Modelling the spatial spread of COVID-19 in a German district using a diffusion model

Moritz Schäfer; Peter Heidrich; Thomas Götz; Moritz Schäfer; Peter Heidrich; Thomas Götz

doi:10.3934/mbe.2023940

Mathematical Biosciences and Engineering

2023, Volume 20, Issue 12: 21246-21266. doi: 10.3934/mbe.2023940

Previous Article Next Article

Research article Special Issues

Modelling the spatial spread of COVID-19 in a German district using a diffusion model

1.
Mathematical Institute, University of Koblenz, 56070 Koblenz, Germany
2.
Magister Laukhard IGS Herrstein/Rhaunen, 55756 Herrstein, Germany

Academic Editor: Shengqiang Zhang

Received: 19 July 2023 Revised: 19 September 2023 Accepted: 12 October 2023 Published: 29 November 2023

In this study, we focus on modeling the local spread of COVID-19 infections. As the pandemic continues and new variants or future pandemics can emerge, modelling the early stages of infection spread becomes crucial, especially as limited medical data might be available initially. Therefore, our aim is to gain a better understanding of the diffusion dynamics on smaller scales using partial differential equation (PDE) models. Previous works have already presented various methods to model the spatial spread of diseases, but, due to a lack of data on regional or even local scale, few actually applied their models on real disease courses in order to describe the behaviour of the disease or estimate parameters. We use medical data from both the Robert-Koch-Institute (RKI) and the Birkenfeld district government for parameter estimation within a single German district, Birkenfeld in Rhineland-Palatinate, during the second wave of the pandemic in autumn 2020 and winter 2020–21. This district can be seen as a typical middle-European region, characterized by its (mainly) rural nature and daily commuter movements towards metropolitan areas. A basic reaction-diffusion model used for spatial COVID spread, which includes compartments for susceptibles, exposed, infected, recovered, and the total population, is used to describe the spatio-temporal spread of infections. The transmission rate, recovery rate, initial infected values, detection rate, and diffusivity rate are considered as parameters to be estimated using the reported daily data and least square fit. This work also features an emphasis on numerical methods which will be used to describe the diffusion on arbitrary two-dimensional domains. Two numerical optimization techniques for parameter fitting are used: the Metropolis algorithm and the adjoint method. Two different methods, the Crank-Nicholson method and a finite element method, which are used according to the requirements of the respective optimization method are used to solve the PDE system. This way, the two methods are compared and validated and provide similar results with good approximation of the infected in both the district and the respective sub-districts.
- epidemiology,
- modelling,
- PDEs,
- parameter fitting,
- metropolis,
- adjoint method
Citation: Moritz Schäfer, Peter Heidrich, Thomas Götz. Modelling the spatial spread of COVID-19 in a German district using a diffusion model[J]. Mathematical Biosciences and Engineering, 2023, 20(12): 21246-21266. doi: 10.3934/mbe.2023940

Related Papers:

Abstract

In this study, we focus on modeling the local spread of COVID-19 infections. As the pandemic continues and new variants or future pandemics can emerge, modelling the early stages of infection spread becomes crucial, especially as limited medical data might be available initially. Therefore, our aim is to gain a better understanding of the diffusion dynamics on smaller scales using partial differential equation (PDE) models. Previous works have already presented various methods to model the spatial spread of diseases, but, due to a lack of data on regional or even local scale, few actually applied their models on real disease courses in order to describe the behaviour of the disease or estimate parameters. We use medical data from both the Robert-Koch-Institute (RKI) and the Birkenfeld district government for parameter estimation within a single German district, Birkenfeld in Rhineland-Palatinate, during the second wave of the pandemic in autumn 2020 and winter 2020–21. This district can be seen as a typical middle-European region, characterized by its (mainly) rural nature and daily commuter movements towards metropolitan areas. A basic reaction-diffusion model used for spatial COVID spread, which includes compartments for susceptibles, exposed, infected, recovered, and the total population, is used to describe the spatio-temporal spread of infections. The transmission rate, recovery rate, initial infected values, detection rate, and diffusivity rate are considered as parameters to be estimated using the reported daily data and least square fit. This work also features an emphasis on numerical methods which will be used to describe the diffusion on arbitrary two-dimensional domains. Two numerical optimization techniques for parameter fitting are used: the Metropolis algorithm and the adjoint method. Two different methods, the Crank-Nicholson method and a finite element method, which are used according to the requirements of the respective optimization method are used to solve the PDE system. This way, the two methods are compared and validated and provide similar results with good approximation of the infected in both the district and the respective sub-districts.

