Ice Regime Variation in the 20th Century Along the Serbian Sector of the Danube and Assessment of Global Climate Change Impact - page 01

Ice Regime Database

The database that contains information about ice phenomena and ice regime parameters of the Danube and its tributaries covers a very long period of time (more than 100 years), and contains chronological data on registered ice phenomena and meteorological and hydrological ice regime parameters, as well as information about ice defense measures.

The chronological data on registered ice phenomena include all periods of ice occurrence on the Danube. The durations by year have been determined based on the date of first occurrence and the date of ice melt (or end of the static ice cover). The long time series enable a statistical analysis as well, to define the frequency and probability of ice formation and static ice cover. For the Serbian sector of the Danube, time series of ice phenomena are available at several gauging stations, but their lengths differ. The longest time series, since the beginning of the 20th century, have been recorded by the stations at Bezdan, Apatin, Bogojevo, Vukovar, Ilok, Novi Sad, Zemun, Pančevo and Smederevo (as well as by Romanian stations at Drencova, Orşova and Turnu Severin).

Meteorological parameters are fundamental drivers of a river's ice regime. The most significant among them is air temperature along the river's course, but also in the entire river basin. Along the Serbian sector of the Danube, measurement of air temperatures did not begin at all stations at the same time. The station at Veliko Gradište provides the longest time series (since the beginning of the 20th century). Measurements at the other stations began after either World War I or II.

River water temperature data belong to a separate category of meteorological parameters. Water temperature is an extremely important hydrodynamic parameter, which does not depend solely on air temperature, but also on other drivers of the river cooling process. Measurement of river water temperature at the gauging stations along the Serbian sector of the Danube began in 1948.

Hydrological parameters of the ice regime include river stages and discharges. All hydrological stations along the Serbian sector of the Danube (Bezdan, Apatin, Bogojevo, Vukovar, Ilok, Novi Sad, Zemun, Pančevo and Smederevo) feature long and statistically representative river-stage gauging periods. At some of the stations the discharges are determined from rating curves. The main problem here is non-homogeneity of the datasets on the two parameters involved. Namely, along the Iron Gate 1 HPP reservoir, the river stage time series is not statistically homogeneous because the stages are a direct result of the HPP's operating mode, which has been modified several times since 1971.

The discharge data are also not homogeneous because ice formations back up the flow and thus interfere with the correlation between river stages and discharges.

The dataset on ice defense measures is rather modest, especially in the more distant past. Before WWII, most of the information is related to the winter of 1928/29, which was the most severe winter of the 20th century. The data situation improved after 1970, or after the Iron Gate 1 HPP was commissioned. The most voluminous ice defense data are available on the winters of 1984/85 and 2011/12.

 

Assessment of Ice Regime Variation

In this study the parameters of the Danube's winter regime were assessed by means of a comparative analysis of the "more distant" (P1) and "more recent" (P2) periods of time. The year 1971 was selected as the cut-off year between the two periods, because it marked the completion of the Iron Gate 1 HPP dam and the formation of the reservoir which altered the ice regime to a considerable extent. Additionally, a significant increase in both thermal and chemical pollution of the Danube (primarily in the Austrian and Hungarian sectors) was recorded in the 1970's.

The main river ice regime indictor is the frequency of ice formation and static ice cover. Figures 1 and 2 show the variations in these parameters along the Serbian sector of the Danube, in parallel for time periods P1 and P2. A large difference in ice frequency between the two periods is more than apparent. The frequency of ice formation in P1 varied from 70 to 90%, and in P2 from 30 to 60%. The frequency of static ice in P1 varied from 30 to 60%, and in P2 from 3 to 18%.

Another important indicator of a river's ice regime is the duration of ice phenomena and static ice cover. Figures 3 and 4 show the variations in maximum durations of ice phenomena and static ice cover along the Serbian sector of the Danube, for P1 and P2. The figures attest to large differences between the two periods. The maximum duration of ice phenomena in P1 was 69–94 days, while in P2 it was 31–54 days. The duration of static ice cover in P1 was 56–85 days, and in P2 10–31 days.

This comparison of P1 and P2 ice regime parameters leads to the conclusion that the differences in ice formation and static ice cover frequencies have been considerably greater than the differences in the duration of these occurrences. This was as expected, because only one (the most critical) year was taken to show the maximum duration in each period. There were many winters with long periods of ice in P1, and only one such winter in P2 (1984/85).

 

Fig01
Figure 1: Ice formation frequency in P1 vs. P2.

Fig02
Figure 2: Static ice cover frequency in P1 vs. P2.

Fig03
Figure 3: Maximum ice duration in P1 vs. P2.

Fig04
Figure 4: Maximum static ice cover duration in P1 vs. P2.