Data assimilation strategies for operational NWP at meso-scale and implication for nowcasting
Stratégies d'assimilation des données pour la prévision numérique opérationnelle du temps à méso-échelle et implication pour la prévision immédiate
Montmerle, T.
<font face="Comic Sans MS, cursive"><font style="font-size: 8pt;" size="1"><span lang="en-US"><span style="font-weight: normal;">Abstract:</span></span></font></font></p><p style="margin-bottom: 0cm" align="JUSTIFY"><font face="Comic Sans MS, cursive"><font style="font-size: 8pt" size="1"><span lang="en-US"><span style="font-weight: normal">The<br>latest developments in supercomputers and observing systems now make<br>it feasible to implement, for operational use, results from the most<br>recent research in mesoscale models physics, dynamics, and data<br>assimilation. The high spatial and temporal resolutions of these<br>models allow realistic representations of different surface features<br>or atmospheric phenomena such as convective systems or fog. Apart<br>from technical issues inherent to numerical and observational data<br>handling, the operational implementation of NWP systems at mesoscale<br>raises many issues: i) operational models representing the surface<br>conditions and the atmospheric flow at larger scales are needed in<br>order to provide boundary conditions, ii) the model state has to be<br>frequently corrected towards the latest available observations to<br>start forecasts from the best initial conditions possible. This last<br>point is the purpose of data assimilation (DA), which aims in<br>retrieving the best initial state (or analysis) from a previous<br>forecast and from various observations, the weight of these two<br>entities being given by their respective error representation.</span></span></font></font></p><br><p style="margin-bottom: 0cm" align="JUSTIFY"><font face="Comic Sans MS, cursive"><font style="font-size: 8pt" size="1"><span lang="en-US">This<br>keynote will firstly focus on the different implementation of DA<br>strategies at mesoscale that are currently considered in different<br>operational NWP centres. Pros and cons of sequential vs. variational<br>approaches will be discussed in this context. The implication that<br>such strategies could have to support nowcasting applications in<br>terms of forecasts suitability and availability will be presented. A<br>forecast becomes meteorologically relevant as soon as the<br>initialization (e.g spin-up) processes have been controlled through a<br>numerical wave filtering. The availability mainly depends on the<br>choice of the cut-off time for observations, of the DA algorithm, of<br>the cycling strategy and of the large scale forecast needed for the<br>coupling. Feedback from three years of operational use of the French<br>AROME model, that runs operationnaly with a 2.5 km horizontal<br>resolution and that makes use of an incremental 3DVar DA system, will<br>then be given, and recent works obtained in this framework will be<br>displayed to illustrate these issues.</span></font></font></p><br>
Accès à la notice sur le site du portail documentaire de Météo-France