Numerical modeling of turbulent melting dynamics: application to sea-ice evolution in the Arctic
Laboratoire de Mecanique de Lille, University of Lille 1, France
Starting date: September 2014
Context, Objectives and Methodology
The extent of sea ice during summer in the Arctic Ocean is among the most sensitive indicators of the
ongoing climate change.
During the last summers satellites have recorded an exceptional reduction of the Arctic sea-ice pack to
an extent that none of the available large-scale numerical Global Climate models could predict. The
observed ice loss has lead scietists to question their quantitative understanding of the dynamical and
thermodynamical processes involved in the summer season melting and with it to reconsider several of
the simplifying assumptions on which the current large-scale computer models are based.
Features related to water ponds forming over melted ice are too small to be directly accounted for in
large-scale models. How does the heat transfer occur in the ponds and how does that affect the large-
scale sea ice mass balance? How does the melt progress on the bottom and on the lateral walls of the
ponds? How does the topography of the ponds, their surface and depth, evolve in the course of the
summer season? All the above questions have been overlooked in the present models. Melt ponds and
their surface topography are an important contributor to the surface albedo of the Arctic Ocean, which
is a key factor for the global climate. Presently, the main challenge in sea ice climate science is to
physically improve the models in order to refine their predictive power.
This research project addresses the problem of the growth process of ice melt ponds in the Arctic
during the summer season by focusing on the small-scale (~ few meters) mechanisms controlling the
evolution of the basin topography of a single melt pond. In particular we study the phenomenology of
the thermal convective flow in the pond, which is known to be highly turbulent, and its interaction with
the phase-change mechanisms at the pond boundaries.
The goal of the present project is to reach a sound understanding on how fluid dynamics and phase-
change processes contribute in determing the pond growth in order to provide useful guidelines for
parametrizations in large-scale models.
The proposed research is based on numerical simulations. A computational-fluid dynamics code is to be
developed (using the Lattice-Boltzmann Method), in order to describe both the turbulent convective
dynamics of melted water, with its inherent salt content, and the ice phase change under realistic
We shall then extend the small-scale simulation results to the medium and large scales. Our objective is
to derive simplified parameterizations suitable for the implementation in large-scale ice-ocean models
used for Earth System modelling.
Research group and funding
The project will be carried out in the Laboratoire de Mécanique de Lille (LML - Université Lille 1), with
the supervision of Prof. Enrico Calzavarini (http://lmlm6-62.univ-lille1.fr/lml/perso/calzavarini/) and
Prof. Silvia Hirata (http://lmlm6-62.univ-lille1.fr/lml/perso/hirata). The research group have a long and
significant experience in computational fluid dynamics, turbulent mixing, convective flows, and
transport phenomena through fluid and porous media.
The project will be carried out in close collaboration with Dr. Martin Vancoppenolle (CNRS), based at
LOCEAN laboratory , Université Paris 6. Dr. Vancopenolle is a geophysicist specialized in the modelling
and development of large-scale sea ice models (https://www.locean-ipsl.upmc.fr/~mvlod).
The PhD student will spend time in both labs to contribute to a shared development of a numerical tool
capable of improving the understanding of melt ponds and sea-ice evolution during the Arctic summer.
The thesis falls within the framework of the SEAS project, funded by the ANR (Agence Nationale de la
Recherche), the French National Research Agency.
The candidate should have a background in fluid mechanics, as well as some experience in numerical
methods and programming. A good knowledge of the English language will be appreciated.
How to apply
The applicant should send a motivation letter, a Curriculum Vitae, a transcript of his master grades, and
a copy of his report of final master project.