Key points are not available for this paper at this time.
Abstract Regional studies are crucial for monitoring and managing the impacts of extreme climatic events. This phenomenon is particularly important in some areas, such as the Mediterranean region, which has been identified as one of the most responsive regions to climate change. In this regard, the analysis of large space-time sets of climatic data can provide potentially valuable information, although the datasets are commonly affected by the issue of missing data. This approach can significantly reduce the reliability of inferences derived from space-time data analysis. Consequently, the selection of an effective missing data recovery method is crucial since a poor dataset reconstruction could lead to misleading the decision makers’ judgments. In the present paper, a methodology that can enhance the confidence of the statistical analysis performed on the reconstructed data is presented. The basic assumption of the proposed methodology is that missing data within certain percentages cannot significantly change the shape or parameters of the complete data distribution. Therefore, by applying several missing data recovery methods whose reconstructed dataset better overlaps the original dataset, larger confidence is needed. After the gap filling procedure, the temporal tendencies of the annual daily minimum temperature (T < 0 °C) were analysed in the Calabria region (southern Italy) by applying a test for trend detection to 8 temperature series over a 30-year period (1990–2019). The results showed that there was a constant reduction in the duration of frosty days, indicating the reliability of the effect of climate change.
Barca et al. (Sat,) studied this question.