Partial Differential Equations

Partial differential equations (PDEs) form the mathematical backbone for models in physics, engineering, biology and finance. They express relationships between the rates of change of a multivariable function and underpin phenomena as diverse as heat conduction, wave propagation, fluid flow and quantum mechanics. Classical second-order PDEs fall into three archetypal classes—elliptic, parabolic and hyperbolic—governed by the sign of the characteristic discriminant. An elliptic equation, such as Laplace’s equation, describes equilibrium states; a parabolic equation, typified by the heat equation, governs diffusion and irreversible processes; and a hyperbolic equation, like the wave equation, encodes finite-speed propagation of disturbances. Beyond these canonical forms lie fully nonlinear PDEs (Monge–Ampère, k-Hessian), fractional-order models capturing long-range interactions, and systems coupling multiple fields (Navier–Stokes, Schrödinger equations). Analytical methods draw on functional analysis in Sobolev and Besov spaces, a priori estimates and barrier constructions, whereas numerical approaches range from finite-difference and finite-element schemes to spectral and meshless methods. Recent decades have also witnessed the emergence of singular coefficient problems, nonlocal operators, obstacle and free-boundary formulations, and data-driven solvers. This breadth reflects the central role of PDEs in describing continuous media across scales—from microstructured metamaterials to planetary atmospheres—and in guiding the design, control and optimisation of modern technologies.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Recent work on scalar conservation laws with discontinuous flux functions has established quantitative compactness estimates for finite-difference schemes, guaranteeing convergence to entropy solutions even when material properties exhibit jumps. Complementing this, a filtered Chebyshev spectral method has been developed to suppress the Gibbs phenomenon near shock interfaces while preserving high-order accuracy; convergence is proved via compensated compactness, and numerical tests on networked flows show marked improvements over standard finite-volume approaches. In the realm of nonlinear wave equations, sharp lifespan estimates have been derived for semilinear Tricomi models with scale-invariant damping and mass terms. These analyses reveal the subtle competition between Strauss and Fujita critical exponents, yielding precise blow-up thresholds and upper bounds on solution lifetimes in both subcritical and supercritical regimes. Meanwhile, advances in stochastic homogenisation have demonstrated that periodising the underlying random ensemble—rather than individual realisations—reduces systematic bias in representative-volume methods. By combining higher-order two-scale expansions of the Green’s function with Malliavin-calculus-based concentration inequalities, error scaling of order L⁻ᵈ is achieved in d dimensions, markedly improving the accuracy of effective coefficient estimates for heterogeneous media.

Partial Differential Equations publication trend

The graph below shows the total number of articles in partial differential equations across all publications each year (not limited to Nature Index journals).

Technical terms

Partial differential equation: An equation involving partial derivatives of a function of several variables, used to model continuous systems.

Elliptic, parabolic, hyperbolic: Classification of second-order PDEs by the sign of the characteristic discriminant, corresponding to equilibrium, diffusive or wave-like processes.

Weak solution: A function that satisfies a PDE in an integral or distributional sense, permitting less regularity than classical derivatives.

Entropy solution: A weak solution to a conservation law that fulfils additional admissibility (entropy) conditions to ensure physical relevance and uniqueness.

Blow-up: The phenomenon whereby a solution to a nonlinear PDE becomes unbounded in finite time, indicating the formation of singularities.

Representative volume element (RVE): A finite sample of a random heterogeneous medium used to approximate its effective macroscopic properties in homogenisation theory.

References

1. A filtered Chebyshev spectral method for conservation laws on network. Computers & Mathematics with Applications (2023).
2. Bias in the Representative Volume Element method: Periodize the Ensemble Instead of Its Realizations. Foundations of Computational Mathematics (2023).