Date : 19/11/2010

Internship proposal for : Master 1 or Master 2

Laboratory

LCVN
UMR 5587 CNRS, Université Montpellier 2
place EugËne Bataillon 34095 Montpellier
Director : Jean-Louis Sauvajol
Website : http://www.lcvn.univ-montp2.fr
Main discipline : Molecular biologyPhysics

Supervisor

Estelle Pitard
email : Cet e-mail est protégé contre les robots collecteurs de mails, votre navigateur doit accepter le Javascript pour le voir
phone : +33 467144933

Subjects / Tools-Methodologies

1 : statistical physics/analytic -numerical simulations
2 : ecology - metapopulations/The LCVN (Laboratoire des Verres, Colloides et Nanomateriaux) is a large physics laboratory, with experimental groups working on nanomaterials and soft condensed matter. The theory group has a wide interest in out-of-equilibrium systems, including glasses, gels, foams, proteins. Several collaborations with biologists have started and the need and interest for such collaborations grow rapidly: they include studies of molecular transport, genetic networks, ecology and evolution dynamics.
3 : analytic -numerical simulations/Public†: students in a Physics master, with a good knowledge in statistical physics. Interest and background in both physics and biology, or physics and complex systems are particularly encouraged.

Supervisers : Estelle Pitard (LCVN, Montpellier, 04 67 14 49 33, Cet e-mail est protégé contre les robots collecteurs de mails, votre navigateur doit accepter le Javascript pour le voir ) and François Munoz (Botanique et Bioinformatique de l\'architecture des plantes -AMAP-, 04 67 61 49 07, Montpellier, Cet e-mail est protégé contre les robots collecteurs de mails, votre navigateur doit accepter le Javascript pour le voir ) The student will be located in the Laboratoire des ColloÔdes, Verres et Nanomatériaux (LCVN, Montpellier) and will interact both with the physicist (E.P.) and the ecologist (F.M.). Context and questions : In ecology, individuals of same species form discrete population in space, that can grow or disappear due to migration and extinction. Such a dynamics can be studied in the framework of metapopulation theory (Hanski and Gilpin 1997, Levins 1969). This theory can predict the density and structure of populations at equilibrium starting from non-linear dynamical equations. The spatio-temporal behaviour of these equations is however largely unknown, although their outcome would be very important for conservation biology in particular. From a physicist point of view, a simplified view is to oppose two different approaches. Spatially implicit metapopulation models (Levins 1969) assume that migration and extinction processes are independent of space -this is a mean field approach-, while spatially explicit approaches involve the influence of the substrate, and the "niche" which is predefined to be a favorable site (Hanski 1998). The main challenge is to reconcile these two approaches. Recently, a few studies have tried to make a link between these ecological questions, and models of statistical physics and reaction-diffusion approaches. For example, Houchmandzadeh (2009) has shown, using a simple model and a simple experimental approach on populations of bacteria, that creation and annihilation processes could give rise to non trivial spatial structures, which can be precisely quantified. In collaboration with ecologists, on can try to extend this type of models to more realistic conditions. For example, one can introduce kinetic constraints that result from steric hindrance or jamming, like what happens in glasses (Ritort and Sollich 2003). Another idea is to add a favorable substrate which has the structure of a percolation cluster (this is compatible with ecological data), and study the metapopulation dynamics on such a substrate. One can also generalize Houchmandzadeh\'s models by introducing disorder in certain parameters for the dynamics. Goals : Using numerical simulations and analytical tools, the student will investigate in detail the results of Houchmandzadeh (2009) and generalized models on the spatial structure of populations resulting from reaction-diffusion processes. Some bibliography work will also be done. The models will be compared to those developed recently in ecology (Hanski and Gilpin 1997, Hiebeler 2000, Munoz 2009, Munoz, et al. 2007). One important aspect will be the choice and discussion of the statistical measures relevant to caracterize the emerging spatial properties: pair correlation function, structure factor, variance to mean ratio, etc..This characterization is essential in ecology (Tilman and Kareiva 1997) but is far from being well defined and is not unified (Holyoak and Ray 1999). By comparing the different models, the goal will be to be able to distinguish the different effects of the environment on the spatial structures, compared to a well-defined reference model. PhD: This internship is part of an interdisciplinary research project associating physics and ecology. PhD grants may be available in 2011. References Hanski, I. A. and Gilpin, M. E. 1997. Metapopulation Biology. - In, Academic Press, pp. 525. Levins, R. 1969. Some demographic and genetic consequences of environmental heterogeneity for biological control. - Bulletin of the Entomological Society of America 15: 237-240. Hanski, I. 1998. Metapopulation dynamics.