Estimates of forecast error covariances are at the heart of any data assimilation system and yet most operational assimilation schemes the forecast error covariance is assumed stationary, homogeneous, and isotropic to overcome the difficulty of estimating the full covariance matrix. These assumptions fail to characterize the true forecast error covariance. Picolo (2011) compared the growth of forecast error from covariances modeled by the Met Office operational four-dimensional variational data assimilation (4DVAR) and ensemble transform Kalman filter (ETKF) methods over a period of 24 h. The forecast error predicted by the implicit evolution of 4DVAR did not grow, regardless of the dataset used to generate the static background error covariance statics. The forecast error predicted by the ETKF grew more rapidly because the ETKF selects balanced initial perturbations, which project onto rapidly growing modes. Localization allowed within the ETKF of the Met Office Global and Regional Ensemble Prediction System (MOGREPS) enables better specification of the ensemble spread as a function of location around the globe (Bowler et al. 2009). The localization of ETKF gives a distribution of the spread as a function of latitude that better matches the forecast error of the ensemble mean. In MOGREPS, model error is represented by using stochastic-physics schemes: a random parameter scheme (RP) and a stochastic kinetic energy backscatter scheme version 2 (SKEB2, Tennant et al. 2011). The RP aims to account for the uncertainty associated with these empirical parameters, and to simulate the non-deterministic processes not explicitly accounted for by the various parameterizations (Bowler et al. 2008). The SKEB2 in MOGREPS aims to introduce some of the kinetic energy that is dissipated by the semi-Lagrangian advection scheme and injected by convective processes into the atmosphere at the mesoscale. The SKEB2 has a positive impact on the root-mean-square error of the ensemble mean and spread of the ensemble with a more realistic representation of model error (Tennant et al. 2011). In this study, we investigate the effects of local ETKF and stochastic-physics scheme in MOGREPS on forecast error covariance modeling over a period of 24 h. It is known in the literature that the localization and the model error term is one of the causes of the imbalance in the ensemble Kalman filter (Mitchell et al. 2002). Imbalances make the forecast error covariance less accurate at initial times. The evolution of the ratio of unbalanced pressure over total pressure errors at 500 hPa represents the remaining unbalanced component of the flow in the system after imposing the geostrophic balance relationship (Bannister 2008). The growth of forecast errors for temperature at 500 hPa and for zonal winds at 850 hPa and 250 hPa over 24 h are shown in addition to the ratio of unbalanced pressure over total pressure. At the initial times, the effects of localization and stochastic-physics on forecast error covariance modeling are illuminated in aspects of imbalance.