models
Differences
This shows you the differences between two versions of the page.
Both sides previous revisionPrevious revisionNext revision | Previous revisionNext revisionBoth sides next revision | ||
models [2017/09/20 16:40] – [DALES] chylik | models [2022/09/14 17:03] – [ICON] burchart | ||
---|---|---|---|
Line 9: | Line 9: | ||
The Dutch Atmospheric Large-Eddy Simulation [[https:// | The Dutch Atmospheric Large-Eddy Simulation [[https:// | ||
- | The current version of DALES ((Heus et al. (2010). // | + | The current |
+ | |||
+ | {{ : | ||
Another activity is to implement the ED(MF)< | Another activity is to implement the ED(MF)< | ||
Line 17: | Line 19: | ||
=== Regional LES and Global climate simulation === | === Regional LES and Global climate simulation === | ||
- | We use the ICON (Icosahedral non-hydrostatic) model in two different version - the LES version ICON-LEM developed during the [[http:// | + | We use the [[https:// |
Apart from its innovative triangular grid, ICON has several advantages over existing models; in particular the combination of a non-hydrostatic core with the option of heterogeneous forcing and non-periodic boundaries creates ideal opportunities for research of scale-adaptive parameterizations. The setup allows to simulate various synoptic situations at different places and a reasonable comparison to observational data - with this the testbed-situations for parameterization development is growing and getting more variable. | Apart from its innovative triangular grid, ICON has several advantages over existing models; in particular the combination of a non-hydrostatic core with the option of heterogeneous forcing and non-periodic boundaries creates ideal opportunities for research of scale-adaptive parameterizations. The setup allows to simulate various synoptic situations at different places and a reasonable comparison to observational data - with this the testbed-situations for parameterization development is growing and getting more variable. | ||
Line 29: | Line 31: | ||
- | ===== SCM ===== | + | ===== Single Column Modeling (SCM) on microgrids |
+ | |||
+ | Single Column Modeling (SCM) is a technique in which only a single vertical column from the grid of a GCM is integrated forward in time. The boundary conditions and larger-scale forcings are prescribed, usually obtained from a larger-scale model and/or observations. The suite of subgrid-scale parameterizations of the GCM is free to act, and can create their own unique model state. This can give insight into the behavior of parameterizations at process level, and can help in understanding model biases as diagnosed in a GCM. | ||
+ | |||
+ | While the highly reduced cost and enhanced transparency of SCM simulation are clear benefits, one should always be aware of its limitations. For example, the interaction with the larger-scale flow is exclusively one-way (down-scale). However, new techniques are being developed to allow the larger-scale flow to respond to the subgrid physics in SCM simulation. One option is to make use of the Weak Temperature Gradient (WTG) approximation in the tropics, in effect coupling the columns horizontally by severely constraining the horizontal temperature gradient in the temperature field. | ||
+ | |||
+ | Another approach for introducing simple feedbacks between resolved and parameterized scales is to simulate a set of neighboring single column models on a small horizontal grid. While in each column the parameterized physics acts as usual, a dynamical core can then takes care of any horizontal interactions that might develop as a result of variations among the columns. The set of columns thus become interactive, | ||
+ | |||
+ | The LES infrastructure lends itself perfectly for SCM on microgrids, as i) it already has a fully non-hydrostatic dynamical core, and ii) it can easily be run on small grids. An additional advantage in this respect is that the LES is a lot simpler in its setup compared to a GCM, which can enhance the transparency of any numerical experiments, | ||
+ | |||
+ | The SCM-on-microgrids approach for developing scale-adaptive convective parameterization has recently been pioneered by InScAPE, by implementing a multi-plume scale-adaptive version of the ED(MF)< | ||
- | {{: | ||
- | Single Column Modeling (SCM) is a technique in which only a single vertical column from the grid of a GCM is integrated forward in time. The boundary conditions and larger-scale forcings are prescribed, usually obtained from a larger-scale model and/or observations. The suite of subgrid-scale parameterizations of the GCM is free to act, and can create their own unique model state. This can give insight into the behavior of parameterization at process level, and can help in understanding model biases as diagnosed in a GCM. | ||
- | While the highly reduced cost and enhanced transparency of SCM simulation are clear benefits, one should always be aware of its limitations. For example, the interaction with the larger-scale flow is exclusively one-way (down-scale). However, new techniques are being developed to allow the larger-scale flow to respond to the subgrid physics in SCM simulation. One of these makes use of the Weak Temperature Gradient (WTG) approximation in the tropics. | ||
- | The InScAPE working group makes use of two SCM codes. One is the ICON-SCM, while the other is the SCM code of the EC-EARTH climate model. |
models.txt · Last modified: 2023/07/12 18:00 by chylik