# Changeset 62 for palm/trunk/DOC/app/chapter_2.0.html

Ignore:
Timestamp:
Mar 13, 2007 2:52:40 AM (15 years ago)
Message:

Id string added to all html files

File:
1 edited

Unmodified
Removed
• ## palm/trunk/DOC/app/chapter_2.0.html

 r54 PALM chapter 2.0

2.0 Basic techniques of PALM chapter 2.0

2.0 Basic techniques of the LES model and its parallelization

LES models generally permit the

LES models generally permit the simulation of turbulent flows, whereby those eddies, that carry the main energy are resolved by the numerical grid. Only the simulated directly (they are explicitly resolved) and their effects are represented by the advection terms.

PALM is based on the

PALM is based on the non-hydrostatic incompressible Boussinesq equations. It contains a water cycle with cloud formation and takes into account infrared radiative cooling in cloudy conditions. The model has six prognostic quantities in total  u,v,w, liquid water potential temperature Θl (BETTS, quantities in total – u,v,w, liquid water potential temperature Θl (BETTS, 1973), total water content q and subgrid-scale turbulent kinetic energy e. The 1993). The water cycle is closed by using a simplified version of KESSLERs scheme (KESSLER, 1965; 1969) to parameterize precipitation processes (MÜLLER and CHLOND, 1996). Incompressibility is processes (MÜLLER and CHLOND, 1996). Incompressibility is applied by means of a Poisson equation for pressure, which is solved with a direct method (SCHUMANN and SWEET, 1988). The Poisson equation horizontal directions. At the lower surface, either temperature/ humidity or their respective fluxes can be prescribed.

The advection terms are treated by the scheme proposed by PIACSEK and WILLIAMS (1970), which conserves the integral of linear and quadratic quantities up to with the third-order Runge-Kutta scheme. A second-order Runge-Kutta scheme, a leapfrog scheme and an Euler scheme are also implemented.

By default, the time step is computed

By default, the time step is computed with respect to the different criteria (CFL, diffusion) and adapted automatically. In case of a non-zero geostrophic wind the coordinate system can be moved along with the mean wind in order to maximize the time step (Galilei-Transformation).

In principle a model

In principle a model run is carried out in the following way: After reading the control parameters given by the user, all prognostic variables are corrected with the help of the pressure solver. Following this, all diagnostic turbulence quantities including possible Prandtl-layerquantities are computed. At the end of a time Prandtl-layer–quantities are computed. At the end of a time step the data output requested by the user is made (e.g. statistic of analyses for control purposes or profiles and/or graphics data). If the given end-time was reached, binary data maybe be saved for restart.

The model is based

The model is based on the originally non-parallel LES model which has been operated at the institute since 1989 Users can choose between a two- and a one-dimensional domain decomposition. A 1D-decomposition is preferred on machines with a slow  network interconnection. In case of a 1D-decomposition, the slow  network interconnection. In case of a 1D-decomposition, the grid points along x direction are distributed among the individual processors, but in y- and z-direction all respective grid points belong to the same PE.

The calculation of central differences or

The calculation of central differences or non-local arithmetic operations (e.g. global sums, FFT) demands communication and an appropriate data exchange y-direction, the data which lie distributed on the individual central processing elements, have to be collected and/or relocated before. This happens by means of the routine MPI_ALLTOALLV. Certain This happens by means of the routine MPI_ALLTOALLV. Certain global operations like e.g. the search for absolute maxima or minima within the 3D-arrays likewise require the employment of special MPI routines (MPI_ALLREDUCE).

Further details of the internal model

Further details of the internal model structure are described in the technical/numerical documentation.

Last change: 14/04/05 (SR)

Last change: \$Id\$

Note: See TracChangeset for help on using the changeset viewer.