Intensive Programme Mathematical Models in Life and Social Sciences

Course: Structured Population Dynamics in ecology and epidemiology

Lecturer: Jordi Ripoll (Universitat de Girona, Spain)
Dates and place: September 7-19 2009 – Alba Adriatica, Italy

Abstract:
Structured population dynamics deals with the time evolution of the population composition
with respect to the state of the individuals. Typically, the structuring variables are age (with
different interpretations, e.g. chronological or time since infection or time since cell division),
size, phenotype, maturity level, space, etc. The simplest structure is given by the age of the indi-
viduals since the evolution of age over time proceeds with speed one. Moreover, age-structured
population models play a central role since, in some cases, models considering other physio-
logical characteristics can be reduced to them. The course encompasses (describes/analyzes)
different types of models: unstructured and structured, continuous and discrete, all of them
coming from ecology (highlighting the interactions between individuals and their environment)
and epidemiology (e.g. the spread of infectious diseases). The course is divided in 4 parts: an
introduction to the abstract setting for the linear case (semigroup approach); early ecological
and epidemiological models (ordinary differential equations); continuously structured popula-
tion models (partial differential equations/integral equations); and matrix population models
(discrete structure). Background: basic knowledge of ordinary differential equations and linear
algebra.

Program:

1. Introduction: time-continuous linear population dynamics. Differential equation and In-

tegral equation. Semigroup approach.

2. Early ecological models. Exponential/logistic population growth. Allee effect. Probabilis-

tic interpretation.

3. Early epidemiological models. Structure according to disease stage: SIS and SIR models.
Epidemic threshold theorem. Basic reproductive number R0 .

4. Continuously age-structured population models. Lotka-Mckendrick equation. Integration

along characteristics. Renewal equation. Asynchronous exponential growth. “Stable”
age distribution. (Generalized) Gurtin-MacCamy equation. Equilibria and linearization.
Kermack-Mckendrick equation.

5. Size-dependent problems. Change of variables. Reduction via linear chain trick.

6. Numerical Simulations. Discretization. Periodic orbits.

7. Matrix population models. Leslie/Usher/Tridiagonal models. Perron-Frobenious theory.

Fundamental theorem of demography. Basic reproduction number R0 . Fibonacci sequence.
 Bibliography: books

[1] F. Brauer and C. Castillo-Chávez. Mathematical models in population biology and epi-
demiology. Texts in applied mathematics 40. Springer-Verlag, New York, 2001.

[2] R. Rudnicki, editor. Mathematical modelling of population dynamics, volume 63 of Banach

Center Publications. Polish Academy of Sciences Institute of Mathematics, Warsaw, 2004.
Papers from the conference held in Bedlewo, June 24–28, 2002.

[3] H.R. Thieme. Mathematics in population biology. Princeton Series in Theoretical and
Computational Biology. Princeton University Press, Princeton, NJ, 2003.

[4] M. Iannelli. Mathematical theory of age-structured population dynamics, volume 7 of

Applied Mathematics Monograph C.N.R. Giardini editori e stampatori, Pisa, 1995.

[5] M. Iannelli, M. Martcheva and F.A. Milner. Gender-structured population modeling. Soci-
ety for Industrial and Applied Mathematics (SIAM), 2005.

[6] G.F. Webb. Theory of nonlinear age-dependent population dynamics, volume 89 of Mono-
graphs and Textbooks in Pure and Applied Mathematics. Marcel Dekker Inc., New York,
1985.

[7] N.F. Britton. Essential mathematical biology. Springer Undergraduate Mathematics Series.
Springer-Verlag London Ltd., London, 2003.

[8] J.A.J. Metz and O. Diekmann. The Dynamics of Physiologically Structured Populations.
Lecture Notes in Biomathematics, Vol. 68. Berlin, Springer, 1986.

[9] J.M. Cushing. An introduction to structured population dynamics. Philadelphia, Pa.

Society Industrial and Applied Mathematics, 1998.

[10] H. Caswell. Matrix population models : construction, analysis, and interpretation. Sun-
derland (Mass.). Sinauer Associates, cop. 2001.

[11] B. Perthame. Transport Equations in Biology. Birkhäuser Verlag, Basel, 2007.

[12] O. Diekmann and J.A.P. Heesterbeek. Mathematical epidemiology of infectious diseases.

John Wiley & Sons Ltd., Chichester, 2000.

Bibliography: papers

[1] F.R. Sharpe and A.J. Lotka. A Problem in Age Distribution, Philosophical Magazine, 21
(1911), 435–438.

[2] W. Feller. On the Integral Equation of renewal Theory, Ann. Math. Stat., 12 (1941),
243–267.

[3] M.E. Gurtin and R.C. MacCamy. Non-linear Age-dependent Population Dynamics, Arch.
Rat. Mech. Anal., 54 (1974), 281–300.
 [4] W.O. Kermack and A.G. Mckendrick. Contributions to the mathematical theory of epi-
demics. Part I (reprinted from Proc. Roy. Soc. A 115, 700-721, 1927). Bull Math. Biol.
53, 33–55, 1991.

