You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The Lorenz 96 2-scale model was first described by Edward Lorenz during a
seminar at the European Centre for Medium-Range Weather Forecasts in the Autumn
of 1995, the proceedings of which were published as Lorenz (1996) [1] the
following year, hence the model is commonly referred to as Lorenz 96.
The model state varies on two separate time scales, one for the X dimension and
another in the Y dimension. It is constructed by coupling together two
implementations of the Lorenz 96 single-scale model. The constant F term in
Lorenz 96 single-scale model is replaced by a term that couples the two scales
together.
Lorenz 96 2-scale is a widely studied model because the differing timescales
can be viewed as an analog of processes that occur on different time and
spatial scales in the atmosphere such as large-scale flow and localized
convection. The references contain some of the earlier studies including
Palmer (2001), [2] Smith (2001), [3] Orrell (2002), [4] Orrel (2003), [5]
Vannitsem and Toth (2002), [6] Roulston and Smith (2003), [7] and Wilks
(2005). [8]
The Lorenz 96 2-scale model has a work/workshop_setup.csh script that
compiles and runs an example. This example may be explored in the
:doc:`DART tutorial <../../theory/readme>`
and is intended to provide insight into model/assimilation behavior.
The example may or may not result in good (or even decent!) results!
Development History
This DART model interface was developed by Josh Hacker as an adaptation of
the Lorenz 96 implementation. The 2-scale model is the second model
described in Lorenz (1996).
Quick Start
To run Lorenz 96 2-scale with its default settings:
Ensure you have the correct settings in mkmf.template in
<DARTROOT>/build_templates/mkmf.template
Build the DART executables using the quickbuild.sh script in the
./work directory.
Once the executables have been built, the two Perl scripts provided in the
./shell_scripts directory, spinup_model.pl and run_expt.pl, can
be used to spin up the model and run an experiment.
Namelist
The model also implements the variant of Smith (2001), which can be invoked by
setting local_y = .true. in the &model_nml namelist in the
input.nml file.
The &model_nml namelist is read from the input.nml file. Namelists
start with an ampersand & and terminate with a slash /. Character
strings that contain a / must be enclosed in quotes to prevent them from
prematurely terminating the namelist.
Palmer, Timothy N., 2001: A nonlinear dynamical perspective on model error: A proposal for non‐local stochastic‐dynamic parametrization in weather and climate prediction models. Quarterly Journal of the Royal Meteorological Society, 127, 279–304. https://doi.org/10.1002/qj.49712757202
Smith, Leonard A., 2001: Disentangling uncertainty and error: On the predictability of nonlinear systems. Nonlinear dynamics and statistics, Alistair I. Mees, Editor, Birkhauser, Boston, USA, 31–64.
Orrell, David, 2003: Model error and predictability over different timescales in the Lorenz '96 Systems. Journal of the Atmospheric Sciences, 60, 2219–2228.
Wilks, Daniel S., 2005: Effects of stochastic parametrizations in the Lorenz ’96 system. Quarterly Journal of the Royal Meteorological Society. 131. 389-407. https://doi.org/10.1256/qj.04.03