Reaction-diffusion spatial modeling of COVID-19: Greece and Andalusia as case examples

We examine the spatial modeling of outbreak COVID-19 in two regions: autonomous community Andalusia Spain and mainland Greece. start with a 0D compartmental epidemiological model consisting Susceptible, Exposed, Asymptomatic, (symptomatically) Infected, Hospitalized, Recovered, deceased populations. emphasize importance viral latent period key role an asymptomatic population. optimize parameters for both regions by comparing predictions to cumulative number infected total deaths via minimizing $\ell^2$ norm difference between observed data. consider sensitivity on reasonable variations initial conditions, addressing issues parameter identifiability. pre-quarantine post-quarantine evolution epidemic time-dependent change transmission rates that arises response containment measures. Subsequently, spatially distributed version form reaction-diffusion equations is developed. that, after localized seeding infection, its spread governed diffusion (and "reactions") symptomatically populations, which decrease imposed restrictive inserted maps regions, we imported population-density data into COMSOL, was subsequently used solve numerically PDEs. Upon discussing how adapt this setting, show these models bear significant potential towards capturing well-mixed, description expansion pandemic regions. Veins refinement assumptions future work are also explored.