Abstract
In the last decades, the Alpine regions have registered several heavy precipitation events, occasionally in combination with high wind speed, which triggered various natural hazards, such as landslides, debris flows, flooding, and forest damage. Some of these events were found to overstrain the risk management capacity of the affected areas and caused cascading impacts and severe consequences in socio-economic systems. There is scientific evidence that overall changes in the distribution, intensity and frequency of precipitation extremes are likely to occur under a warming climate. However, the analysis of the long-term changing signal on a regional level is still challenging due to the rarity of the events and their heterogeneous spatial structure, which requires an adequate data coverage in space and time. Existing studies analysing extreme precipitation changes in the Alpine region based on in-situ observations (e.g., Scherrer et al., 2016; Zeder and Fischer, 2020) report an overall increase in extreme precipitation intensity and frequency over the last century even though with some differences in spatial and temporal distribution of resulting trends. However, only a few studies focused in detail on long-trend analysis in the north-eastern Italian Alps where Trentino South Tyrol is located (e.g., Brugnara et al., 2012), and most of them do not include the latest years, when some of the most impactful events occurred. In this context, the daily precipitation observations for more than 60 rain gauges located in Trentino South Tyrol and surrounding areas are collected, checked and used to assess the spatio-temporal variability and trends in intensity and frequency of heavy precipitation events over the period 1956-2020, as well as to relate them to large-scale temperature anomalies. Only precipitation series with less than 6 incomplete years (i.e., with more than 15 % of missing daily values) over the 65-year period are included in the analyses. The annual and seasonal 1-day precipitation maxima and the number of daily exceedances of the 97th percentile are first analysed at each station showing distinctive spatial patterns in both magnitude and seasonality. Maximum precipitation values are higher in the southern part of the region where they primarily occur during the autumn season and in areas where the influence of the Mediterranean climate is more pronounced. In the northern and more mountainous portion of the region, the seasonal occurrence of annual precipitation maxima is equally distributed between autumn and summer. Almost two thirds of the stations analysed through Theil-Sen and Mann-Kendall tests depict an increasing tendency over 1956-2020 in either the intensity of 1-day precipitation maxima or the frequency of daily precipitation exceedances, even though the resulting trends are statistically significant in less than 20 % of the locations. The greatest and most significant increases are obtained in the northern part of Trentino South Tyrol, where trends in 1-day annual precipitation maxima for some locations are above + 50 % per decade. Similar results are also obtained when precipitation maxima over longer intervals (e.g., 5-day maxima) are considered. In order to analyse the influence of large-scale thermodynamic state of the atmosphere on extreme precipitation changes and compare results with previous studies in surrounding countries, all 1956-2020 series of annual 1-day maxima are scaled with annual mean temperature anomalies (with respect to the 1961- 1990 reference period) for the northern hemisphere derived from the HadCRUT5 dataset. By considering the Clausius-Clapeyron hypothesis, the local scaling at each station site is calculated by fitting a Generalized Extreme Value (GEV) distribution to series of annual and seasonal maxima with temperature anomalies as a covariate and by assuming a linear dependency. By averaging over all stations, the mean scaling values for 1-day annual, summer and autumn maxima are about 3.8, 4.9 and 7.7 % °C-1 , respectively, that have to be compared with the theoretical value of 7 % °C -1 . Local scaling exhibits a large spread over the region with the greatest and statistically significant scaling values in the northern part. Results are overall in line with pre-existing studies in bordering countries even though differences in analysed time periods and regional features should be considered. Finally, the extreme value distributions based on the 65-year records of 1-day annual precipitation maxima are analysed at both regional and local scale. In particular, regional growth curves are derived for homogeneous sub-regions of Trentino South Tyrol that are identified through a cluster analysis of station sites based on local precipitation features. The resulting curves are thus used to assess the regional distribution of precipitation intensities associated to different return periods. By re-running the analysis for two subsequent periods (1961-1990 and 1991-2020), a preliminary evaluation of changes in statistical properties of annual maxima is performed. The extreme value analysis is also used to characterize and discuss the recorded precipitation intensities in the region during storm Vaia (October 2018), one of the most exceptional events which hit the north-eastern Alps in the last decades and caused severe short and long-term consequences. In summary, the results indicate that intensity and frequency of extreme precipitation events in Trentino South Tyrol are increasing. First outcomes represent a starting point towards a more detailed investigation of non-stationary conditions that can be included in precipitation extreme analyses in order to properly extend the evaluation to future climate scenarios. However, spatial heterogeneities – as expected given the orographic complexity of the region – have to be taken into account both in the evaluation of resulting trends and in the assessment of non-stationarity in the extreme value analysis. The contribution shows and discusses the most relevant findings and provides a final outlook towards the implementation of compound extreme analysis, e.g. heavy precipitation events combined with exceptional wind speed, and their link to impacts. The research leading to these results has received funding from Interreg Alpine Space Program 2021-27 under the project number ASP0100101, “How to adapt to changing weather eXtremes and associated compound and cascading RISKs in the context of Climate Change” (X-RISK-CC).