Common HIRLAM and ALADIN-M/F physics plan (08/05)

1
Common HIRLAM and ALADIN-M/F physics plan (08/05)

• J-F Geleyn, B. Hansen-Sass, G. Hello L. Rontu
• Oslo mini-workshop
• Mesoscale physics and diagnostic tools
• 12/12/2005

2
Basic guidelines

• I) We all want to develop physics towards the
  finer scales with a need to increase the realism
  of the parameterised phenomena.
• II) It is understood that each contributing group
  is also in a continuous process of improving its
  physics package. Each group will be responsible
  for maintaining the physics in a state working
  safely within the IFS coding framework.
• III) We would like to also make progress towards
  a common coding framework. This implies some
  concrete coordination efforts over the years to
  come.
• IV) In order to be able to develop the physics
  schemes with regard to physical realism,
  numerical stability, accuracy and economy it is
  essential to develop common tools for different
  types of validation.
• V) Clean inter-comparisons of different schemes,
  e.g. using common tools and rules, are a
  prerequisite for common conclusions.

3
Split of planned actions

• From longer term to short term
• Basic scientific issues
• Algorithmic issues
• Interfacing and diagnostic issues
• Transversal challenges.
• Concrete tasks only for 2005-2006
• This mini-workshop as kick-off meeting

4
Basic scientific issues (1/2)

• Selected themes
• Representation of small cumulus clouds
  (non-precipitating convection)
• Central topic (too?) many ideas.
• Convection for the scales where it still needs to
  be parameterised
• Grey zone definition work on entrainment rate.
  
• Turbulent scheme
• Attention on stable cases march towards 3rd
  order.
• Increased realism, tuning and improvement of the
  microphysics
• Validation methods how to avoid proliferation?
• Organisational aspects scientific mini-workshops
  (Tartu is a super example).

5
Basic scientific issues (2/2)

• Special theme
• Link with the surface externalisation
• Joint topic with another community, currrent
  partial lack of visibility, intrinsic complexity
  of the issue gt
• A careful scientific watch for 2006.

6
Algorithmic issues

• Selected themes
• Radiation
• Better strategy for intermittent types of
  computation.
• Better algorithmic of the microphysical
  computations (towards time steps gt 60s)
• Interplay with convection modularity.
• Turbulence (Study of staggering, fibrillation and
  implicitness issues for prognostic TKE schemes)
• Trying to sort out the CBR vs. CBR problem.
• Turbulence (3D minus 1D aspects)
• Immediate need to simply document the issue.
• Organisational aspects prepare convergence via
  some anticipating work

7
Interfacing and diagnostic issues

• Selected themes
• Choices for the non-hydrostatic aspects of the
  interfacing
• Stabilisation of the link between DDH and
  interfacing equations
• Anchor point for the development of an ambitious
  common diagnostic strategy.
• Realisation of the main interfacing routines
• Starting backwards in order to disconnect as
  long as possible from previous item link with
  SURFEX calendar constraint.
• Organisational aspects permanent WG needed (the
  work is ongoing but scattered)

8
Transversal challenges (1/2) The development of
common tools appears very important. It might
turn out that the degree of collaboration success
depends on how well such developments can be
handled in the future.

• Selected themes
• Development of common validation tools
  well-adapted to test several physical options in
  the same dynamical core
• 1D model
• ALAPIA-type model
• One (at least) 3D benchmark case.

9
Transversal challenges (2/2)

• Selected themes (continued)
• Long term aspects of the interfacing and
  portability rules
• Rules D E of the Winter 04/05 PDI document.
• Consistency of the Méso-NH and AROME/ALARO/HIRLAM
  development constraints
• Item of growing importance needing first a better
  definition.

10
Conclusion

• Let us all go back from here with a clear view
  of our common actions for 2006. The ensuing
  progress and redefinition tasks will follow
  naturally.