Atmospheric Physics

The physical processes associated with radiative transfer, convection, clouds, turbulent mixing, sub grid-scale orographic drag and non-orographic gravity wave drag have a strong impact on the large-scale flow of the atmosphere.  However, these mechanisms are often active at scales smaller than the resolved scales of the model grid.  The effect of sub-scale physical processes on weather systems is expressed in the model in terms of resolved model variables using parameterisation.  This involves both statistical methods and simplified mathematical-physical models, (e.g. the air closest to the earth’s surface exchanges heat with the surface through turbulent diffusion or convection, which adjusts stability in the lowest layers).

Moist Physics

An upgrade of the moist physics package (the so-called "New Moist Physics") was introduced within Cycle 47R3 (implemented 12 October 2021).  This aims to improve the representation of complicated interactions between turbulence in the lowest part of the atmosphere, convective motions and the cloud physics.  It is important that these processes interact with each other in a physically consistent way to represent real-world processes effectively.  The upgrade adjusts the following:

  • the physical representation of the convective boundary layer and modelling of Cu-topped, Sc-topped, and clear boundary layers and their transitions,
  • deep convection, cloud and precipitation in the forecast,
  • cloud amounts and humidity,
  • general precipitation through better parameterisation of hydrometeor interactions and processes,
  • the activity of the model.  The ensemble spread is increased more strongly in regions where convection dominates perturbation growth (i.e. in the northern hemisphere (in summer) and in the tropics) and here, the increase in spread persists out to day 15.

In most cases, but not all, the changes listed above manifest themselves as improvements in model output fields, but for more forecaster-oriented detail on product impacts see here.

Aspects of the moist physics package are given below:


Fig2.1.5-1: Outline of revisions to the physics within Cycle 47R3 upgrade.  Schematic of the representation of the cloudy convective boundary layer in the IFS, consisting of moist convective mass flux in the moist convection scheme and dry mass flux and turbulent diffusion in the turbulence scheme.  Turbulent diffusion is confined within the dry boundary layer.  However, it is extended to the cloud top of Sc where there is also an additional contribution from radiative cooling and cloud top entrainment. 

The Convective Boundary Layer

The turbulent and convective mixing in the convective boundary layer is formulated in a simple and consistent way but there are several important differences which have been introduced:

  • the mixed layer top is typically:
    • near the inversion top in the clear boundary layer.
    • near the cloud base for Cu-topped boundary layers.
    • near the cloud top for Sc‑topped boundary layers. 
  • the computed strength of the temperature inversion is used to distinguish between Sc and Cu topped boundary layers.
  • entrainment at the top of the mixed layer is proportional to the surface buoyancy flux (20%) for all types of boundary layers.  Additionally for Sc, there is an increased contribution to turbulent mixing from entrainment driven by radiation effects at the cloud surface.
  • all cloud processes are handled by the cloud scheme.

Saturation adjustment and the cloud scheme

The moist physics upgrade has brought:

  • a reduction in the number of overactive quasi-stationary precipitation cells.
  • a simpler and more consistent representation of cloud fraction tendencies across the model.    
  • some beneficial increases in the cloud cover in deep cloud systems and in humidity in the mid-to-upper troposphere.
  • improvement of the heat tendencies that are included in the Stochastically Perturbed Parameterisation Tendencies (SPPT) scheme resulting in an increase in the spread of the ensemble in the first few days of the forecast. 

Microphysical  processes and interaction with radiation

The moist physics upgrade improves the parameterisation of microphysical processes by:

  • introducing additional processes for:
    • the depositional growth of precipitating snow,
    • the evaporation of cloud ice,
    • the collision-collection of rain and snow particles.
  • the warm-rain collision–coalescence process at supercooled temperatures can form supercooled rain drops.  This will enables the IFS to predict freezing drizzle (but at present this is not an ECMWF product). Freezing rain is produced through a different process.

Even small changes in the processes can have significant impacts on the cloud and precipitation field and can affect shortwave and longwave radiation.

Deep convection and mesoscale convective systems

The moist physics upgrade improves the parameterisation of deep convection, especially addressing: 

  • representation of propagating mesoscale convective systems and their diurnal cycle.
  • insufficient night-time convection over land.

These include the coupling between the convection and the dynamics, which is particularly delicate in the case of mesoscale convective systems that propagate and regenerate by producing their own horizontal convergence.

New and revised products.

Cycle 47r3 included:

  • a revision of several forecast products, including an improved calculation of:
    • visibility, reducing biases in fog, rain and snow,
    • wind gusts, reducing overestimation,
    • precipitation-type, improving the diagnosis of ice pellets and freezing rain,
    • the peak wave period for ocean waves when there are multiple peaks.
  • introduction of new or improved products: 
    • MUCAPE (most-unstable convective available potential energy) which uses virtual potential temperature,
    • two new variants of mixed-layer CIN (convective inhibition) and CAPE,
    • a clear air turbulence (CAT) diagnostic.

Additional Sources of Information

(Note: In older material there may be references to issues that have subsequently been addressed)