary, the two formulations agree closely. But as time passes, they diverge dramatically. By ten days of simulated transport, the Neumann solution retains nearly twice as much mass in the domain as the Dirichlet solution, and the peak concentration ratio climbs to nearly three. The reason is physical: a Dirichlet boundary continuously drains mass from the system, steepening gradients and artificially removing contaminant, whereas the Neumann formulation preserves mass and allows the plume to spread more evenly. Compared with the equivalent Dirichlet model, the Neumann formulation predicts a 22 percent lower peak concentration, a peak located 18 percent farther away, and up to 28 percent lower near-field concentrations at early times. For long-term, dispersion-dominated simulations, the choice of boundary condition is not a technicality; it can flip the answer.

The adsorption-desorption comparison carries practical weight as well. When both processes are included, concentrations decline gradually over time, reflecting a dynamic, reversible exchange between dissolved and adsorbed phases. With adsorption alone, concentrations drop faster because metals are irreversibly retained on solid surfaces. Without either process, the decline is steepest of all, governed only by advection, dispersion, and decay. Desorption, in other words, acts as a slow-release reservoir that keeps contaminant levels elevated long after the source has faded, a mechanism that any credible risk assessment or remediation plan must account for.

To demonstrate real-world relevance, the authors applied the model to a demonstrative case of lead leakage from a landfill. The simulation showed that the plume advances only a few meters over a decade, with concentrations falling below drinking-water limits beyond 20 meters from the source. That kind of quantitative result directly supports the design of buffer zones and monitoring networks around waste facilities, giving engineers a defensible basis for deciding how far a safety perimeter must extend.

The authors are candid about the model’s limits. It assumes homogeneous, isotropic soil, steady uniform flow, linear equilibrium adsorption, and no chemical reactions beyond first-order decay. Natural aquifers violate all of these assumptions to some degree: they are heterogeneous, prone to preferential flow, and often exhibit nonlinear, kinetically limited sorption. Transient recharge and pumping further complicate real flow fields. The researchers therefore position the solution as an effective-medium approximation and a screening tool, valuable for benchmarking numerical codes, conducting preliminary site assessments, and building more elaborate models through superposition of multiple sources. Future extensions could incorporate scale-dependent dispersion, nonlinear desorption isotherms, and time-varying velocities. For now, the work delivers something the field has lacked: an exact, fast, mass-conserving answer to how reactive contaminants spread from a leaking source, and a clear demonstration that the boundary conditions we write into our equations shape, sometimes decisively, the environmental futures we predict.

Subject of Research: Analytical modeling of two-dimensional reactive heavy metal contaminant transport in soil under flux-controlled boundary conditions with landfill leakage application

Article Title: Analytical solution for two-dimensional reactive contaminant transport under a Neumann flux boundary: Adsorption, release mode, and landfill leakage application

Article References: Li, Y., & Liang, R. (2026). Analytical solution for two-dimensional reactive contaminant transport under a Neumann flux boundary: Adsorption, release mode, and landfill leakage application. Results in Engineering, 32, Article 113278. https://doi.org/10.1016/j.rineng.2026.113278

Image Credits: AI Generated

DOI: Not provided

Keywords: groundwater contamination, heavy metals, advection-diffusion equation, Neumann boundary condition, adsorption, desorption, landfill leakage, analytical solution, retardation factor, soil pollution, contaminant transport, sensitivity analysis