The Piedmont flash flood occurred on 4-6 November 1994 was studied by means of:
The work highlights:
Their combined effect produced heavy precipitation phenomenon. Prefrontal event was nowcasted through METEOSAT images and hindcasted by means the thermodynamic analysis and precipitation measurements. The cells recognised above Tyrrhenian sea, because of their structure and atmospheric stability indexes, can be compared with quasi-tropical supercells. This kind of event seems to be always more frequent at mid-latitude. Therefore they should be monitored by Civil Protection to alert people even though it's difficult to define the exact coordinates (time and space) of the development of these cells.
The westerly flow present on Central Europe on Nov. 2nd is coming undulate on Nov. 3rd. A well-extended trough had its meridian axes from Greenland southward Iberica peninsula; while on Europe and Mediterranean area a ridge extended own influence from Tunisian coast up to Norway. The two axes had a meridian distance of 20\xfb only. The Eastern high pressure system produced a blocking situation with a strong warm and moist southerly advection especially in lower levels (low jet). On Nov. 3rd surface chart shows, to Southwest of Ireland, a low leading frontal system. On Nov. 4th the cold front across Spanish region, while during Nov. 5th splits in two fronts moving toward Northwest Italy. The situation in the upper level was similar to Arno flash-flood (November 1966).
Figure 1. METEOSAT images of prefrontal squall line at 13:00 UTC (a) and at 18:00 UTC (b) of 4 November 1994.
The METEOSAT images shown in Fig. 1 clearly indicate the situation. On Nov. 4th at 13:00 UTC some small cells along a meridian axis from the Tunisian coast up to the North Alpine region can be seen. At 18:00 UTC (same day) a broken squall line is clearly visible and the cells seen at 13:00 UTC are larger and quasi-stationary (there is a little North-Northeast motion). Very important was the cells above the South Piedmont-Ligurian area; during it stay (more then 6 hours), 150 mm of rain were collected at Ponzone (Tanaro basin). An other similar cell was developed and stays more than 8 hours above Northeast of Corsica island.
Figure 3. Sounding at Cagliari, 12:00 UTC (a) and at Milan 12:00 UTC (b). Dashed lines are wet temperature profiles, while dashed area is CAPE at Gagliari and warm-moist advection respect Milan.
The frontal system was splitted during Nov. 5th in two cold fronts. The first front was stronger than the second, it caused heavy precipitation on Tanaro basin. These new precipitation added to previous one producing the flash flood of Tanaro river.
The only two Italian radiosounding (Cagliari and Milan) performed along the squall line are considered. The skew-T log p diagrams and Tw of the Nov. 4th are shown in Fig. 2 where it's possible to see:
In the case studied the NWP like ECMWF and DWD pointed out a heavy precipitation due to frontal system in Northwest Italy, but they didn't well describe prefrontal precipitation. METEOSAT images permitted nowcasting following the slow parcel motion, taking in to account the slow convective instability.
In this contribution we analysed the meteorological situation that produced Piedmont 1994 flash-flood. It's remarkable that the main and unforeseen precipitation were due to a stationary cells (likely supercells) with long life that triggered along a squall line ahead the main front. During the last years well localised quasi-tropical precipitations on North Italy increased their frequency. To improve nowcasting satellite images and radar measurements should be considered. The final and key question is where and when this kind of cell trigger.
Boccolari M., P. Frontero, L. Lombroso, S. Morelli, R. Santangelo, 1992: Orographics effects on the precipitation at the border between Po valley and Apennines mountains, XXII International Conference on Alpine Meteorology, Tolosa (F), 256-260.
Frontero P. et al. 1992: An integrated system for forecasting Arno river flash floods, Natural Hazard 5, 79-192.
Giacobello N., G. Todisco, 1979: Caratteristiche Sinottiche di alcune situazioni alluvionali, Rivista di Meteorologia Aeronautica V. XXXIX no 2.
Prodi F., V. Levizzani, 1990: Previsione meteorologica con elaborazione
coordinata di immagini radar e satellite (nowcasting), Boll. Geof.
I/90.
MAP Data Centre - April '05 - MAP WebMaster