A Gap Filled Procedure for the Analysis of Frost Days in Southern Italy

The studies on the climate changes effects aim at prevent their often-harmful consequences for people and ecosystems. Therefore for this important impact, the quantity and the structure of data involved in those studies have to follow precise and well defined standards. Since long time, many countries are endowed with networks of climatic gauges to track the dynamics of climatic variables in space and time. Unfortunately, the efficiency maintenance of those networks requires a constant control of the tools that collect data to avoid measuring errors and missing data which negatively impacts on studies based on robust statistical methodologies. In order to overcome the problem of missing data, in recent times powerful statistical procedures were born. These approaches contain novel and effective methodologies that, under suitable boundary conditions, are capable to reconstruct with good reliability the set of missing data. Within this context, the aim of this study is to perform a reliable trend analysis of the minimum temperature by means of a gap-filled database of 8 temperature series collected in the Calabria region, in Southern Italy. Specifically, considering the past studies, which evidenced that winter conditions are changing more rapidly than any other season, in this study particular attention has been paid to the temporal changes of the frost days, i.e. the annual count of daily minimum temperature < 0 °C. In fact, decline in frost days could have lasting impacts on ecosystems, especially in mountainous and forested area such as the Calabria region, in which about 42% of the regional area is located over 500 m a.s.l.. Indeed, the number of frost days recorded during the winter is particularly important for the health of mountain forests (for example by killing parasites and harmful insects, or mitigating summer water stress) and for the conservation of the snowpack, which insulates the soil, providing subnivean shelter to animals and prevents freezing of roots and microorganisms. With this aim, in the present work, a statistic-probabilistic methodology has been presented and applied to select the most effective data reconstruction method among those known from current literature. After the missing data reconstruction, the Mann-Kendall non-parametric test has been applied evidencing a declining trend of frost day’s number for six out of eight series, thus confirming an overall rise of the minimum temperature in the study area