Integrated use methods of genetic and landscape graphs for the analysis of landscape connectivity
Started in october 2017
Funding: CIFRE thesis with the company ARP-Astrance, Paris
Supervisors: Jean-Christophe Foltête (directeur, HDR), UMR TheMA, Besançon ; Stéphane Garnier (co-directeur), UMR Biogéosciences, Dijon ; Hervé Moal (encadrement opérationnel), directeur innovation et partenariats, ARP-Astrance
Ecological networks are integrated in public policies aiming at preserving biodiversity and taken into account in French land planning via the "Trame Verte et Bleue (TVB)" juridic framework. Among the models developed by scientists to represent these networks, landscape graphs are a promising potential support for ecological networks analysis and managment. However, to reinforce their reliability, these methods should be better supported by field data.
To achieve this validation, population genetics is relevant because this study field relies on empirical data to identify structures resulting in part from individuals' dispersal on a time scale of several generations. Confronting these structures to landscape data, landscape genetics aims at identifying indirectly landscape role on dispersal fluxes. As this study field relies greatly on graph theory to represent links between groups of individuals, landscape graphs are tightly linked to it, but few studies have been carried out on this topic.
In that context, the thesis project objectives are to develop, experiment and compare methods allowing to couple landscape graphs-based ecological networks modelling and populations spatial and genetic structure analysis. This coupling should enable to reinforce landscape graphs ecological dimension, and therefore their aptitude to answer conservation and management operational questions. The project aims at answering the following questions:
- Do landscape graphs and genetic graphs have similarities? Under which conditions and at which scale level can landscape graphs approximate genetic spatial structures?
- How can genetic graphs be integrated into landscape graphs construction process in order to improve their reliability?
- Do the different models coupling landscape graphs and genetic graphs offer the same interest in terms of decision making to address conservation issues?
graphs, landscape, genetics, ecology, networks
Thesis advisory panel
Marie-Josée Fortin, University of Toronto
Eve Afonso, UMR Chrono-environnement
Catherine Labruère, Mathematics Institute of Burgundy
Aurélie Khimoun, UMR Biogéosciences, Dijon
Gilles Vuidel, UMR TheMA