Seasonal forecasts at the regional scale on the basis of the continental scale average circulation

There is an increasing interest for the development of seasonal climate forecasting systems adopting methodologies ranging from complex global ocean-atmospheric models to easier statistical approaches. Relationships between the North Atlantic Oscillation (NAO) index, associated with changes in atmospheric circulation at the north-Atlantic and western Europe scale, and seasonal rainfall over much of Europe were found in many studies. In this work monthly rainfalls in some hydrological basins in Tuscany, Italy, expressed in terms of rainfall depth, are compared with previous patterns of large scale circulation, represented by the NAO index, and central Mediterranean sea surface temperature anomalies (SSTa) for the time period 1951-1991. Various statistical methods are employed, from simple linear correlation to contingency tables up to the calculation of skill scores such as POD (Probability Of Detection) and FAR (False Alarm Ratio). The aim of the work is to quantitatively define the role of NAO and SSTa as potential predictors in a regional seasonal forecasting system.