References

[1]	P. Heidrich, M. Schäfer, M. Nikouei, T. Götz, The COVID-19 outbreak in Germany–Models and Parameter Estimation, Commun. Biomath. Sci., 3 (2020), 37–59. https://doi.org/10.5614/cbms.2020.3.1.5. doi: 10.5614/cbms.2020.3.1.5
[2]	M. Schäfer, K. P. Wiyaya, R. Rockefeller, T. Götz, The impact of travelling on the COVID-19 infection cases in Germany, BMC Infect. Dis., 2 (2022). https://doi.org/10.1186/s12879-022-07396-1 doi: 10.1186/s12879-022-07396-1
[3]	A. Viguerie, G. Lorenzo, F. Auricchio, D. Baroli, T. J. R. Hughes, A. Patton, et al., Simulating the spread of COVID-19 via a spatially-resolved susceptible–exposed–infected–recovered–deceased (SEIRD) model with heterogeneous diffusion, Appl. Math. Lett., 111 (2021), 106617. https://doi.org/10.1016/j.aml.2020.106617 doi: 10.1016/j.aml.2020.106617
[4]	H. Wang, N. Yamamoto, Using a partial differential equation with Google Mobility data to predict COVID-19 in Arizona, Math. Biosci. Eng., 17 (2020), 4891–4904. https://doi.org/10.3934/mbe.2020266 doi: 10.3934/mbe.2020266
[5]	A. Elsonbaty, Z. Sabir, R. Ramaswamy, W. Adel, Dynamical analysis of a novel discrete fractional SITRS model for COVID-19, Fractals, 29 (2021). https://doi.org/10.1142/S0218348X21400351 doi: 10.1142/S0218348X21400351
[6]	N. Ahmed, A. Elsonbaty, A. Raza, M. Rafiq, W. Adel, Numerical simulation and stability analysis of a novel reaction–diffusion COVID-19 model, Nonlinear Dynam., 106 (2021), 1293–1310. https://doi.org/10.1007/s11071-021-06623-9 doi: 10.1007/s11071-021-06623-9
[7]	C. Kuehn, J. Mölter, The influence of a transport process on the epidemic threshold, J. Math. Biol., 85 (2020), 37–59. https://doi.org/10.1007/s00285-022-01810-7 doi: 10.1007/s00285-022-01810-7
[8]	K. Logeswari, C. Ravichandran, K. S. Nisar, Mathematical model for spreading of COVID-19 virus with the Mittag–Leffler kernel, Numer. Meth. Partial Differ. Equations, (2020), 1–16. https://doi.org/10.1002/num.22652 doi: 10.1002/num.22652
[9]	P. J. Harris, B. E. J. Bodmann, A mathematical model for simulating the spread of a disease through a country divided into geographical regions with different population densities, J. Math. Biol., 85 (2022). https://doi.org/10.1007/s00285-022-01803-6 doi: 10.1007/s00285-022-01803-6
[10]	H. Berestycki, J. M. Roquejoffre, L. Rossil, Propagation of epidemics along lines with fast diffusion, Bull. Math. Biol., 83 (2020). https://doi.org/10.1007/s11538-020-00826-8 doi: 10.1007/s11538-020-00826-8
[11]	H. Abboubakar, R. Racke, N. Schlosser, A Reaction-Diffusion Model for the Transmission Dynamics of the Coronavirus Pandemic with Reinfection and Vaccination Process, Konstanzer Schriften in Mathematik, KOPS Universität Konstanz, 2023.
[12]	Y. Nawaz, M. S. Arif, K. Abodayeh, W. Shatanawi, An explicit unconditionally stable scheme: application to diffusive COVID-19 epidemic model, Adv. Differ. Equations, 2021 (2021). https://doi.org/10.1186/s13662-021-03513-7 doi: 10.1186/s13662-021-03513-7
