🌍 Spatial Epidemiology: Reaction–Diffusion Models for Contagion Wavefronts 🌊

Partial Differential Equation (PDE) models are used in mathematical epidemiology to move beyond the simplifying assumption of homogeneous mixing, allowing for the representation of disease dynamics in continuous space (x) and time (t). Reaction–Diffusion systems describe both the biological progression of disease (reaction) and the spatial dispersal of hosts (diffusion). This framework is essential for analyzing geographic disease propagation and determining the speed and structure of epidemic wavefronts.

---

📐 Compartmental Structure and Flow Explanation

In spatial epidemic models, traditional epidemiological compartments such as Susceptible (S), Infectious (I), and Recovered (R) are represented as spatial density functions U(x, t) defined over a domain Ω.

The rate of change of any compartmental density ∂U/∂t is governed by two main processes:

1. Reaction Term, f(U)
   Represents local disease processes such as infection, recovery, and demographic change. These terms mirror those found in standard ODE-based epidemic models.
2. Diffusion Term, DΔU
   Represents the random spatial movement of individuals, modeled using Fick’s Law of diffusion. The Laplace operator Δ describes how individuals disperse from areas of higher to lower concentration.

Together, reaction and diffusion define the spatial-temporal evolution of the epidemic across Ω.

---

🧮 Mathematical Formulation (Diffusive SIR System)

A spatial SIR model defined on a bounded domain Ω with diffusion coefficients Dₛ, Dᵢ, and Dᵣ is expressed by the following PDE system:

∂S(x, t)/∂t = Dₛ ΔS(x, t) − β S(x, t) I(x, t) − μ S(x, t)
∂I(x, t)/∂t = Dᵢ ΔI(x, t) + β S(x, t) I(x, t) − (γ + μ) I(x, t)
∂R(x, t)/∂t = Dᵣ ΔR(x, t) + γ I(x, t) − μ R(x, t)

The model typically uses homogeneous Neumann boundary conditions (∂U/∂n = 0 on ∂Ω), meaning there is no flux of individuals across the domain boundary.

---

⚗ Parameter Definitions and Ranges

Parameter	Definition	Typical Range	Context
Dₛ, Dᵢ, Dᵣ	Diffusion coefficients	Positive values	Represent random movement speeds of each compartment.
β	Transmission rate	0.2 – 1.0 per day	Determines infection likelihood per contact.
γ	Recovery rate	0.14 – 0.5 per day	Inverse of mean infectious period (approximately 2–7 days).
μ	Natural mortality rate	Approximately 1/27,000 per day	Often negligible for short epidemic horizons.

---

🌐 Applicability and Limitations

Applicability

1. Spatial Spread Dynamics
   Reaction–diffusion models quantify the speed and direction of disease spread across landscapes, enabling the characterization of epidemic wavefronts.
2. Targeted Spatial Control
   Useful for evaluating geographically focused interventions such as buffer zones, targeted vaccination corridors, or spatial suppression barriers.
3. Persistence and Stability Analysis
   Enables mathematical investigation of disease-free and endemic equilibrium states within spatially structured populations.

Limitations

1. Fickian Random Movement Assumption
   Assumes undirected, diffusive movement, which may not adequately reflect human travel patterns or constrained movement behaviors.
2. Local Homogeneity
   Models assume uniform mixing within infinitesimally small regions, potentially oversimplifying heterogeneity at fine spatial scales.
3. High Mathematical and Computational Complexity
   PDE systems generally lack closed-form solutions and require intensive numerical methods, which increases computational costs for simulation and calibration.

---

📚 Selected References

1. Kermack, W. O., & McKendrick, A. G. A contribution to the mathematical theory of epidemics.
2. Murray, J. D., Stanley, E. A., & Brown, D. L. On the spatial spread of rabies among foxes.
3. Mollison, D. Dependence of epidemic and population velocities on basic parameters.
4. Wu, J. Spatial Structure: Partial differential equation models.
5. Lotfi, E. M., Maziane, M., Hattaf, K., & Yousfi, N. Partial differential equations of an epidemic model with spatial diffusion.