1 Definition and basic idea

The entropy condition is an admissibility rule used in the analysis of partial differential equations, especially hyperbolic conservation laws. Its main purpose is to distinguish physically relevant solutions from mathematically possible but unrealistic ones. In many problems, the governing equations admit several weak solutions, and the entropy condition selects the one that behaves in a manner consistent with dissipation and the irreversible character of real processes.

1.1 Weak solutions and nonuniqueness

Many conservation laws develop discontinuities even when the initial data are smooth. In such cases, classical differentiable solutions may cease to exist, so the equations are interpreted in a weaker integral sense. This broader framework allows solutions to include shocks and other singular features. However, once weak solutions are admitted, the same initial data can produce more than one mathematical solution, so an additional criterion is needed.

1.2 Need for an admissibility criterion

Without a selection principle, some weak solutions may violate expected wave behavior. For example, a discontinuity may propagate in a direction or with a structure that contradicts physical intuition. The entropy condition serves as a filter that rejects such solutions and preserves those that are consistent with the irreversible nature of many real systems. In this way, it restores well-posedness in a practical sense.

1.3 Physical interpretation of entropy

The term entropy comes from thermodynamics, where it is associated with irreversibility and the tendency of isolated systems to evolve toward states of greater disorder. In mathematical PDE theory, the analogy is conceptual rather than literal. An entropy condition usually imposes an inequality that prevents artificial decreases in entropy-like quantities across discontinuities. The condition therefore encodes a preferred direction of evolution for shocks and related structures.

2 Entropy conditions in conservation laws

Conservation laws describe quantities that are preserved in the absence of sources or sinks, such as mass, momentum, or energy. When solutions remain smooth, the conservation law is straightforward. When discontinuities appear, entropy conditions become central to identifying the correct weak solution.

2.1 Scalar conservation laws

For a scalar conservation law, a single unknown evolves according to a flux function that determines how the quantity moves through space. These equations often provide the clearest setting for understanding entropy criteria, because the wave structure is relatively simple and can be analyzed in detail.

2.1.1 Shock waves

A shock wave is a discontinuity across which the conserved quantity changes abruptly. Shocks arise naturally when characteristics converge and cause compression. Although a shock may satisfy the conservation law in weak form, not every discontinuity is acceptable. The entropy condition ensures that the shock has the correct orientation and that characteristics enter rather than leave the discontinuity in an unphysical way.

2.1.2 Rarefaction waves

Rarefaction waves are smooth expanding solutions that occur when characteristics spread apart. They often replace discontinuities that would otherwise be unstable or inadmissible. In scalar problems, the entropy condition typically favors rarefaction over expansion shocks, which are mathematically possible but physically incorrect. This distinction is one of the clearest examples of entropy selection.

2.2 Systems of conservation laws

In systems, several conserved quantities interact, making the wave structure more complex. Multiple characteristic families may appear, and shocks, contact discontinuities, and rarefactions can coexist. Entropy conditions in this setting are correspondingly more sophisticated.

2.2.1 Entropy-entropy flux pairs

For a conservation law system, an entropy is usually a convex scalar function of the state variables, paired with a corresponding entropy flux. Together, they satisfy an additional conservation or balance relation for smooth solutions. For weak solutions, this relation becomes an inequality. The convexity requirement helps ensure that the entropy inequality selects stable and physically meaningful behavior.

2.2.2 Entropy inequalities

Entropy inequalities generalize the admissibility condition to discontinuous solutions. Instead of requiring exact conservation of the entropy quantity, the inequality allows entropy to be produced or dissipated in a controlled manner. This reflects the irreversible nature of many physical processes and excludes solutions that would create entropy in the wrong direction or fail to dissipate it appropriately.

2.3 Kruzhkov entropy condition

The Kruzhkov entropy condition is a standard criterion for scalar conservation laws. It uses a family of entropy inequalities built from absolute-value functions centered at arbitrary constants. This formulation is especially powerful because it yields existence and uniqueness results for broad classes of initial data. It also provides a robust framework for proving stability under perturbations.

3 Classical entropy conditions

Several classical formulations of entropy admissibility are widely used in the theory of shocks and conservation laws. These conditions are often equivalent in standard settings, though each emphasizes a different aspect of wave behavior.

3.1 Lax entropy condition

The Lax entropy condition characterizes admissible shocks by comparing shock speed with characteristic speeds on both sides of the discontinuity. For a compressive shock, characteristics should enter the shock from both sides rather than spread out from it. This condition is especially important in systems of hyperbolic conservation laws, where it helps identify physically relevant discontinuities.

