Potential of sequential EnKF for the short-range prediction of a maritime severe weather event

The Western Mediterranean coastlands are persistently affected by severe phenomena related to maritime convective systems. Areas with low density of observations around highly populated regions pose serious forecasting challenges due to the risk of misrepresenting crucial structures. This forecast problem is exemplified by the squall line that affected Mallorca (Spain) on 4th October 2007. Ensemble Kalman Filter (EnKF) assimilation algorithms exploit the statistical information conveyed by ensembles and are specially suited for regions with poor knowledge about climatological error statistics and covariances. We investigate the potential for predictability improvement from the assimilation of standard observations in the squall line event. Ensemble forecasts are assessed in terms of probabilistic products which clearly bring out the differences between assimilation and control experiments. Results show the large improvetnents rendered by the EnKF system in terms of severe weather threat. The attribution of these improvements is discussed in terms of the environmental ingredients linked to squall line formation. Experiments reveal that forecast improvements are fully attributable to the ability of EnKF to accurately represent the convergent flow over the Alboran Sea responsible for the thunderstorm, initiation. Additional sensitivity experiments are performed to confirm the hypothesised primary role of the terrestrial observations in the accurate representation of the low-level convergent flow. These experiments confirm the ability of the sequential assimilation system in conveying crucial observational information from terrestrial to marine areas, and thus bestowing the EnKF a central role in future upgrades of high impact weather prediction systems in the Western Mediterranean region. (C) 2016 Elsevier B.V. All rights reserved.