🌐 𝐃∇² Diffusion Dynamics: Spatiotemporally Continuous Models

Partial Differential Equations (PDEs) provide a rigorous mathematical framework for modeling infectious disease transmission when epidemic dynamics evolve continuously in both space and time. In contrast to ordinary differential equation models, which assume homogeneous mixing, and metapopulation models, which discretize space into patches, PDE-based approaches describe the smooth spatial propagation of pathogens. These models are fundamentally compartmental in nature and extend classical SIR or SEIR systems by incorporating spatial diffusion, thereby enabling the study of epidemic wave formation and spatial heterogeneity.

────────────────────────────────────────────

🧱 Compartmental Structure and Flow

Reaction–diffusion epidemiological models retain the classical compartmental interpretation but treat each compartment as a spatially varying density rather than an aggregate count. The population state depends explicitly on spatial location and time.

Susceptible Population Density, S(x, t)
Represents the density of individuals vulnerable to infection at location x and time t. This density decreases locally due to infection and spreads spatially through diffusion.

Infectious Population Density, I(x, t)
Represents the density of infectious individuals. It increases due to local transmission, decreases due to recovery or removal, and diffuses across the spatial domain.

Recovered Population Density, R(x, t)
Represents individuals removed from the transmission process. This compartment also diffuses spatially, reflecting post-infection movement.

Spatial Spread Mechanism
Diffusion is represented by the Laplacian operator ∇² acting on each compartment, scaled by a diffusion coefficient. The term D∇²C models random movement of individuals in continuous space.

────────────────────────────────────────────

📐 Mathematical Formulation: Reaction–Diffusion System

A standard reaction–diffusion formulation for a spatial SIR model is given by the following system of partial differential equations:

∂S/∂t = Dₛ ∇²S − β S I

∂I/∂t = Dᵢ ∇²I + β S I − γ I

∂R/∂t = Dᵣ ∇²R + γ I

Here, S(x, t), I(x, t), and R(x, t) denote the spatial densities of susceptible, infectious, and recovered individuals. The Laplacian operator ∇² captures spatial diffusion, while the reaction terms describe local epidemiological processes governed by transmission and recovery.

────────────────────────────────────────────

🔢 Table 1. Parameter Definitions

Parameter	Definition
β	Local transmission rate governing infection between susceptible and infectious individuals
γ	Recovery or removal rate, equal to the inverse of the average infectious period
Dₛ	Diffusion coefficient for susceptible individuals
Dᵢ	Diffusion coefficient for infectious individuals
Dᵣ	Diffusion coefficient for recovered individuals
∇²	Laplacian operator representing spatial diffusion

────────────────────────────────────────────

📊 Table 2. Typical Parameter Ranges

Parameter	Typical Range (per day)	Interpretation
β	0.10 – 0.30	Moderate to high transmission intensity
γ	0.05 – 0.10	Infectious period of approximately 10–20 days
D	Context-specific	Governs speed of spatial spread; depends on population mobility and spatial scale

────────────────────────────────────────────

🌍 Applicability and Limitations

Applicability
Reaction–diffusion PDE models are particularly well suited for spatiotemporally continuous analyses of epidemic spread. They are ideal for studying wave-like propagation of infections, local outbreak expansion, and the influence of spatial heterogeneity at neighborhood, city, or regional scales.

Analytical and Numerical Insights
These models enable spatially resolved simulations and support stability and wave-speed analyses. Advanced numerical techniques, such as adaptive mesh refinement, can be used to resolve fine-scale spatial dynamics while maintaining computational efficiency.

Limitations
A central limitation of diffusion-based models is the assumption that population movement occurs only through local, random dispersal. This framework does not naturally capture long-range travel or network-driven mobility, such as air travel or commuting. In addition, numerical simulation of PDEs typically requires discretization of space and time, which can be computationally demanding and sensitive to resolution choices.

────────────────────────────────────────────

📚 Selected References

Guglielmi, N., Iacomini, E., and Viguerie, A. Delay differential equations for the spatially resolved simulation of epidemics.

Kevrekidis, P. G., et al. Reaction–diffusion spatial modeling of epidemic spread.

Oshinubi, K., et al. Spatial epidemiological modeling approaches in infectious disease research.

Grave, M., and Coutinho, A. L. G. A. Adaptive mesh refinement for diffusion–reaction epidemiological models.

Kammegne, B., et al. Spatial diffusion modeling of epidemics with vaccination dynamics.