Spatial populations with seed-bank

In this project we analyse the effect of a seed-bank on the genetic evolution of

a genetic population subject to resampling and migration. Individuals live in colonies, labelled by the vertices of a finite connected graph, and are subject to resampling and migration as long as they are active. Resampling means that individuals adopt the type of a randomly chosen ancestor. Migration means that this ancestor may be drawn from other colonies as well.  

 Each colony has a seed-bank, into which individuals can retreat to become dormant, suspending their resampling and migration until they become active again. The goal is to analyse the equilibrium behaviour of the population, its genealogy backwards in time, as well as its multi-scale behaviour, when the graph becomes very large. In the continuum limit, when the number of individuals per colony tends to infinity, the evolution is modeled via a system of interacting diffusions. Duality properties of this system play a crucial role. The genealogy with seed-bank is drastically different from that without seed-bank. A systematic analysis of the role of the underling graph is targeted as well.