Metapopulation (or Patch) Models are essential frameworks in mathematical epidemiology for incorporating discrete spatial heterogeneity into the analysis of infectious disease transmission. Rather than assuming homogeneous mixing across one large population, these models divide the geographic domain into distinct spatial units or patches (such as regions, cities, communities) and explicitly model how disease spreads within and between these patches. This structured approach is crucial for understanding how local outbreaks scale into regional or global epidemics.
📐 Compartmental Structure and Flow Explanation
Metapopulation extensions of classical compartmental systems (SIR, SEIR) track the population of each patch using epidemiological variables such as Sᵢ, Eᵢ, Iᵢ, and Rᵢ. While the within-patch transitions follow standard infection and recovery dynamics, spatial heterogeneity arises from inter-patch coupling.
1. Intra-Patch Flow
Individuals progress through local disease stages (Sᵢ → Eᵢ → Iᵢ → Rᵢ) under patch-specific disease parameters.
2. Inter-Patch Coupling
Disease spreads between patches either:
• Implicitly, through a force of infection Λᵢ that accounts for infectious individuals in other patches, or
• Explicitly, by modeling the movement of individuals between patches.
A susceptible individual in patch i experiences a cumulative force of infection Λᵢ that integrates all infectious pressures from interacting patches j, along with contact or travel rates.
Mathematical Formulation (Coupled SIR System)
Consider an n-patch SIR framework where Sᵢ, Iᵢ, and Rᵢ represent the populations in patch i. Transmission into each patch is driven by a force of infection Λᵢ that depends on infection across all patches.
The governing system is:
dSᵢ/dt = − Sᵢ Λᵢ(I₁, …, Iₙ)
dIᵢ/dt = Sᵢ Λᵢ(I₁, …, Iₙ) − γᵢ Iᵢ
dRᵢ/dt = γᵢ Iᵢ
In a multi-patch mass-action model, the force of infection often takes the form:
Λᵢ = Σⱼ βᵢⱼ (Iⱼ / Nⱼ)
where βᵢⱼ represents contact or transmission coupling between patches i and j.
Parameter Definitions and Ranges
Metapopulation models require both epidemiological and spatial parameters:
| Parameter | Definition | Typical Range | Context |
|---|---|---|---|
| γᵢ | Patch-specific recovery or removal rate | 0.14 – 0.5 per day | Reflects infectious period of approximately 2–7 days. |
| Nᵢ | Patch population size | Variable | Must be large enough for deterministic models to apply. |
| βᵢⱼ | Transmission or contact rate from patch j to i | Data-derived | Depends on contact intensity or travel patterns. |
| R₀ | Basic reproduction number | 1.5 – 4.0 | Computed as the dominant eigenvalue of the Next Generation Matrix. |
Applicability and Limitations
Applicability
- Modeling Spatial Propagation
Metapopulation models describe how outbreaks spread geographically over time, capturing regional diffusion patterns and epidemic wave fronts. - Targeted Intervention Strategies
They help evaluate location-specific measures such as regional vaccination, school closures, or travel restrictions, informing resource allocation and prioritization. - Capturing Heterogeneity
Patch models account for differences in socioeconomic, demographic, or epidemiological characteristics across regions, enabling more realistic forecasts than single-population models.
Key Assumptions and Weaknesses
- Within-Patch Homogeneity
Although inter-patch differences are captured, each patch typically assumes uniform mixing internally. - Data Requirements
Parameterizing inter-patch movement and contact requires detailed spatial and demographic data that may be incomplete or unavailable. - Computational and Analytical Complexity
As the number of patches increases, the dimensionality and complexity of the system grow rapidly, making simulation and analysis more challenging.
📚 Selected References
- Kermack, W. O., & McKendrick, A. G. A contribution to the mathematical theory of epidemics.
- Anderson, R. M., & May, R. M. Infectious Diseases of Humans: Dynamics and Control.
- Mossong, J., et al. Social Contacts and Mixing Patterns Relevant to the Spread of Infectious Diseases.
- van den Driessche, P., & Watmough, J. Reproduction numbers and sub-threshold endemic equilibria for compartmental models of disease transmission.
- Prem, K., Cook, A. R., & Jit, M. Projecting social contact matrices in 152 countries using contact surveys and demographic data.