El Nino spielt mit dem Hagel. Rolle des Klimawandels unklar

Unser Thema heute: Der Hagel. Was macht der Klimawandel mit dem Hagel? Wird er mehr, wird er weniger? Wir schauen auf die Statistiken und betreiben Ursachenforschung.

Australien
Soderholm et al. 2017 analysierten die Hagelhäufigkeit für die letzten 18 Jahre und fanden einen Zusammenhang mit dem El Nino, dem meteorologischen Christkind.

USA
Auch in den USA wird der Hagel vom El Nino beeinflusst. Allen et al. 2015 haben die Zusammenhänge genau angeschaut und sehen gute Prognosemöglichkeiten für den Hagel in den USA auf Basis der El Nino-Entwicklung. Siehe Pressemitteilung der Columbia University hier (oder hier auf Science Daily). Weitere Quellen zum Hagel in den USA: Das SPPI stellte 2012 einige Arbeiten zu Hageltrends in den USA vor (pdf hier).  Siehe auch Bericht von Stanley Changnon. Die Webseite hailtrends.com scheint auch Daten für die USA vorzuhalten, man muss sich aber zunächst (kostenfrei) registerieren.

Es ist bisher kein robuster Zusammenhang zwischen Hagel und Klimawandel belegbar. Tippett et al. 2015:

Climate and Hazardous Convective Weather
Substantial progress has been made recently relating the large-scale climate system and hazardous convective weather (HCW; tornadoes, hail, and damaging wind), particularly over the USA where there are large societal impacts and a long observational record. Despite observational data limitations, HCW has shown to be influenced by the climate system and the tropical atmosphere via the Madden-Julian Oscillation and El Niño-Southern Oscillation. Analysis of the atmospheric environments favorable to HCW (e.g., convective available potential energy and vertical wind shear) avoids observational and model limitations. While few robust trends are seen over recent decades, future climate projections indicate increased frequency of such environments over the USA, Europe, and Australia, suggesting increased future HCW activity. A recent increase in the year-to-year variability of US tornado occurrence is striking, but not yet understood. Dynamical downscaling to convection-permitting resolutions promises improved understanding of the relationships between large-scale climate and HCW occurrence.

In der Arbeit selber machen die Autoren wenig Hoffnung, dass der Hagel in absehbarer Zukunft verstanden werden könnte. Es gibt offenbar keine verlässlichen Registriermethoden, die flächendeckend den Hagel erfassen könnten:

“The primary difficulty in relating hazardous convective weather (HCW; tornadoes, hail, and damaging winds) with large-scale climate signals is the lack of high-quality observations of HCW. There is no foreseeable solution to this problem, and current observation collection practices remain inadequate. The US HCW report database contains poorly characterized, non-physical variability in time and space which is substantially greater than that expected from climate sources. Other HCW databases have more serious deficiencies.”

Gensini & Allen 2018 meinen für den Hagel in den USA zudem einen Zusammenhang mit den Windregimen gefunden zu haben, dem sogenannten Index der Global Wind Oscillation.

 

Europa
Die schlechte Beobachtungsdatenbasis ist natürlich auch in Europa ein Problem, wie Mohr et al. 2015 zu bedenken geben. Die Autoren entwickelten daher ein Modell, das jedoch für die vergangenen 60 Jahre keinen Trend finden kann:

Hail potential in Europe based on a regional climate model hindcast
Due to the local-scale nature of hail and a lack of appropriate observation systems, comprehensive, reliable, and consistent information about hail frequency and intensity in Europe is not available. To overcome this constraint, we developed a logistic hail model that quantifies the potential of the atmosphere to form hailstorms. The model is based on a combination of appropriate hail-relevant meteorological parameters. This paper presents the application of an adjusted version of the logistic model with the objective being to estimate the hail potential across Europe based on dynamically downscaled National Centers for Environmental Prediction/National Center for Atmospheric Research1 reanalysis over a long-term period of 60 years (1951–2010). The model output, in terms of the potential hail index (PHI), identified several hot spots that are well known from other observational studies. Time series of the PHI over the 60 year period show a high correlation at different sites across Europe and high annual and multiannual variability, but no overall trend.

Brazdil et al. 2016 haben die Hageltrends für die Tschechische Republik für die letzten 600 Jahre beschrieben. Richtig robust wird die Statistik erst ab 1925. Insgesamt hat die Hagelhäufigkeit in den letzten 100 Jahren in der Tschechischen Republik offenbar abgenommen:

A long-term chronology of summer half-year hailstorms for South Moravia, Czech Republic
Climatological analyses of hailstorms, as phenomena of local or regional occurrence with associated damage, depend strongly on the quality and density of meteorological observations. Documentary sources, both historical and modern, including insurance company records, can be used to complement existing meteorological data or extend them into the period prior to continuous meteorological observations. This paper employs such aids to compile a long-term hailstorm chronology for the summer half-year (April-September) in South Moravia (Czech Republic) based on derivations from various types of documentary evidence together with systematic meteorological records. Although the first single hailstorm record dates back to 17 August 1435, the number of hailstorms detected only increases significantly after the 18th century. Documentary sources favour reports of particularly damaging hailstorms, so frequency increases with the number of surviving documents; obviously, this can never achieve the coverage maintained in the period of organised meteorological observations. The best temporal coverage of hailstorm days during the summer half-year in South Moravia starts in 1925 and expresses an overall decreasing trend of -0.05 d per 10 yr up to 2015, more marked after 1961 (-1.4 d per 10 yr). Particularly damaging hailstorms, on 20 June 1848, 1 July 1902, 10 July 1902 and 19 July 1903, are described. Finally, uncertainties in the hailstorm chronology are discussed, and differences related to various aspects of hailstorm days detected from documentary and meteorological data in three 40 yr periods are analysed.