Stochastically Perturbed Parametrisation (SPP)

The Stochastically Perturbed Parametrisation (SPP) scheme has been developed over several years to improve characteristics of the perturbed forecasts, such as spread. The Stochastically Perturbed Parametrization (SPP) scheme, introduced in Cy49r1 in autumn 2024, replaced the previously used Stochastically Perturbed Parametrization Tendency (SPPT).   

SPP simulates model uncertainty due to physics parameterisations within the IFS ensemble system.  It randomly perturbs the values from the physical parameterisation schemes.  This is done to represent uncertainties in the effects of under-resolved processes.  These uncertainties arise from either or both:

• parameterisation scheme assumptions. These incorporate bulk descriptions of sub-grid scale processes active within an individual grid box or column.
• approximations that are necessary to describe poorly constrained processes.

The physics schemes operate through an entire grid box column.  The shape of the unperturbed column of values is preserved by multiplying by a single random number.

SPP does not perturb fluxes and this may cause certain inconsistencies and different behaviour between perturbed and unperturbed forecasts.  SPP perturbations do not explicitly depend on the current synoptic pattern.  

Stochastically Perturbed Parametrisation:

• represents uncertainties due to the model integration. 
• gives a methodology for representing model uncertainty in ensemble forecasts.
• gives a capability of reducing systematic error through the concept of nonlinear noise-induced rectification.

The same Stochastically Perturbed Parametrization configuration is used in:

• the creation of the Ensemble of Data Assimilations (EDA) as part of the representation of uncertainties in the forecast initial conditions.
• these ensemble forecast systems (medium, sub-seasonal).

The ensemble control member is unperturbed and does not use these "stochastic physics" perturbations during execution.  

The current SPP scheme generally has beneficial impacts compared to the SPPT scheme:

• on 200hPa and 925hPa.
• on synoptic scale evolution.
• on 2m temperature and 2m specific humidity (SPPT tended to focus its perturbations above the boundary layer).
• increases spread of near surface fields and precipitation (increases capture of extremes).
• decreases spread of mid/upper troposphere fields (users should asses important features).
• increased spread in tropical cyclones (beneficial for strong systems).
• positive impact in MJO at week 3-4 in monthly forecasts.  

In the past with SPPT the ensemble used to sometimes show a small risk of extreme weather beyond what is synoptically reasonable (e.g. winter maritime convective heating was dampened and some members were unrealistically cold as a result), but SPP has improved this aspect.

It should be stressed that overall stochastic perturbations undoubtedly do deliver clear improvements in particular aspects of ensemble performance.

Additional Sources of Information

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

• Read more on stochastic representation of model uncertainties.
• Read more on stochastic parameterisation and model uncertainty.
• Read more on stochastic methods for representing model uncertainties in the IFS.

(FUG Associated with Cy49r1)