Spatial distribution of temperature trends in Sicily

Abstract

Climate change resulting from the enhanced greenhouse effect is expected to have great impacts on hydrological cycle and consequently on ecosystems. The effects of climate variability have direct implications on water management, as water availability is related to changes in temperature and precipitation regimes. At the same time, this kind of alterations drives ecological impacts on flora and fauna. For these reasons, many studies have been carried out to investigate the existence of some tendency in temperature and/or precipitation series in different geographic domains. In order to verify the hypothesis of temperature increase in Sicily (Italy), temperature data from about 80 spatially distributed weather stations have been deeply analysed. In this study, trend of annual, seasonal and monthly temperature time series have been examined for the period 1924–2006 to investigate possible evidences of climate changes in this region. In addition, also a long series (more than 200 years) has been analysed in order to individuate possible anomalies in the 20th century and to verify the presence, in the last decades, of a temperature increase larger than in the past. The Mann–Kendall non-parametric statistical test has been used to identify trends in temperature time series data. The test has been applied at local and regional scale for three different confidence level, considering the influence of serial correlation as well. The field significance of the regional results has been evaluated using a bootstrap technique of resampling that allows to eliminate the influence of data spatial correlation on Mann–Kendall test. The application of Mann–Kendall test on temperature data provides the evidence of a general warming in Sicily during the analysed period. The analysis of the long series demonstrates that the temperature trend is mainly due to the strong rising observed in the last years of the past century. In order to determine the spatial patterns of temperature trends and identify areas with a similar temperature evolution, the detected trends have been first subjected to the spatial auto correlation analysis and then interpolated using spatial analysis techniques in a GIS framework. Temperature trend maps have allowed to argue on the risk of aridity increase, in particular in the central and western part of the island. Copyright © 2013 Royal Meteorological Society