# Adaptive parameterisations

Parameterisations for sub-scale processes are indispensible parts of any weather prediction or climate model.
The term *adaptive parameterisation scheme* stands for a parameterisation scheme, which uses spatial
and temporal correlations in geophysical fields to make the parameterisations computationally more efficient
and thus to be able to include more physics in the parameterisations. We have worked on an adaptive radiative transfer scheme and are now also working on an adaptive soil module.

In an adaptive parameterisation scheme, the computation is split into a more complex, intrinsic parameterisation and a simple, adaptive extrinsic parameterisation. The scheme makes an intrinsic parameterisation at a fraction of the time steps or only in a part of the grid boxes or columns. Due to the reduced number of calls to the intrinsic parameterisation, its total computational cost goes down. As a consequence the intrinsic calculation can be made more complex and physical. To generalise the results to the full domain, an adaptive generalisation method is used that utilises the results of nearby intrinsic calculations. This adaptive generalisation may be a simple computationally light parameterisation that would lead to biased results if used by itself, i.e. without being part of an adaptive scheme. We hope that this way of thinking can make parameterisation more physical without increasing their total computational costs. The ideas might also be usefull for non-atmospheric complex models.

## Radiative transfer parameterisation

We developed two adaptive parameterisation schemes (Venema et al., 2007) for radiative transfer (RT) in the limited area model COSMO (used for short range numerical weather prediction and regional climate modelling). The temporal perturbation scheme, for instance, starts by computing the full radiation field with a delta-two-stream scheme. Then it regularly calculates the radiative changes with a simple regression scheme. Large (local) biases typical for regression schemes are avoided because the regression algorithm is only employed to calculate changes. **More information on the radiative transfer schemes.**

## Soil module

In the MiKlip program a seamless decadal climate prediction system will be developed. We are responsible for its soil module, which after the ocean represents the main memory component of the climate system. The intrinsic parameterisation will be based on a 3D hydrological model and the common land model with dynamic vegetation. Such a model can only be run in a few catchments in Germany. The extrinsic parameterization, a simple version of the CLM, will cover the full model domain. In the catchments with the intrinsic computations, the biases of the extrinsic parameterization can be assessed. **More information on the soil module.**

## Future developments

If you would like to be informed about future developments, please send me an e-mail, victor.venema@uni-bonn.de.

## References

In Venema et. al. (2007) two adaptive parametrization schemes are detailed and tested on model output. In Schomburg et. al. (2011) one of these schemes (the spatial adaptive scheme) is implemented in the COSMO model and validated on one day case studies. The UK MetOffice also developed an adaptive radiative transfer scheme; see Manners et al. (2009). The graduation thesis of Annika Schomburg provides more detail on the temporal adaptive radiative transfer scheme.

### Articles

Venema, Victor, Annika Schomburg, Felix Ament, and Clemens Simmer.
Two adaptive radiative transfer schemes for numerical weather prediction models.
*Atmospheric Chemistry and Physics, 7, 5659-5674, doi: 10.5194/acp-7-5659-2007, 2007.*

A. Schomburg, V. Venema, F. Ament, and C. Simmer.
Application of an adaptive radiative transfer scheme in a mesoscale numerical weather prediction model. *Quarterly Journal of the Royal Meteorological Society*, **138**, pp. 91–102, doi: 10.1002/qj.890, 2012.

Manners, J., J.-C. Thelen, J. Petch, P. Hill, and J.M. Edwards.
Two fast radiative transfer methods to improve the temporal sampling of clouds in numerical weather prediction and climate models.
*Q.J.R. Meteorol. Soc., 135, pp. 457-468, 2009*

Schomburg, Annika. An adaptive radiative transfer parameterisation for the Lokal-Modell. Diploma (Master) thesis, 2006