Urban heat island bias in Italian air temperature series

A new data-set of Italian monthly minimum and maximum temperature series for the period 1865-1996 has recently been set up and analysed for trend within the project "Reconstruction of the climate of the past in the Mediterranean area" of the National Research Council (CNR). The project allowed the completion and updating of the Ufficio Centrale di Ecologia Agraria (UCEA) secular series data-set that was set up in the '70s. Although the data were tested for homogeneity and sometimes homogenised, the presence of some urban series may cause urban heat islands to partially affect the observed trends. In this framework the aim of this paper is the analysis of the effects of this possible bias in the period 1951-1996 in regional average series calculated for two different geographic areas: Northern and Central-Southern Italy. The analysis is performed comparing Northern and Central-Southern seasonal average temperature anomalies calculated using the UCEA-CNR data-set with the same series calculated using the Italian Air Force data-set that includes only rural and airport stations.