[13]	M. Grave, A. Viguerie, G. F. Barros, A. Reali, A. Coutinho, Assessing the Spatio-temporal Spread of COVID-19 via Compartmental Models with Diffusion in Italy, USA, and Brazil, Arch. Comput. Meth. Eng., 28 (2021), 4205–4223. https://doi.org/10.1007/s11831-021-09627-1 doi: 10.1007/s11831-021-09627-1
[14]	Robert-Koch-Institute, COVID-19 Dashboard, 2020. https://experience.arcgis.com/experience/478220a4c454480e823b17327b2bf1d4
[15]	W. O. Kermack, A. G. McKendrick, Contributions to the mathematical theory of epidemics–I. 1927., Bull. Math. Biol., 53 (1991), 33–55. https://doi.org/10.1007/bf02464423 doi: 10.1007/bf02464423
[16]	M. Martcheva, An Introduction to Mathematical Epidemiology, Springer, New York, 2015.
[17]	X. He, E. H. Y. Lau, P. Wu, X. Deng, J. Wang, X. Hao, Temporal dynamics in viral shedding and transmissibility of COVID-19, Nat. Med., 26 (2020), 672–675. https://doi.org/10.1038/s41591-020-0869-5 doi: 10.1038/s41591-020-0869-5
[18]	N. Britton, Reaction-Diffusion Equations and Their Applications to Biology, Academic Press, London, 1986.
[19]	C. M. Oishi, J. Y. Yuan, J. A. Cuminato, D. E. Stewart, Stability analysis of Crank-Nicolson and Euler schemes for time-dependent diffusion equations, BIT Numer. Math., 55 (2015), 487–513. https://doi.org/10.1007/s10543-014-0509-x doi: 10.1007/s10543-014-0509-x
[20]	S. MacNamara, G. Strang, Operator splitting, in Splitting Methods in Communication, Imaging, Science, and Engineering, Springer, (2016), 95–114.
[21]	G. Evans, J. Blackledge, P. Yardley, Numerical Methods for Partial Differential Equations, Springer, London, 1999.
[22]	J. R. Dormand, P. J. Prince, A family of embedded Runge-Kutta formulae, J. Comput. Appl. Math., 6 (1980), 19–26.
[23]	Overpass-Turbo. https://overpass-turbo.eu/
[24]	P. Heidrich, T. Götz, Parameter Estimation via Adjoint Functions in Epidemiological Reaction-Diffusion Models, in Progress in Industrial Mathematics at ECMI 2021, Springer, (2022), 115–122. https://doi.org/10.1007/978-3-031-11818-0_16
[25]	N. Metropolis, A. W. Rosenbluth, M. W. Rosenbluth, A. H. Teller, E. Teller, Equation of State Calculations by Fast Computing Machines, J. Chem. Phys., 21 (1953), 1087–1092. doi: https://doi.org/10.1063/1.1699114 doi: 10.1063/1.1699114
[26]	A. Gelman, J. B. Carlin, H. S. Stern, D. B. Rubin, Bayesian Data Analysis, 2nd Edition, Chapman and Hall, London, 1996.
[27]	W. R. Gilks, S. Richardson, D. J. Spiegelhalter, Markov chain Monte Carlo in Practice, Chapman and Hall/CRC, London, 1996.
[28]	M. Schäfer, T. Götz, Modelling Dengue Fever Epidemics in Jakarta, Int. J. Appl. Comput. Math., 6 (2020). https://doi.org/10.1007/s40819-020-00834-1. doi: 10.1007/s40819-020-00834-1
[29]	D. N. Rusatsi, Bayesian analysis of SEIR epidemic models, Ph.D thesis, Lappeenranta University of Technology, 2015.

Reader Comments

Your name:*

Email:*
© 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)