Isotopic composition of precipitation in Northern Italy: Reverse effect of anomalous climatic events

SummaryA few years ago the first comprehensive map of the isotopic composition of precipitation in Italy was published, based on the results obtained over several years at 77 different locations in Italy. The lack of financial support made it impossible to continue this study extended to the whole country. However, the collection of composite monthly samples continued at 12 locations in Northern Italy. The main purpose was to check the variability through time of the yearly mean isotopic values calculated for some of these stations according to previous data and to add a couple of stations in the western section of Alps, along the Italy–France border for which we had no data. We report here the data obtained for the years 2002–2004. This period is particularly interesting since, during 2003, most of Europe, and particularly Italy, experienced the hottest summer of the last century. In principle, this climatic event should have determined a marked enrichment in heavy isotopes in the annual weighted means for that year. On the contrary, all but one of the collection sites yielded yearly mean isotopic values that were considerably more negative than the mean values calculated for the previous years. This is essentially due to the marked decrease in the yearly amount of precipitation and particularly in that of spring and summer so that the mean weighted yearly isotopic values are heavily affected by the amount of precipitation and the isotopic values of the winter and autumn months. This shift was particularly marked at the collection sites located along the northern slope of the Apennines where a marked “shadow” effect related to westerly air masses moving over the Apennines exists. An important aspect of these data concerns the interpretation of isotopic values of palaeoprecipitation: in the case of palaeo-isotopic changes determined by environmental conditions similar to those reported here, a “normal” interpretation of the data may be misleading.