Global Modeling and Assimilation Office
NASA Goddard Space Flight Center
Last Updated: May 27 2011
Title: Use of heterogeneous background error covariances accounting for precipitations at convective scale
Authors: Thibaut Montmerle (CNRM-GAME (Meteo-France/CNRS))
Loik Berre (CNRM-GAME (Meteo-France/CNRS))
This study focuses on the simultaneous use of different background error covariances matrices that are representative of different meteorological phenomena in the incremental 3DVar of the French operational AROME model at convective scale.
At first, background error covariances have been computed for precipitating cases using forecasts provided by an AROME ensemble 3D-Var assimilation, coupled with the operational French global ensemble 4D-Var assimilation AEARP. Results obtained from 3h forecast differences performed separately for precipitating and non-precipitating columns will be discussed. Explicit convection and comprehensive microphysical parameterization explain in particular the large discrepancies in correlation lengths, error variances and in the coupling between humidity, temperature and divergence errors between the different areas.
These covariances are then used in different geographical locations, following a formulation that allows heterogeneity. This consists in expressing the analysis increment as the sum of two terms, one for precipitating and one for non-precipitating areas, making use of a mask deduced from observations. The interest of considering background error covariances and balance relationships that are representative of rainy areas for the assimilation of precipitation-related observations, e.g reflectivities and radial velocities from radars, will be discussed. This method allows more localized anisotropic increments to be retrieved in a way coherent with convection processes in precipitating areas, and spin-up is mitigated. Impacts on analyses and on forecasts will be shown in a cycled assimilation framework and compared to the operational AROME suite.
| |
Global Modeling and Assimilation Office NASA Goddard Space Flight Center |
Last Updated: May 27 2011 |