[5] C.-K. Li and H. Schneider. Applications of Perron-Frobenius theory to population dynamics,

J. Math. Biol. 44, 450–462 (2002).

[6] À. Calsina and C. Perelló. Equations for biological evolution Proc. Roy. Soc. Edinburgh
Sect. A, 1995, 125, 939–958

[7] À. Calsina and J. Saldaña. A model of physiologically structured population dynamics with
a nonlinear individual growth rate J. Math. Biol., 1995, 33, 335–364

[8] À. Calsina and J. Saldaña. Asymptotic behaviour of a model of hierarchically structured
population dynamics J. Math. Biol., 1997, 35, 967–987

[9] À. Calsina and J. Saldaña. Global dynamics and optimal life history of a structured popu-
lation model SIAM J. Appl. Math., 1999, 59, 1667–1685

[10] À. Calsina and J. Ripoll. Evolution of age-dependent sex-reversal under adaptive dynamics.
J. Math. Biol., Online First, 2009.

[11] À. Calsina and J. Ripoll. A general structured model for a sequential hermaphrodite
population. Math. Biosci., 208(2):393–418, 2007.

[12] M. Iannelli and J. Ripoll. Asynchronous growth and competition in a two-sex age-
structured population model. CRM prep. 854, April 2009.

[13] EPICO Working Group. Scenarios of diffusion and control of influenza pandemic in Italy.
Istituto Superiore di Sanità 2006. Rapporti ISTISAN 06/33 (in Italian). Roma, Italy.

[14] J. Ripoll, J. Saldaña, and J.C. Senar. Evolutionarily stable transition rates in a stage-
structured model. An application to the analysis of size distributions of badges of social
status. Math. Biosci., 190(2):145–181, 2004.

[15] À. Calsina and J. Ripoll. Hopf bifurcation in a structured population model for the sexual
phase of monogonont rotifers. J. Math. Biol., 45(1):22–36, 2002.

[16] D. Juher, J. Ripoll, and J. Saldaña. Analysis and Monte Carlo simulations of a model for
the spread of infectious diseases in heterogeneous metapopulations. Phys. Rev. E, 2009.

Material not covered in the course: delay-differential equations, Lotka-Volterra type models,
Biological motion and chemotaxis, Adaptive dynamics of phenotypic traits, Selection-Mutation
equations, and Complex networks.
 Final Test (Master Students):
Consider the following linear problem of the Fibonacci’s rabbits in continuous age and time:
Z ∞
ut (a, t) + ua (a, t) = 0 , u(0, t) = u(a, t) da , (1)
2

where u(a, t) is the age-density of pairs of rabbits. Notice that there is no mortality term and
the birth rate (the influx of newborns) is given by the mature pairs (age a ≥ 2).

1. Find the eigenvalues and the eigenfunctions of the linear system (1). Hint: compute the
solutions with separate variables.
P (t+1)
2. Compute the “stable” age distribution of the pairs of rabbits and lim , where P (t)
t→∞ P (t)
is the total population.

Exercises of the slides:

page 6 For the Allee model, find the relation between population x and time t (implicit solution).

page 8 Draw the qualitative behaviour of the time-continuous models (Malthus, Verhulst, Allee).

page 11–12 Find the relation of the parameter β with respect to infectiveness, total population and
the contact rate when transmission is limited (e.g. c = 1) and when it is non-limited (e.g.
c = N ). Find the units of all parameters in both cases.

page 14 Compute and interpret the growth rate r and the carrying capacity K for the particular
logistic equation of the SIS model.

page 17–18 Integrate Lotka-Mckendrick system (1) along the characteristic lines a = t + c and obtain
system (2). Hint: write a linear ode for ν(t) := u(t + c, t) and solve it with initial condition
at tc = max{0, −c}.

page 18 Obtain the solutions with separate variables letting u(a, t) = T (t) A(a). Show that any
complex solution λ = x + yi of the characteristic equation is such that Re(λ) < α∗ . Hint:
use the properties of the complex exponential function.

page 19 Derive the renewal equation (3) from the boundary condition of (1) and using system (2).

page 21 Solve the “linear” ode for the normalized eigenfunction of the adjoint operator.

page 26 Write an extension of the SIS model with infection age as the structuring variable.

page 28 Find the expressions for the new fertility and the new mortality for the reduced age-
dependent problem.

page 29 Check the terms appearing in the matrix of the ode. Hint: multiply the equation by a
suitable weight function in age and then integrate over the age span.

page 30 Write the numerical scheme for the case of the Lotka-Mckendrick (linear) equation. Is it
an implicit scheme?
page 33 Write the time-discrete version of the SIR and SIS models [pages 12 and 14]. Hint: dis-
cretize the ode and convert rates into probabilities. You can obtain a linear system if the
density-dependent probabilities are fixed by the initial condition.

page 36 Show the Fundamental theorem of demography when the projection matrix is such that
P = W · Λ · V T.

page 39 Check the expressions of R0 for the Leslie and Usher matrices. Hint: compute the matrix
R.

page 41 Write the linear recurrence equation for the total number of pairs of rabbits as a two-
dimensional matrix population model.