Question

Importance of flow dependent background error covariance

Answer 1

AROME-France is a convective-scale numerical weather prediction system which has been running operationally at Météo-France since the end of 2008. It uses a 3D-Var assimilation scheme in order to determine its initial conditions. Static background-error covariances are calculated for the 3D-Var using differences between AROME forecasts from an ensemble data assimilation


Modern numerical weather prediction (NWP) systems require accurate estimates of the initial state of the atmosphere, in order to produce realistic simulations of the weather. They are produced using data assimilation (DA) systems that attempt to optimally combine two sources of information, the observations and a background.

background-error statistics Among other things, these background-error covariances depend on:

The model and its resolution. For instance, Stefănescu, et al. (2006) have compared background-error statistics for the French global model ARPEGE and its limited-area counterpart ALADIN. Brousseau, et al.(2011) have compared covariances for ALADIN and the AROME-France higher-resolution model. In both studies, differences were found in the small-scale part of the error variance spectra.

The geographical area. Error covariances differ between midlatitude, tropical (Derber and Bouttier, 1999; Montmerle, et al., 2006), and polar (Michel and Auligné, 2010) regions.