 |
|
|
Projects:
- (a) Multiscale Methods for Near Surface Vertical Mixing
Processes in the Ocean (Dipl.-Math. Dipl.-Inf.
Madlen Kimmritz) In cooperation with the Excellence Cluster
"The Future Ocean"
The biotic uptake of carbon of the ocean crucially depends on near surface
vertical mixing processes. Vertical mixing transports the essential nutrients
from the aphotic to the euphotic zone where photosynthesis can take place.
This upward flux of nutrients is (in steady state) in turn balanced by
the export of organic materials down to the aphotic zone, i.e. it is directly
related to the biotically induced carbon drawdown in the ocean (the
biological carbon pump).
In ocean models these processes have to be parameterized since not all spatial
scales can be resolved. These parametrizations are still a source of large
uncertainties concerning the carbon uptake of the oceans.
In this project, we aim to improve the parameterization of near surface
vertical mixing processes. We propose to apply a 3-D non-hydrostatic
Large-Eddy Simulation (LES) model to the surface ocean to explicitly resolve
spatial scales ranging from 500\,m down to 1\,m. The results will be
compared with different state-of-the-art parameterizations of vertical mixing
as used in the existing global ocean circulation models (e.g. Kiel Climate
Model, FLAME). The intention is to optimize parameters associated with the
parameterizations of vertical mixing and, subsequently, to examine the
sensitivity of modeled carbon uptake (as modeled with an ecosystem coupled to
a global ocean circulation model) on the optimized parameterizations.
- (b) Model Adaptivity for Atmospheric Transport Models
(Dipl.-Math. Nico Taschenberger,
DFG priority program SPP 1276 MetStroem)
Numerical simulation of the transport and chemical
reactions of pollutants in the atmoshere demands an extremly high
numerical costs, because of the large number of chemical species
(e.g. 58 species ans 201 reactions for a mesoscale chemical transport
model). In this project the separation of scales of the processes
should be used in order to perform an error-controlled dimension reduction.
The error control is based on adjoint equation which allows to measure
the effect of modelling and discretization error with respect to
user-defined output funtionals. Possible functionals are e.g. maximal concentrations
of ozone or nitrogen oxides, since these are important quantities for
the numerical simulation.
- (c) Flow Control with Stabilized Finite Elements (Dipl.-Math. Benjamin Tews,
DFG priority program
SPP 1253)
Due to the high numerical costs for the numerical solution
of optimization problems with partial differential equations (PDE),
the discretization is of extreme importance. The investigation and analysis of
discretization schemes for optimization problems with
flow problems is still in its infancy. The topic of this project is the
development and analysis of stabilized finite element schemes in connection
with optimization problems.
- (d) Simulation of Thermohaline Convection in the Ocean's Crust
with Adaptive Finite Elements
(Dr. Jaime Carpio, Post-Doc project of the Excellence Cluster
"The Future Ocean"
2008-2009)
Phd position: Numerical simulation is an indispensable tool of investigation
for the understanding of the physical and chemical processes involved in
the formation of seafloor
resources, such as polymetallic sulfides and gas hydrates.
State-of-the-art simulations of seafloor resources are in 2-D only and
without the possibility to perform local mesh refinement. Due to the
hetereogenities in the subsurface more flexible simulation tools
are needed.
Hence the development of advanced numerical
techniques for obtaining accurate
and robust numerical solutions for this type of partial differential equations
is a crucial aspect and must be
addressed.
In this project we target the development of numerical software for
the 3-D simulation of thermohaline convection in the ocean's crust
with adaptive finite elements, integrating chemical and biological reactions.
|
|
|
