An example of this situation is shown in Figure 1b – R99p for the warm season. The regional time series of R95 and R99 are produced by summing the numbers of events at all stations in the region: 7 stations belong to the western region, 13 to the central region and 20 to the eastern region. The 99th percentiles of daily precipitation

distributions for Estonian stations vary between 18.9 mm and 25.3 mm in the warm season (Figure 1b) and 9.9 mm and 15.8 mm in the cold season (Figure 1a); the corresponding 95th percentiles are 9.3–13.1 mm and 5.2–8.8 mm. The R99p and R95p for the whole year fall into the 15.7–20.6 mm and 7.7–10.4 mm ranges this website respectively. Approximately the same values can be seen in Figures 2a and 2b, which show histograms of the daily precipitation see more distributions at the Viljandi and Vilsandi stations, together with the annual values R95p and R99p. These stations were selected as examples of typical stations with low (Vilsandi) and high (Viljandi) percentile values. Figures 3a, 3b and 3c show the interannual variability of R99 and

R95 at Viljandi. The R95 and R99 usually go hand in hand from one year to the next. The reason for this synchronous movement is that during years with a lot of extreme events, both very wet days and extremely wet days occur more often, and also that extremely wet days are counted among the very wet days. In Figure 3, especially in Figure 3b for the cold season, two different periods between 1961 and 2008 can be distinguished: one with lower values beginning from 1961 (or before) and ending around 1980, and the

Bortezomib cell line other with higher values beginning in the 1980s and lasting till the present day. This pattern is also apparent in the other time-series. Among the temporal changes in the series from individual stations, tendencies were evident in both directions as regards very wet and extremely wet days, but none of the falling trends was significant. Whereas summing the events over the whole country yields more stable trends (see Table 1), grouping the stations in regions allows us to refer to regions where these trends are more pronounced. If we look at the trends of the Estonian mean, then they are all significant at least at the 5% level. The trends for very wet days are always larger than for extremely wet days. This is also the case in all the regions taken separately. As we can see in Figure 4a, the number of very wet days in the warm season has increased by 5.2% at a significance level of α = 0.05. On average, events over R95p take place 9.3 times during the warm season, so the 5% increase is relatively small in absolute terms. Even more so, the same scenario applies to values over R99p during the warm season. As on average there are 1.9 events over R99p per station during the warm season, its 2.2% increase at α = 0.01 is not especially remarkable.