Chapter 1. Overview

IFS documentation Front Page

Table of contents

Chapter 1. Overview

Chapter 2. Radiation

Chapter 3. Turbulent diffusion and interactions with the surface

Chapter 4. Subgrid-scale orographic drag

Chapter 5. Convection

Chapter 6. Clouds and large-scale precipitation

Chapter 7. Land suface parametrization

Chapter 8. Methane oxidation

Chapter 9. Climatological data

REFERENCES

Next Section
Previous Section

1.1 Introduction

Figure 1.1 Schematic diagram of the different physical processes represented in the IFS model.

The physical processes associated with radiative transfer, turbulent mixing, subgrid-scale orographic drag, moist convection, clouds and surface/soil processes have a strong impact on the large scale flow of the atmosphere. However, these mechanisms are often active at scales smaller than the horizontal grid size. Parametrization schemes are then necessary in order to properly describe the impact of these subgrid-scale mechanisms on the large scale flow of the atmosphere. In other words the ensemble effect of the subgrid-scale processes has to be formulated in terms of the resolved grid-scale variables. Furthermore, forecast weather parameters, such as two-metre temperature, precipitation and cloud cover, are computed by the physical parametrization part of the model.

This part (Part IV `Physical processes') of the IFS documentation describes only the physical parametrization package. After all the explicit dynamical computations per time-step are performed, the physics parametrization package is called by the IFS. The physics computations are performed only in the vertical. The input information for the physics consists of the values of the mean prognostic variables (wind components, temperature, specific humidity, liquid/ice water content and cloud fraction), the provisional dynamical tendencies for the same variables and various surface fields, both fixed and variable.

The time integration of the physics is based on the following:

1) it has to be compatible with the adiabatic part of the IFS;

2) the tendencies from the different physical processes are computed in separate routines;

3) as a general approach, the value of a prognostic variable is updated with the tendency from one process and the next process starts from this updated value, in what is usually referred to as the `method of fractional steps' (details are different for different processes);

4) explicit schemes are used whenever possible, but if there are numerical stability problems the scheme is made as implicit as necessary.

The radiation scheme is described in Chapter 2 `Radiation' and is the first process to be called in the physics. To save time in the rather expensive radiation computations, the full radiation part of the scheme is currently called every 3 hours. This is when the computation of the shortwave transmissivities and the longwave fluxes is performed, using the values of temperature, specific humidity, liquid/ice water content and cloud fraction at time-step

, and a climatology for aerosols, CO₂ and O₃. The computation of the fluxes is not necessarily done at every grid-point but is only performed at sampled points, using a sampling algorithm that is latitude dependent. The results are then interpolated back to the original grid using a cubic interpolation algorithm. The shortwave fluxes are updated every time-step using synchronous values of the zenith angle. The radiation scheme takes into account cloud- radiation interactions in detail by using the values of cloud fraction and liquid/ice water content, at every level, from the prognostic cloud scheme. The radiation scheme produces tendencies of temperature.

The turbulent diffusion scheme is called just after radiation (Chapter 3 `Turbulent diffusion and interactions with the surface' ). The surface fluxes are computed using Monin-Obukhov similarity theory. The computation of the upper-air turbulent fluxes is based on the

-diffusivity concept. Depending on the atmospheric stability different formulations for determining the

-coefficients are used: a

-profile closure for the unstable boundary layer and a -number dependent closure for the stable boundary layer. Because of numerical stability problems the integration of the diffusion equation is performed in an implicit manner. In fact, it uses a so-called `more than implicit' method, in which the `implicitness factor'

(which takes the value 0 in a fully explicit scheme and 1 in a fully implicit one) is set to 1.5. During the integration it uses the values of the prognostic variables at

to compute the

-coefficients but uses the tendencies updated by the dynamics and radiation on the right hand side of the discretized diffusion equation. The turbulent diffusion scheme also predicts the skin temperature and the apparent surface humidity. The turbulent diffusion scheme produces tendencies of temperature, specific humidity and wind components. It does not compute fluxes or tendencies of the cloud variables (liquid/ice water content and cloud fraction).

The subgrid-scale orographic drag scheme is called after the turbulent diffusion and is described in Chapter 4 `Subgrid-scale orographic drag' . The subgrid-scale orographic drag parametrization represents the low- level blocking effects of subgrid-scale orography and the transports due to subgrid-scale gravity waves that are excited when stably stratified flow interacts with the orography. Numerically the scheme is similar to the turbulent diffusion and also requires an implicit treatment. In this case the `implicitness factor'

is set to 1. The subgrid-scale orographic drag scheme produces tendencies of the wind components and temperature.

The moist convection scheme is described in Chapter 5 `Convection' . The scheme is based on the mass-flux approach and is divided in deep, mid-level and shallow convection. For deep convection the convective mass-flux is determined by assuming Convective Available Potential Energy (CAPE) is adjusted towards zero over a specified time-scale. For mid-level convection the cloud base mass-flux is directly related to the large scale vertical velocity. The intensity of shallow convection is estimated by assuming an equilibrium of moist static energy in the sub- cloud layer. The convection scheme provides tendencies of temperature, specific humidity and wind components.

In Chapter 6 `Clouds and large-scale precipitation' the prognostic cloud scheme is described. It solves two prognostic equations for liquid/ice water content and cloud fraction. The cloud scheme represents the cloud formation by cumulus convection, the formation of boundary layer and stratiform clouds. The scheme also takes into account several important cloud processes like cloud-top entrainment, precipitation of water and ice and evaporation of precipitation. In the numerical integration of the equations the terms depending linearly on the values of liquid/ice water and cloud fraction are integrated analytically. The cloud scheme produces tendencies of all the prognostic variables.

The soil/surface scheme is described in Chapter 7 `Surface parametrization' . The scheme includes prognostic equations for temperature and moisture in four soil layers and snow mass. The soil equations use an implicit time integration scheme. An interception layer collects water from precipitation and dew fall. The evaporative fluxes consider separately the fractional contributions from snow cover, wet and dry vegetation and bare soil.

Chapter 8 `Methane oxidation' describes a simple parametrization of the upper-stratospheric moisture source due to methane oxidation. A parametrization representing photolysis of vapour in the mesosphere is also included.

Chapter 9 `Climatological data' describes the distributions of climatological fields.

Next Section
Previous Section

05.04.2002