3.2 Oleinik entropy condition

The Oleinik entropy condition is often expressed as a one-sided bound that controls how rapidly a solution can change. In scalar conservation laws, it can be formulated as a monotonicity or slope restriction that suppresses nonphysical expansion shocks. The condition is useful because it gives a direct analytical handle on admissibility and supports uniqueness arguments.

3.3 Rankine-Hugoniot relation

The Rankine-Hugoniot relation gives the speed of a discontinuity that satisfies the conservation law in weak form. It is derived by integrating across the jump and balancing flux differences against the motion of the interface. This relation alone does not determine admissibility, but it is a necessary ingredient for any shock solution.

3.3.1 Connection to discontinuity speed

The Rankine-Hugoniot formula links the jump in the conserved quantity to the velocity of the discontinuity. It specifies how fast the interface must move so that the total conserved amount remains balanced. In practical terms, it provides the kinematic law governing shock propagation.

3.3.2 Selection of admissible shocks

Although the Rankine-Hugoniot relation identifies possible shocks, it does not eliminate all nonphysical ones. The entropy condition must be added to select the correct subset. Together, these two requirements determine both the speed and the admissibility of the discontinuity.

4 Mathematical properties

Entropy conditions play a central role in the modern theory of nonlinear hyperbolic PDEs. They support rigorous results on existence, uniqueness, and regularity, and they connect analytic theory with computational methods.

4.1 Existence of entropy solutions

For many conservation laws, one can construct entropy solutions through approximation schemes such as vanishing viscosity, finite difference methods, or front-tracking procedures. These methods introduce controlled regularization or discretization and then pass to a limit. The entropy condition is preserved in the limit, ensuring that the resulting weak solution is admissible.

4.2 Uniqueness and stability

A major advantage of entropy formulations is that they often restore uniqueness. In many scalar problems, the entropy solution is the only weak solution compatible with the entropy inequalities. Stability estimates also become available, meaning that small changes in initial data lead to small changes in the solution. This property is crucial both for theory and for numerical approximation.

4.3 BV and regularity considerations

Solutions with bounded variation, often abbreviated BV, are especially important in entropy theory. BV bounds help control oscillations and make it possible to analyze shocks and wave interactions precisely. While entropy solutions may still contain discontinuities, BV regularity provides enough structure to study their evolution and to prove compactness in approximation arguments.

5 Applications

Entropy conditions appear in many models where waves, shocks, and discontinuities arise naturally. They are indispensable in both theoretical analysis and applied modeling.

5.1 Gas dynamics

In gas dynamics, entropy conditions help determine the physically correct behavior of compressible flow. Shocks, expansion fans, and contact discontinuities all occur in this setting. The admissibility criterion is essential for describing how pressure, density, and velocity change across moving fronts.

5.2 Fluid mechanics

In fluid mechanics, entropy methods are used in simplified and idealized models of flow. They help select the proper solution when abrupt changes in velocity or density appear. Even when the full Navier-Stokes equations are not treated directly, entropy conditions remain important in reduced hyperbolic models.

5.3 Traffic flow models

Traffic flow can often be modeled by conservation laws for vehicle density. In these models, shocks may correspond to traffic jams, while rarefactions describe the spreading of cars after congestion eases. Entropy conditions prevent unrealistic behaviors such as backward-moving jams that would violate the intended interpretation of the model.

5.4 Nonlinear wave propagation

Nonlinear waves in acoustics, elasticity, and related areas can also develop steep fronts and discontinuities. Entropy conditions help determine which wave patterns are stable and physically realizable. They are particularly useful when wave steepening leads to the formation of shocks.

Entropy conditions are connected to several broader ideas in mathematics and physics. These connections help explain why the terminology is shared across different disciplines.

6.1 Second law of thermodynamics

The second law of thermodynamics states that entropy production in isolated systems is nonnegative. Mathematical entropy conditions echo this principle by imposing a directional inequality across shocks. The analogy is strongest in continuum models where dissipation and irreversibility are built into the formulation.

6.2 Variational principles

Although entropy conditions are not the same as variational principles, both are selection mechanisms. Variational methods choose an extremum of an energy or action functional, while entropy conditions choose an admissible weak solution among many possibilities. In some settings, the two ideas interact through convexity and stability arguments.

6.3 Shock admissibility

Shock admissibility is the general problem of deciding which discontinuities may occur in a solution of a conservation law. The entropy condition is one of the principal admissibility criteria. It is used to rule out shocks that satisfy the balance laws algebraically but fail to represent physically consistent wave motion.

6.4 Entropy solution

An entropy solution is a weak solution that satisfies the relevant entropy inequalities. This is the standard notion of solution for many nonlinear conservation laws. It combines the flexibility needed to include discontinuities with the admissibility needed for uniqueness and physical relevance.