Atmosphere Overview

The ICON atmosphere model predicts the spatio-temporal evolution of the atmospheric state in terms of the prognostic variables virtual potential temperature, 3D wind, total air density and mass fractions of atmospheric water constituents and trace gases. In addition, the model provides a comprehensive set of diagnostic quantities, such as surface pressure, wind gusts or potential vorticity just to name a few. An extensive, but still incomplete list of available output variables is provided in Appendix A of the ICON Tutorial 2024.

In mathematical terms, the ICON atmosphere model solves the fully compressible non-hydrostatic Navier-Stokes equations on the sphere. The explicitly resolved scales of motion are treated by the so called Dynamical Core. The latter is accompanyied by a set of physical parameterizations which account for the effect motions that fall below a chosen mesh size. ICON offers two different physics packages which are known as the AES Physics Package and the NWP Physics Package.

The AES Physics Package was originally derived from the physics package of the ECHAM model and subsequently further developed. Applications range from general circulation model on long time scales to km-scale, storm-resolving climate simulations. The AES Physics Package can be chosen by setting the namelist parameter iforcing=2.

The NWP Physics Package parameterizations were chosen from different sources, most notably the COSMO-Model and the IFS model. Originally developed for numerical weather prediction, the NWP Physics Package was extended in the framework of the ICON-XPP project for seamless application across scales. The NWP Physics Package can be activated by setting iforcing=3.

Dynamical Core

The dynamical core can be considered as the foundations of any numerical model. It predicts the evolution in space and time of all atmospheric motions which are resolvable on a given mesh. To this end, the dynamical core solves the Euler equations which is a set of partial differential equations describing adiabatic and inviscid flow. There exist various approximative forms of the Euler equations in which certain types of motion are filtered out that are difficult to handle numerically. Well known examples are the hydrostatic form or the Boussinesq form, both of them not supporting the propagation of acoustic waves.

The ICON dynamical core solves the fully compressible form of the Euler equations, which does support acoustic waves. Notable approximations to the exact Euler equations relate to the treatment of the Earth as a spherical geoid and the shallow atmosphere approximation (see e.g. Thuburn & White 2013). The latter may be deactivated with the Namelist switch ldeepatmo=.TRUE., leading to the so called deep atmosphere form of the governing equations (Borchert et al. 2019). Given a suitable set of physical parameterizations, this allows for simulations on model domains reaching all the way up to the lower thermosphere.

The governing set of equations is discretized on an Icosahedral-triangular C-grid in horizontal directions, while in vertical direction a height-based terrain following coordinate with Lorenz-type staggering of the prognostic variables is used. The discrete numerical operators, such as the divergence, gradient or laplace operator, are constructed using a mixture of finite difference and finite volume discretizations of mostly second order-accuracy. They are combined with a predictor-corrector type two-level time integration scheme, leading to a discretization which is mass conserving, but not strictly energy conserving.

For additional details on the dynamical core, the reader is referred to Zaengl et al. 2015 and Chapter 3 of the ICON Tutorial 2024.

Tracer Transport

The tracer transport module is an important building block of any weather prediction or climate model. It solves the tracer mass continuity equation, in order to describe the redistribution of gaseous, liquid or solid atmospheric constituents such as water vapour or rain water due to air motion or gravitational settling (sedimentation).

In ICON, finite volume methods of second order accuracy in time and up to fourth order accuracy in space are applied to construct mass conserving and mass consistent transport schemes. If needed, these schemes can be combined with monotonicity or positivity preserving limiters.

More details on the tracer transport module can be found in the ICON reports (Reinert 2020, Reinert & Zaengl 2021) and Section 3.6 of the ICON Tutorial 2024.

Physical Parameterizations

AES Physics Package

to be added

NWP Physics Package

Overview NWP physics package

Parameterization	References	Namelist Parameter
Radiation	RRTM (Mlawer et al. 1997, Barker et al. 2003), ecRad (Hogan & Bozzo 2018)	inwp_radiation
Non-Orographic Gravity Wave Drag	Orr et al. 2010	inwp_gwd
Sub-grid scale Orographic Drag	Lott & Miller 1997	inwp_sso
Cloud Cover	-	inwp_cldcover
Microphysics	Single Moment (Doms et al. 2011), Double Moment (Seifert & Beheng 2006), SBM (Khain & Sednev 1996, Khain et al. 2004)	inwp_gscp
Convection	Tiedtke 1989, Bechtold et al. 2008	inwp_convection
Turbulent Transfer	Prognostic TKE (Raschendorfer 2001), 3D Smagorinsky (Smagorinsky 1963, Lilly 1962)	inwp_turb
Land	See Land Parameterizations	inwp_surface

to be extended

You can find a brief overview on the NWP physics package in chapter 3 of the ICON Tutorial 2024.

Waves (NWP)

to be added

Glossary of Namelist Parameters

Operational NWP setting marked by

iforcing

(&run_nml) Forcing of dynamics and transport by parameterized processes. 2: AES forcing, 3: NWP forcing

ldeepatmo

(dynamics_nml) Switch for deep-atmosphere modification of non-hydrostatic atmosphere. Specific settings can be found in &upatmo_nml.

inwp_radiation

(&nwp_phy_nml) Radiation parameterization. 1: RRTM radiation, 4: ecRad radiation

inwp_gwd

(&nwp_phy_nml) 1: Orr et al. scheme

inwp_sso

(&nwp_phy_nml) 1: Lott-Miller scheme

inwp_cldcover

(&nwp_phy_nml) 1: Diagnostic PDF 5: All or nothing scheme (grid-scale clouds)

inwp_gscp

(&nwp_phy_nml) 1: Single moment 2: Single moment incl. graupel 4: Double moment 8: Warm spectral bin microphysics

inwp_convection

(&nwp_phy_nml) 1: Tiedtke-Bechtold

inwp_turb

(&nwp_phy_nml) 1: Prognostic TKE (COSMO) 5: 3D Smagorinsky diffusion

inwp_surface

(&nwp_phy_nml) 1: TERRA