![Hier klicken, um den Treffer aus der Auswahl zu entfernen](images/unchecked.gif) |
Titel |
Level set methods for meteorological multi scale problems |
VerfasserIn |
Eckhard Dietze, Bjorn Stevens, Juan Pedro Mellado, Norbert Peters, Heiko Schmidt |
Konferenz |
EGU General Assembly 2010
|
Medientyp |
Artikel
|
Sprache |
Englisch
|
Digitales Dokument |
PDF |
Erschienen |
In: GRA - Volume 12 (2010) |
Datensatznummer |
250040521
|
|
|
|
Zusammenfassung |
Reasonable modeling of shallow cumuliform and stratocumulus convection is important to
both weather and climate forecast. Indeed recent work suggests that shallow cumulus
convection is the major point of departure among climate-model representations of climate
sensitivity, and a principal means by which the atmospheric aerosol imprints itself on
larger-scale circulations. Unfortunately our understanding of these clouds is limited, in part,
by our ability to simulate them with great confidence or fidelity. A standard tool for
understanding both the dry and the moist boundary layer behavior is Large Eddy Simulation
(LES), but fundamental issues emerge in quantities of interest (e.g. exact boundary layer
heights or Albedo). A reason for this is that current LES cannot resolve the interface physics
due to insufficient grid resolution. Even elaborate physically based subgrid models are
numerically smeared out, so that a distinction between numerical (e.g. discretization
errors) and physical effects (e.g. entrainment) is impossible. Here, we focus on the
numerical effects and explain a heterogeneous multi scale concept for the modeling of
Stratocumulus clouds that has been successfully used in combustion science. The cloud top
interface which is driven by large scale motions is explicitly followed (tracked) using
a level set ansatz. In parallel we solve the underlying flow equations in a finite
volume sense. The level set and the flow solutions are coupled via an approach
called in-cell-reconstruction. The approach keeps an option to include small scale
processes like e.g. entrainment via (super-) parameterizations in a modular fashion. |
|
|
|
|
|