engleski

Dynamical processes in the upper-troposphere and lee cyclogenesis in the western Mediterranean

Cyclones that appear in the Adriatic Sea basin strongly influence the climate and weather conditions in the area. A classification various types of cyclone tracks on the meso-β scale was performed based on the analysis of four years (2002 ? 2005) of operational ECMWF T511 dataset. The analysis indicates that four types of cyclones over the Adriatic Sea can be identified: (1) Type A: Genoa cyclones, with subcategories (I) continuous track and (II) discontinuous track. (2) Type B: cyclones developed in situ over the Adriatic Sea without any connections with other pre-existing cyclones in the surrounding area, with subcategories (I) northern Adriatic cyclones and (II) middle Adriatic cyclones. (3) Type AB: two cyclones co-exist and stride over the Apennines (?twin? or ?eyeglass? cyclones). (4) Type C: cyclones moving from the Mediterranean Sea, but not from the Gulf of Genoa (non-Genoa cyclones), with subcategories (I) continuous track and (II) discontinuous track. The results reveal that the greatest number of cyclones that appear in the Adriatic are initiated in the lee sides of mountain ranges in the western Mediterranean (such as the Alps, the Atlas, the Pyrenees and the Apennines), associated with an upper-level trough traversing over the mountain range. The influence of the upper-level trough on the lifecycle of the typical Mediterranean cyclone initiated to the lee of the Alps (12-15 Nov 2004) is investigated through a numerical analysis with the use of factor separation and piecewise potential vorticity inversion methods. The upper-level trough is shown to be the necessary ingredient of the event and primary deepening factor in the mature stage of cyclone development. However, the Atlas orography is the main contributor to the deepening in the first phase of deepening. Therefore, despite the differences in synoptic setting, creation of thermal anomaly and deepening rates through the generation phase, the existence of two phases of the deepening results suggest the dynamical resemblance of the cyclogenesis to the lee of the Atlas and the Alps. In addition, the analysis of the ensemble of simulations with macroscale and mesoscale perturbations to the upper-level potential vorticity anomaly suggests that cyclone track and intensity are controlled the most by the strongest mesoscale upper-level potential vorticity anomaly cores (local maximums of potential vorticity) embedded in the trough. The sensitivity of the MAP IOP 15 Genoa lee cyclogenesis event (06-09 Nov 1999) to the initial-analysis uncertainties in the upper-level precursor is numerically analyzed though the ensemble of sensitivity simulations. Modifications in the initial conditions are created with the use of 90th percentile of derived potential vorticity global model analysis error statistics, reflecting uncertainty in both intensity and position of the upper-level trough. The maximal spread of intensities in the ensemble reaches almost 50% of the cyclone intensity in the control run, and occurs during the most rapid deepening phase of the cyclone. In contrast, the strongest spread of cyclone centre positions is present in the late mature and dissipation phases, reaching up to 750 km, due to different upper-level cut-off dynamics, modified by the Alpine orography. The Bora strength is strongly sensitive to the cyclone intensity and position, but also to the properties of the background flow, with a resulting spread of wind speeds of ?30% compared to the Bora intensity in the control run. The variability of the background flow throughout the ensemble of simulations is evidenced through the (non)existence of synoptic critical levels and their varying altitudes as well as a span of the integral background Froude number over several flow regimes. Therefore, the initial-analysis uncertainties in the macroscale dynamics propagate to a range of smaller-scale phenomena in a chain of dynamically related mesoscale events, having a strong potential to decrease the mesoscale numerical forecast accuracy in the region. This suggests that the considerable improvement of the quality of the mesoscale numerical weather prediction in the area is very likely to be reached through a limited area model ensemble prediction system.

lee cyclogenesis; numrerical analysis; predictability; potential voriticty inversion

nije evidentirano