Dynamics of concentration in a population model structured by age and a phenotypical trait - Archive ouverte HAL Access content directly
Journal Articles Acta Applicandae Mathematicae Year : 2018

Dynamics of concentration in a population model structured by age and a phenotypical trait

Abstract

We study a mathematical model describing the growth process of a population structured by age and a phenotypical trait, subject to aging, competition between individuals and rare mutations. Our goals are to describe the asymptotic behaviour of the solution to a renewal type equation, and then to derive properties that illustrate the adaptive dynamics of such a population. We begin with a simplified model by discarding the effect of mutations, which allows us to introduce the main ideas and state the full result. Then we discuss the general model and its limitations. Our approach uses the eigenelements of a formal limiting operator, that depend on the structuring variables of the model and define an effective fitness. Then we introduce a new method which reduces the convergence proof to entropy estimates rather than estimates on the constrained Hamilton-Jacobi equation. Numerical tests illustrate the theory and show the selection of a fittest trait according to the effective fitness. For the problem with mutations, an unusual Hamiltonian arises with an exponential growth, for which we show existence of a global viscosity solution, using an uncommon a priori estimate and a new uniqueness result.
Fichier principal
Vignette du fichier
age-trait-model-v2.pdf (710.14 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-01493068 , version 1 (20-03-2017)
hal-01493068 , version 2 (15-04-2020)

Identifiers

Cite

Samuel Nordmann, Benoît Perthame, Cécile Taing. Dynamics of concentration in a population model structured by age and a phenotypical trait. Acta Applicandae Mathematicae, 2018, 155 (1), pp.197-225. ⟨10.1007/s10440-017-0151-0⟩. ⟨hal-01493068v2⟩
678 View
386 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More