This thesis investigates the influence of low ozone episodes on UV radiation in Austria during the period 1999 to 2014. To this aim observations of total ozone, available from one site in Austria (Sonnblick) and three sites in neighboring countries (Arosa, Switzerland; Hradec Kralove, Czech Republic; Hohenpeissenberg, Germany), and erythemal UV radiation, available from twelve sites of the Austrian UV monitoring network, are analyzed. Various definitions for low ozone episodes appeared in the scientific literature. As each definition has its strengths and weaknesses, and none is particularly suited to investigate effects on UV, a novel threshold approach is developed. As this approach considers anomalies, it is applicable to both ozone and UV data and provides thereby a joint framework for the analysis of extremes. The results show a negative correlation between total ozone and UV extremes, although modulating effects of sunshine duration impact the robustness of the statistical relationship. Therefore information on relative sunshine duration, available at (or nearby) UV monitoring sites, is included as explanatory variable in the analysis. The joint analysis of UV, total ozone and relative sunshine duration across sites shows that more than 60% of observations with ozone anomalies smaller than minus one led to positive UV anomalies independent of sunshine duration. To further investigate the relationship between UV and ozone a so-called anomaly amplification factor has been introduced, established over observations during clear-sky and partly-cloudy conditions. The mean anomaly amplification factor was found to be 0.32 with a mean absolute deviation of 0.05. Further individual low ozone episodes are studied in detail, highlighting the modulating effects of total ozone and sunshine duration on UV radiation.